• TABLE OF CONTENTS
HIDE
 Title Page
 Table of Contents
 Overview
 The setting of the study
 The organization of sorghum and...
 The allocation of women's labor...
 A comparative perspective
 Policy implications
 Selection of the sample
 Endnotes
 Reference






Group Title: Paper / Nutrition and Development Project ;, no. 83-1
Title: The impact of the SEMRY I irrigated rice production project on the organization of production and consumption at the intrahousehold level
CITATION PAGE IMAGE ZOOMABLE PAGE TEXT
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00071909/00001
 Material Information
Title: The impact of the SEMRY I irrigated rice production project on the organization of production and consumption at the intrahousehold level
Series Title: Paper Nutrition and Development Project
Physical Description: 139 p. : ; 28 cm.
Language: English
Creator: Jones, Christine Winton
United States -- Agency for International Development. -- Bureau for Program and Policy Coordination
Publisher: The Agency
Place of Publication: Washington D.C.?
Publication Date: 1983
 Subjects
Subject: Rice trade -- Cameroon   ( lcsh )
Rice farmers -- Cameroon   ( lcsh )
Irrigation -- Economic aspects -- Cameroon   ( lcsh )
Household surveys -- Cameroon   ( lcsh )
Genre: federal government publication   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )
Spatial Coverage: Chad
 Notes
Bibliography: Bibliography: p. 136-139.
Statement of Responsibility: Christine W. Jones ; prepared for the Agency for International Development.
General Note: "September 1983."
Funding: Paper (Nutrition and Development Project (U.S.)) ;
 Record Information
Bibliographic ID: UF00071909
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 20634565
lccn - 89110701

Table of Contents
    Title Page
        Title Page
    Table of Contents
        Table of Contents
    Overview
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
    The setting of the study
        Page 6
        Selection of the villages
            Page 6
            Page 7
            Page 8
            Page 9
            Page 10
            Page 11
            Page 12
        Selection of the sample
            Page 13
        Data collection
            Page 14
            Page 15
            Page 16
            Page 17
            Page 18
            Page 19
            Page 20
            Page 21
    The organization of sorghum and rice production
        Page 22
        The organization of sorghum production
            Page 22
            Land
                Page 22
                Page 23
            Labor
                Page 24
                Page 25
                Page 26
                Page 27
            Capital
                Page 28
        The organization of rice production
            Page 29
            The installation of SEMRY
                Page 29
            SEMRY's current problems
                Page 30
                Page 31
                Page 32
                Page 33
            Present organization of rice production
                Page 34
                Page 35
                Page 36
                Page 37
                Page 38
            Summary of the major differences between sorghum and rice production
                Page 39
        The agricultural calendar
            Page 40
            Page 41
            Page 42
            Page 43
            Page 44
            Page 45
            Trade-offs between sorghum and rice cultivation
                Page 46
                Page 47
                Page 48
                Page 49
    The allocation of women's labor to rice production
        Page 50
        The remuneration of women's rice cultivation labor
            Page 51
            Page 52
            Page 53
            Opportunity cost of women's labor
                Page 54
                Page 55
        The intrahousehold responsibility for food provision
            Page 56
            Page 57
            The Massa diet
                Page 58
                Page 59
            Responsibility for providing grain
                Page 60
                Page 61
                Page 62
                Page 63
                Page 64
                Page 65
            Changes in the division of the responsibility for sauce ingredients
                Page 66
                Page 67
                Page 68
                Page 69
                Page 70
                Page 71
                Page 72
        Allocation of labor between sorghum and rice production
            Page 73
            comparison of married and independent women's labor allocation
                Page 73
                Page 74
                Page 75
                Page 76
                Page 77
                Page 78
                Page 79
                Page 80
                Page 81
                Page 82
        A bargaining approach to household resource allocation
            Page 83
            Page 84
            Page 85
            Page 86
            Page 87
            Page 88
    A comparative perspective
        Page 89
        Page 90
        Page 91
        Page 92
        Control over income-generating resources
            Page 93
            Page 94
            Page 95
            Page 96
        Mobilization of women's labor
            Page 97
            Page 98
            Page 99
            Page 100
            Page 101
            Page 102
            Page 103
            Page 104
        Allocation of responsibility for household maintenance
            Page 105
            Page 106
            Page 107
            Page 108
            Page 109
            Page 110
    Policy implications
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
    Selection of the sample
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
    Endnotes
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
    Reference
        Page 136
        Page 137
        Page 138
        Page 139
Full Text
- t


I/, Cc2-








THE IMPACT OF THE SEMRY I IRRIGATED RICE PRODUCTION PROJECT

ON THE ORGANIZATION OF PRODUCTION AND CONSUMPTION
AT THE INTRAHOUSEHOLD LEVEL






Christine W. Jones
Harvard University





September 1983





Prepared for the Agency for International Development
under Contract OTR-0096-00-2232-00







Agency for International Development
Bureau for Program and Policy Coordination

Nutrition and Development Project
Paper No. 83-1







TABLE OF CONTENTS

Page
Section 1. Overview 1.

Section 2. The Setting of the Study 6.

2.1 Selection of the Villages 6.
2.2 Selection of the Sample 13.
2.3 Data Collection 14.

Section 3. The Organization of Sorghum and Rice Production 22.

3.1 The Organization of Sorghum Production 22.
Land 22.
Labor 24.
Capital 28.
3.2 The Organization of Rice Production 29.
The Installation of SEMRY 29.
SEMRY's Current Problems 30.
Present Organization of Rice Production 34.
Summary of the Major Differences between
Sorghum and Rice Production 39.
3.3 The Agricultural Calendar 40.
Trade-offs between Sorghum and Rice Cultivation 46.

Section 4. The Allocation of Women's Labor to Rice Production 50.

4.1 The Remuneration of Women's Rice Cultivation Labor 51.
Opportunity Cost of Women's Labor 54.
4.2 The Intrahousehold Responsibility for Food Provision 56.
The Massa Diet 58.
Responsibility for Providing Grain 60.
Changes in the Division of the Responsibility for
Sauce Ingredients 66.
4.3 Allocation of Labor between Sorghum and Rice Production 73.
Comparison of Married and Independent Women's Labor
Allocation 73.
4.4 A Bargaining Approach to Household Resource Allocation 83.

Section 5. A Comparative Perspective 89.

5.1 Control over Income-Generating Resources 93.
5.2 Mobilization of Women's Labor 96.
5.3 Allocation of Responsibility for Household Maintenance 105.

Section 6. Policy Implications 111.


Appendix The Selection of the Sample 121.

Endnotes 126.


References


136.








SECTION 1. OVERVIEW



Policymakers usually assume that a project which benefits the

household as a unit benefits each member of the household. This

assumption reflects the key assumption on which the neoclassical model

of the household is based: namely, that the household is a joint

decision-making unit which allocates its resources and spends it income

according to a mutually agreed upon, and therefore identical, set of

priorities. Recent scholarship, however, challenges the view that

household members have identical preferences (Dwyer, 1983). Indeed, if

household members' preferences do not always coincide, then a new model

of household economic behavior is called for.

Bargaining models have been proposed as an alternative to the

joint household utility function model of the household (Manser and

Brown, 1979 and 1980; McElroy and Homey, 1981; Folbre, forthcoming).

Bargaining models recognize that household members may have conflicting

as well as complementary interests. The "weight" attached to an

individual household member's preferences depends on his or her

bargaining power. Thus, a bargaining model of the household forces one

to pay attention to those variables which give some household members

greater leverage in determining the household resource allocation and

expenditure pattern.

The saliency of the bargaining approach to household economic

behavior and its implications for policymaking are brought out in this

case study. It analyzes the impact of an irrigated rice production

project in Cameroon on women's labor allocation, intrahousehold income

distribution, and the intrahousehold division of responsibility for food








provision. Its aim is (1) to delineate the factors which determine

women's access to cash-cropping resources and the terms under which

their labor is mobilized by their husbands, and (2) to analyze the

impact of the intrahousehold incentive structure on the efficiency of

resource allocation. The major conclusion the study is that, despite

the fact that rice cultivation appears to have benefitted women, they

allocate their labor inefficiently because of intrahousehold conflict

over the distribution of rice income. The costs of ignoring or assuming

away such conflicts at the project design stage are high if they

contribute to the failure of projects to achieve their welfare and

production objectives.

The analytical framework for the study is based on the

following set of questions, which are addressed to the extent permitted

by the data available:

-To what extent is women's labor mobilized for cash crop

production?

--What activities have they forgone on account of cash-cropping?

For example, have they forgone childcare, food preparation,

leisure, or income-generating activities?

--Has anyone else in the household taken on the responsibility

for the domestic labor activities women forwent?

--What was the opportunity cost of forgoing these activities?

--How much income have women forgone from their own income-

generating activities on account of cash crop production?

--Are they compensated for their cash-cropping labor? How does

the intrahousehold rate of compensation influence women's

labor allocation pattern?


-2-








--If women's incomes have decreased or increased on account of

cash-cropping, how have they readjusted their expenditure

pattern, what are the welfare costs of these readjustments and

by whom are they borne?

--How have other household members' expenditure patterns shifted

in response to alterations in women's expenditure pattern?

--If women do not directly control the income generated by their

labor in the new activity, how would the household expenditure

pattern differ if they did control the income and what would

the likely impact on household welfare be?



As these questions demonstrate, the relevant issue is not whether cash-

cropping increases women's agricultural labor, but how it alters their

labor allocation pattern in conjunction with the intrahousehold income

distribution and household expenditure patterns. Shifts in all the

economic dimensions open to intrahousehold negotiation must be factored

into any assessment of the impact of cash-cropping on women.

The study is organized as follows. Background information on

the selection of villages, sample and household structure is presented

in Section 2. Section 3 describes the organization of sorghum

production, the traditional subsistence crop, and the organization of

rice production, a recently introduced crop controlled by a semi-

autonomous government authority, SEMRY. It concludes with a discussion

of the agricultural calendar to indicate the periods of the year during

which the two crops compete for labor. In the area of North Cameroon in

which this study was conducted, labor is the scarce resource;.additional

land for both sorghum and rice production can generally be obtained by


-3-








farmers in the project area.

Women's labor is mobilized by their husbands for irrigated rice

production at the expense of sorghum production and other income-

generating activities. Section 4 relates the amount of compensation

that women receive from their husbands in return for their labor on rice

production to the opportunity cost of their labor in nonrice income-

generating activities. The data indicate that women are compensated at

a rate greater than the opportunity cost of their labor. Moreover, the

increase in their incomes is real. An analysis of the changes in the

food provision pattern suggest that rice cultivation has not increased

women's repsonsibility for the provision of food. Despite the increase

in their real income, however, they do not allocate their labor

efficiently, as a comparison of the labor allocation pattern of women

whose husbands control the disposition of income from rice production

and women who control the disposition themselves shows. A bargaining

model of the household, which does not assume that household members'

preferences coincide, is proposed as an explanation for the inefficient

pattern of resource allocation.

Section 5 draws on comparative material from elsewhere in Africa

to place this case study in context. Using the perspective afforded by

a bargaining model of the household, it examines the factors which

influence the extent to which women are able to capture a share of the

gains from cash crop production. In particular, it argues that account

must be taken of how the structural position of women, within their

households and the wider society, influences the degree to which they

exercise control over resources, the terms under which their labor is

mobilized, and the categories of household maintenance needs for which


-4-




* 0


they are responsible. As Section 5 points out, the impact of cash-

cropping on women has been by no means been uniform, because of the

variation in women's bargaining power and the particular circumstances

surrounding the introduction of a cash crop in a given area.

Section 6 concludes with a brief discussion of the policy

implications of the bargaining approach to household decision-making,

particularly with respect to women. To a certain extent, policymaking

in the area of "women in development" has been hampered by the lack of

an alternative to the neoclassical household decision-making model.

Under the assumption that an increase in household income is translated

into an increase in individual household member welfare, the

neoclassical perspective rarely concerns itself with the distribution of

benefits within the household. A bargaining model, however, is based on

the recognition that household members have different preferences and

that, due to their differential bargaining power, some members'

preferences have greater weight than other members' in determining the

household pattern of resource allocation and income distribution. Thus,

the challenge for policymakers is first, to understand the conflicts in

household members' preferences and second, to find a means of increasing

the bargaining power of the household member whose preferences are most

closely identified with the goal of the policymaker. Since preferences

and bargaining power are socially constructed along the lines of gender

as well as other variables, the failure to take gender into account in

the design of policy may result in inefficiencies and inequities neither

foreseen nor desired by the policymaker.


-5-




4 4


SECTION 2. THE SETTING OF THE STUDY



2.1. Selection of villages

Formal survey work was carried out in three villages, which were

selected according to their relative degree of involvement in rice

cultivation. Two villages bordering the project area were chosen to be

representative of rice-cultivating villages, and one village was chosen

from outside the project area to serve as.a control. Location vis-a-vis

the rice fields is the major factor which distinguishes villages in the

project area. The SEMRY I project area extends northward from the town

of Yagoua, where SEMRY's headquarters are located, along the western

bank of the Logone River which forms the border with Chad.

Villages located on the eastern side of the project perimeter

are confined to a relatively narrow piece of land bounded on one side by

the Logone River and on the other by the rice fields (see Map 2.1).

Farmers from Vele, the village which was chosen to be representative of

villages on the eastern side, are at most about a half-hour's walk from

their rice fields. Because of the proximity of the rice fields, and the

difficulty of extensifying sorghum cultivation, virtually all households

located along the eastern perimeter (and particularly those in the

southern half of the perimeter closest to Yagoua where population

density is higher) cultivate some rice in addition to sorghum.

However, many households on the western side of the perimeter

choose not to cultivate rice. They are daunted by the long and arduous

walk in the rainy season through mud and hip-deep water and attracted by

the possibilities of extensifying their sorghum cultivation and growing


-6-




4 4


SECTION 2. THE SETTING OF THE STUDY



2.1. Selection of villages

Formal survey work was carried out in three villages, which were

selected according to their relative degree of involvement in rice

cultivation. Two villages bordering the project area were chosen to be

representative of rice-cultivating villages, and one village was chosen

from outside the project area to serve as.a control. Location vis-a-vis

the rice fields is the major factor which distinguishes villages in the

project area. The SEMRY I project area extends northward from the town

of Yagoua, where SEMRY's headquarters are located, along the western

bank of the Logone River which forms the border with Chad.

Villages located on the eastern side of the project perimeter

are confined to a relatively narrow piece of land bounded on one side by

the Logone River and on the other by the rice fields (see Map 2.1).

Farmers from Vele, the village which was chosen to be representative of

villages on the eastern side, are at most about a half-hour's walk from

their rice fields. Because of the proximity of the rice fields, and the

difficulty of extensifying sorghum cultivation, virtually all households

located along the eastern perimeter (and particularly those in the

southern half of the perimeter closest to Yagoua where population

density is higher) cultivate some rice in addition to sorghum.

However, many households on the western side of the perimeter

choose not to cultivate rice. They are daunted by the long and arduous

walk in the rainy season through mud and hip-deep water and attracted by

the possibilities of extensifying their sorghum cultivation and growing


-6-

















SEMRY I PROJECT AREA: YAGOUA, CAMEROON


DOREISSOU


CHAD


0

S


--
--
-
--
( -.


YAGOUA *


0 5 km


I,


LEGEND


Road
""* Dike Road
--. Border

0 Village


Rice Field


l River


r I Mayo
L-_J (seasonal)


- "^










cotton. For all these reasons, only about 19% of the compounds in

Widigue, the village chosen to be representative of those on the western

side of of the project perimeter, cultivated rice in 1980 (Sisson and

Ahlers, 1981). The remaining households could have cultivated rice if

they had chosen to, since each year many fields remain uncultivated due

to lack of farmer interest. The compounds in Widigue which do cultivate

rice are generally those located closest 'to the rice fields and furthest

from the village's cotton fields.

Zebe, the third village, located outside the project area, was

selected as a control. Like Vele, it is located along the Logone River

but it is southwest of Yagoua.

The three villages are similar in many respects. They are

ethnically quite homogeneous. Although there is a small quarter in each

village which is inhabited by the Fulbe, the rest of the population

identify themselves as Massa. Virtually all the rice cultivators in

approximately the'southernmost three-quarters of the project area are

Massa.

Almost without exception, every compound in each of the three

villages surveyed cultivates sorghum, which is the mainstay of the diet.

Millet is also cultivated on the sandier soils found in Zebe and on the

western side of Widigue. In general, soils in Zebe are much poorer than

in the other two villages, and as a consequence sorghum yields are much

lower. According to the 1980 and 1981 farm management surveys (Sisson

and Ahlers, 1981; Bikoi, 1982), the average sorghum yield in Zebe were

310 kg/ha in 1980 and 401 kg/ha in 1981. In contrast, the average

sorghum yield in Widigue was 988 kg/ha in 1980 and 806 kg/ha in 1981.











Vounaloum, a village bordering along the Logone about 10 km upriver from

Vele, reported yields were 360 kg/ha in 1980 and 1034 kg/ha in 1981.

Sorghum yields in Vele were apparently exceptionally good in 1981--1616

kg/ha. While it is certainly not inconceivable that some Vele farmers

obtained sorghum yields of that magnitude, a comparison with the yields

obtained in Vounaloum, which is quite similar to Vele both in soil type

and labor allocation patterns, suggests that the average yields reported

for Vele may be overstated.

In the last thirty years the Massa to the north of Yagoua have

adopted a variety of transplanted dry season white sorghum, dongolonga,

long cultivated by those to the south of Yagoua and by the populations

of the Diamare plain of North Cameroon (de Garine, 1964:66). Dongolonga

can only be cultivated in fields which are flooded during the rainy

season. These fields are in short supply in the project area. At the

end of the rainy season, the young seedlings are transplanted into

ten-inch-deep holes into which a cup of water is poured. They flourish

throughout the dry season on only this water and the retained moisture

in the soil, which has a high clay content. Yields average about .9

t/ha.

It is not entirely clear what the impetus was for the adoption

of dongolonga. Increasing population pressure and the loss of soil

fertility may have been two factors that made its cultivation

attractive. Also, increasing contact with the Toupouri and the Fulbe

(some young Massa men migrate seasonally to work on the dongolonga

fields of the Fulbe) may have played a part. In addition, compounds

which found it expedient to abandon some of their rainy season sorghum


-9-










fields to undertake rice cultivation may have taken up dongolonga as a

substitute for red sorghum. The resistance to dongolonga cultivation

long manifested by the Massa broke down once irrigated rice cultivation

was introduced. Not every compound has access to dongolonga fields,

however, since many compounds lost their existing or potential fields

when the land was taken over by SEMRY for rice cultivation.

Cotton is an option only for farmers in Widigue, because the

parastatal responsible for cotton production does not operate in

villages along the eastern perimeter of the rice fields where sorghum

land is limited nor in the villages to the southwest of Yagoua along the

Logone River. Aside from rice and cotton, the major agricultural

difference between the three villages is the extent to which tobacco, a

dry season crop, is grown. In Zebe there is a stream bed that dries up

several months after the rains end. The soil is very propitious to the

cultivation of tobacco, and it is easy to dig shallow wells in the

stream bed to obtain water for irrigation. In Vele and Widigue, tobacco

is cultivated on a much smaller scale, mostly for home consumption. All

women (and several men) in Zebe grow tobacco to obtain cash in order to

purchase enough grain to make up the substantial deficit in grain they

face each year. Virtually all households in Widigue (rice and nonrice-

cultivating alike) and most households in Vele, however, produced enough

grain in the 1980 rainy and dry season to meet their subsistence needs.

Commercial opportunities are probably the best for households in

Zebe, since it is only about 11 km from Yagoua. In addition to the

small weekly market in Zebe, villagers can take advantage of the large

weekly market in Yagoua which draws people in from all the surrounding


-10-










villages. There is also a daily market in Yagoua. Only at the height

of the fishing season, however, do women from Zebe who sell fish walk to

the Yagoua market each day. In general, people from Zebe attend the

Yagoua market at most once a week.

Walking from Vele to Yagoua takes considerably more time, since

Vele is about 25 km from Yagoua. However, there is a very small market

which operates daily in Vele where fresh fish can often be bought. The

weekly market in Vele is the largest in any village along the eastern

perimeter. After the rice harvest, however, villagers find their way to

the Yagoua market.

Of the three villages, Widigue is the most disadvantaged

commercially. There are two small weekly markets in Widigue as well as

one in the town north of Widigue which attract traders from Yagoua.

Many of those from Widigue walk the 17 km to Yagoua in the dry season to

sell their surplus grain. There is less opportunity to purchase fresh

fish in Widigue.

Some basic agricultural information on the three villages is

presented in Table 2.1. The results for Widigue and Zebe were taken

from the census carried out in conjunction with a farm management survey

done in 1980 (Sisson and Ahlers, 1981), and information for Vele was

obtained from a census taken in 1981 (Bikoi, 1982).


-11-












Table 2.1 Agricultural Characteristics of the
Three Villages Selected for the Survey


Zebe Widigue Vele


Number of Compounds 126 320 201

Crops Grown
Rainy season rice 3% 19% 98%
Dongolonga 38% 70% 47%
Millet 98% 47% 0%
Tobacco 100% 37% 57%


There are small differences in the social amenities found in the

three villages. Vele has a primary school built out of cement, Zebe is

constructing one, and Widigue began a primary school, which is built out

of mud and thatch, several years ago at the initiative of parents. Few

of the children in any of the villages go to the secondary school

located in Yagoua. It is rare to find a man who speaks more than a few

words of French in any of the villages, and even rarer to find a woman

who does. There is a well-established Catholic mission in Vele, but

both Protestant and Catholic church services are held in the other

villages as well. Few of the villagers have converted to Islam, and

strong support for the Protestant and Catholic religions is not

widespread.

Medical services in the villages are limited to a small (and

quite ill-equipped) dispensary in Vele. For serious medical problems

some villagers seek attention in Yagoua, but many others rely on

traditional remedies. It is difficult for villagers to obtain even

anti-malarial pills.


-12-










Little money is invested in housing. Only several of the huts

of the compound heads in the three villages had tin roofs. Grain is

stored in granaries built out of mud and covered with thatch which are

raised above the ground, but even so there is some loss of grain due to

insect damage. The low level of scholarization, the infant mortality
2
rate and the overall standard of living suggest that the Massa are

relatively disadvantaged when compared to other ethnic groups in

Cameroon.



2.2. Selection of the Sample

After a census was taken in the three villages, a sample of 102

women was selected for the labor allocation and food expenditure survey.

Table 2.2 shows the composition of the final sample. A random

stratified sample of compounds was chosen using an interval selection

process based on the number of adult workers in the compound. The

sample was stratified to obtain a sufficient number of cases for

intravillage comparisons on the basis of marital status and/or

agricultural activities. Women were then randomly chosen from compounds

within each stratum. The selection procedure is described in more

detail in the Appendix.


-13-












Table 2.2 Composition of the Sample of Women


Zebe married women 16
widowed women 14

Widigue rice cultivators 16
nonrice cultivators 14

Vele married women who work on the
fields of other household members 26

married women who work on their
own fields 5

widows 11


2.3 Data Collection

Two formal surveys were conducted, one a survey of women's labor

allocation to agricultural activities and the other a survey of

household food expenditures and women's earned income. The labor

allocation survey began in mid-May with the rains and continued until

the end of the rainy season rice harvest in late December. The

information gathered was basic. Women were asked what crop they worked

on, what activity they performed, whose field they worked on, when they

left their compound to go work on the field and when they returned to

their compound, and what kind of and how much remuneration they

received, if any.3 Rice cultivators were interviewed every other day at

the end of the day about that day's and the previous day's agricultural

activities from the middle of May to the end of December. Nonrice

cultivators, due to budgetary reasons, were interviewed only during the

period of peak labor demand, mid-May through the end of August. The


-14-










short recall period was chosen to yield high quality data to compensate

for the small sample size.

All the sorghum and millet fields of the women in the sample

were measured. The size of a household's rice fields and the weight of

the paddy sold were indicated on the slip it received from SEMRY for the

sale of its paddy. Data on total yield and quantity of paddy retained

by the household for consumption and in-kind payments were obtained from

interviews with household members. The average returns to rice and

sorghum labor were calculated using aggregate data from the farm

management survey. The data on women's labor allocation were to have

been supplemented by the gender-specific data on labor inputs to all

household agricultural enterprises collected in the 1981 farm management

survey. However, those data are not yet available. This unfortunately

limits the comparisons which can be made between men's and women's labor

allocation patterns.

No data were collected on the allocation of women's time to

nonagricultural activities. The decision to forgo the collection of

these data was based on financial, time, and managerial constraints.

Obviously, in assessing the.impact of a project on women, changes in

their domestic labor allocation pattern are important because of thier

potential impact on household welfare. However, I did interview women

about whether rice cultivation had altered their domestic labor

allocation pattern. Women in Widigue reported that the long walk to the

rice fields, and hence their late return in the day, sometimes after

dark, left them too tired on occasion to prepare dinner. On the other

hand, women frequently prepared porridges of rice, particularly in the


-15-










morning. Often the porridges were made with grains of rice, rather than

rice ground into flour, the method used to prepare porridges of sorghum.

Not having to grind grain into flour saves women a considerable amount

of time and drudgery. It should also be noted that the survey results

(see Section 3.3) indicate that rice cultivation substitutes for, rather

than adds to, sorghum labor during the period of peak labor demand.

Data on household food expenditures were also collected at five

points during the year for a two-week period, using a two-day recall

interval. The expenditure survey is described in more detail in Section

4. It was designed to determine how rice cultivation has affected the

intrahousehold food expenditure pattern. The decision to limit the

questionnaire to expenditures on food reflected a desire to simplify the

questionnaire to the extent possible given the frequent interview

schedule. It also reflects the assumption that any shift in the

household expenditure pattern in women's disfavor would in all

likelihood be most pronounced in the category of food expenditure, since

most of women's income is spent on food. No new category of expenditure

appears to have been introduced concurrently with rice production;

households spend little money on schooling, health care, or their

dwellings. A questionnaire was also administered on women's earned

income simultaneously with the food expenditure survey. The income

questionnaire was designed to determine the magnitude and sources of

women's income, particularly from sales of grain, and also to acquire

some sense of the returns to women's labor from nonagricultural

activities.


-16-










2.4. The Domestic Unit

Villages extend for four or five kilometers along a road and

another two kilometers or so on each side of the road. Compounds tend

to be widely dispersed throughout the village. Within each village

(which is basically an administrative unit established by the colonial

state), there are small territorially-based groups of compounds which

are composed of agnates descended from a common ancestor several

generations back. Rights to the land surrounding their compounds is

vested in these small descent groups. Marriages are prohibited between

members of the same descent group.

The basic residential unit of the Massa is the zina, or

compound. Zina refers to the collection of huts, kitchens, granaries,

etc., which are arranged in a circle around a central granary controlled

by the head of the compound, and also by extension to the people who

reside in the compound. The male head of the compound, if there is one,

is called the bum zina. He resides in the compound with his wives and

their unmarried children. In many instances one or more of his married

sons or his younger brothers and their wives and children also reside in

the compound.

A very small percentage of compounds are headed by women, all of

whom are widows. Upon her husband's death, a widow is usually

incorporated in the household of another male. Most frequently, she is

inherited by one of her husband's junior agnates; the Massa practice

levirate. If her son resides in the compound, however, a woman past

childbearing age will often remain with him, whether or not she is

inherited. A few childless widows work for many years to accumulate


-17-










enough cattle to return their bridewealth to their husband's family so

that they can return to their natal village. Thus, a women becomes a

compound head only if she is past childbearing age and does not have a

son who resides in the same compound.

Throughout this study the word "household" will be used to refer

to the group composed of a married man and his wives and their children

or the rare case of a widow-headed household. However, there is no word

in Massa which refers to the conjugual unit. Compound members refer to

themselves as belonging to a certain zina. The part of the zina which

is a woman's personal domain, her hut, kitchen and granary, is denoted

by the word diguiligna. There is no part of the compound, however, that

in some sense belongs to the household. Each wife cooks for herself and

her children, takes turns cooking for her husband, fetches her own

water, and does her own laundry. Sorghum fields are cultivated either

individually or by the compound as a unit. The grain that a woman

prepares may come from either her own field, the field of her husband,

or the collective sorghum field, depending on the time of year. Thus,

members of a conjugual household can not be said to eat out of the same

pot, i.e, eating grain from a common household field. A husband and

wife (or wives) constitute neither a unit of production nor a unit of

consumption. The major exception to this is rice cultivation, where a

husband and wife (or wives) work on the same fields together. The

differences in the organization of sorghum and rice production are

discussed in Section 3.

Table 2.3 indicates the number of compounds headed by men and

women in Vele. Vele was the only village censused in its entirety in


-18-










1981 (see the Appendix), and the only village, therefore, for which data

on the number of female headed-compounds and the rate of polygyny (see

Table 2.5) are available on a village-wide basis. As Table 2.3 shows,

the percentage of female-headed compounds in Vele is very small. Very

few female-headed compounds were encountered in the other two villages.


Heads of Compounds in Vele, Broken
Down by Marital Status and Gender


No of compounds headed by widowed women:


No. headed by unmarried men:

divorced
young men living with widowed mothers
widowers
older men never married


No. headed by married men:


5 (2%)


19 (9%)

7
8
8
1


177 (88%)


Table 2.4 indicates the number of multi-household compounds in

the three villages. The data for Widgue and Zebe were obtained from the

1980 farm management census (Sisson and Ahlers, 1981).



Table 2.4 Number of Households per Compound


Number of compounds Vele Widigue Zebe


1 Household/Compound 100 (50%) 59% 72%

2 Households/Compound 60 (30%) 25% 19%

3+ Households/Compound 41 (20%) 16% 9%


-19-


Table 2.3










Women in the villages tend to marry at a very early age, generally by

sixteen at the latest. Given the high bridewealth which is required,

however, men tend to marry much later than women. Most men would prefer

to be polygynous though some of the young men who have converted to

Christianity are choosing not to be. As Table 2.5 shows, the majority

of married men in Vele have only one wife, usually because they have not

yet accumulated enough cattle to marry another. Still, a substantial

minority have more than one wife.




Table 2.5 Number of Wives of Married Men in Vele


No. of married men with:

one wife 248 (70%)
two wives 83 (23%)
three wives 16 (5%)
four or more 7 (2%)



What is interesting is that the percentage of men with more than one wife

is almost identical to the one cited by the first ethnographer of the Massa

twenty-five years ago. In a small sample of compounds from the village

immediately to the north of Vele, De Garine (1964:159) found that 72% of

the married men had one wife, 21% had two, 5% had three wives and only 1%

had four or more wives.

Evidently it seems to be as difficult to acquire a wife today as it

was twenty-five years ago, despite the inflow of cash from rice production

into the economy. Although the number of cattle (usually ten) required for

bridewealth has not increased in the past century (de Garine, 1964:151),

the price of cattle has. Even if men's real incomes have increased as a


-20-











result of rice cultivation, it would probably be many years before an

increase in the rate of polygyny would become apparent, since the

bridewealth payment is so high relative to men's income. The proceeds of

about ten seasons of rice cultivation are required to buy the cattle

necessary for bridewealth payment. Farmers in Vele, for example, have been

cultivating rice for only about eight years, and they certainly have not

spent their entire earnings on cattle. Thus, it is highly unlikely that

rice cultivation has had a significant impact on the rate of polygyny to

date..


-21-










SECTION 3. THE ORGANIZATION OF SORGHUM AND RICE PRODUCTION


This section describes the organization of sorghum and rice

production. Despite major differences between the two crops in the

organization of production-unlike sorghum production, rice production

is controlled by a semi-autonomous government authority, SEMRY-the two

crops have one feature in common. The major constraint to increased

production of both crops in the rainy season is not land, but labor. It

is important, therefore to understand the factors which determine the

set of constraints, opportunities, and incentives faced by particular

categories of producers--men, women, elders, juniors--with respect to

the two crops.

Section 3.1 describes the mobilization of land and labor for

sorghum production. Section 3.2 focuses on the structure imposed by

SEMRY on the production and commercialization of rice. Section 3.3

discusses the constraints on labor allocation imposed by the

agricultural calendar. Section 4 takes up the issue of the

intrahousehold relations of production and distribution and their effect

on women's labor allocation.



3.1. The Organization of Sorghum Production

Land. The Massa distinguish two categories of fields: those in

the vicinity of the compound which are inherited by the bum zina, the

head of the compound, belong to his descent group and cannot be

alienated, and those which individuals in the compound clear in

unoccupied territory beyond the immediate compound and are alienable.


-22-










SECTION 3. THE ORGANIZATION OF SORGHUM AND RICE PRODUCTION


This section describes the organization of sorghum and rice

production. Despite major differences between the two crops in the

organization of production-unlike sorghum production, rice production

is controlled by a semi-autonomous government authority, SEMRY-the two

crops have one feature in common. The major constraint to increased

production of both crops in the rainy season is not land, but labor. It

is important, therefore to understand the factors which determine the

set of constraints, opportunities, and incentives faced by particular

categories of producers--men, women, elders, juniors--with respect to

the two crops.

Section 3.1 describes the mobilization of land and labor for

sorghum production. Section 3.2 focuses on the structure imposed by

SEMRY on the production and commercialization of rice. Section 3.3

discusses the constraints on labor allocation imposed by the

agricultural calendar. Section 4 takes up the issue of the

intrahousehold relations of production and distribution and their effect

on women's labor allocation.



3.1. The Organization of Sorghum Production

Land. The Massa distinguish two categories of fields: those in

the vicinity of the compound which are inherited by the bum zina, the

head of the compound, belong to his descent group and cannot be

alienated, and those which individuals in the compound clear in

unoccupied territory beyond the immediate compound and are alienable.


-22-










SECTION 3. THE ORGANIZATION OF SORGHUM AND RICE PRODUCTION


This section describes the organization of sorghum and rice

production. Despite major differences between the two crops in the

organization of production-unlike sorghum production, rice production

is controlled by a semi-autonomous government authority, SEMRY-the two

crops have one feature in common. The major constraint to increased

production of both crops in the rainy season is not land, but labor. It

is important, therefore to understand the factors which determine the

set of constraints, opportunities, and incentives faced by particular

categories of producers--men, women, elders, juniors--with respect to

the two crops.

Section 3.1 describes the mobilization of land and labor for

sorghum production. Section 3.2 focuses on the structure imposed by

SEMRY on the production and commercialization of rice. Section 3.3

discusses the constraints on labor allocation imposed by the

agricultural calendar. Section 4 takes up the issue of the

intrahousehold relations of production and distribution and their effect

on women's labor allocation.



3.1. The Organization of Sorghum Production

Land. The Massa distinguish two categories of fields: those in

the vicinity of the compound which are inherited by the bum zina, the

head of the compound, belong to his descent group and cannot be

alienated, and those which individuals in the compound clear in

unoccupied territory beyond the immediate compound and are alienable.


-22-










The land that surrounds the compound is controlled by the bum zina. It

is usually divided into one large field, sinema ngolla, which is

collectively worked under his direction, and into smaller fields which

are individually worked.

When one of the junior men in the compound marries, the compound

head gives him a field usually in the vicinity of the compound for the

use of his household, which he then divides up between his wife and

himself. If land is in short supply so that a married man cannot

maintain his household, he might move his household to a locale where

there is sufficient land for compound fields and install himself as bum

zina. Alternatively, members of his household might decide to clear

uncultivated land, sinema fulla. These fields are generally located at

some distance from the compound and do not benefit from manuring in the

dry season by the constant passage of the compound's cattle. Both men

and women can establish sinema fulla, although the right to alienate the

field remains with a woman's husband. Still another strategy which is

employed is for a man to ask his wife's parents for a field which she

then returns home to cultivate at various times throughout the rainy
1
season.

In general, sorghum land is available, albeit at some distance

from the compound. Construction of the dike along the Logone River

increased the area which could be put under sorghum cultivation and

reduced the variation in yields by controlling flooding. However,

villagers along the eastern perimeter have the least opportunity to

extensify their sorghum cultivation.

At present, no land market has developed, even in villages along


-23-










the eastern perimeter of the rice fields. Occasionally, a field is lent

when its proprietor is unable to cultivate it (in the case of illness,

for example), or when a compound ceases to exist. In these cases,

recognition of the proprietor's (or the lineage's) claim to the field is

made by a token offering of grain or a small cash payment after the

harvest. Except in the very rare case of an outright sale of a sinema

fulla, the field can be reclaimed in the future by the proprietor.

Labor. With the exception of the collective field to which

every member of the compound is expected to contribute several days'

work of planting, weeding and harvesting, sorghum fields are generally

cultivated on an individual basis. However, variants of the typical

pattern do exist. The.unmarried sons of the bum zina usually cultivate

the collective field with him if it is sufficiently large. Otherwise,

they clear their own fields. Sometimes married men also cultivate with

the bum zina, but they usually have their own fields. Rarely does a man

cultivate a sorghum field with his wife. The exceptional case of joint

cultivation is usually due to extenuating circumstances such as illness

or childbirth which forces one spouse to take on the responsibility for

the field of the other spouse during part of the agricultural season.

If a man has more than one wife, each wife (except perhaps in the first

year of her marriage) will have at least one field of her own, generally

in the vicinity of the compound.

In Vele, however, unlike Widigue or Zebe, there were several

compounds that had anomalous patterns of cultivation which are not

explainable by the extenuating circumstances described above. Of the

thirty-six compounds from which women were chosen, seventeen of the


-24-









compounds were composed of only one conjugal household and the other

nineteen of more than one. Among the former group, only one exception

to the general pattern was observed: that of a husband who had no

sorghum field himself but aided his two wives in the cultivation of

theirs. He had his own granary, in which he stored purchased sorghum.

Among the nineteen multi-household compounds, there were seven

that had no collective field on which all compound members worked.

There are several factors which explain why some of these compounds did

not have collective fields. In one case the compound head was old and

nearly blind, and his brother only slightly less so, so that each of

their wives cultivated a portion of the compound's fields. Another case

was a large compound composed of five households with relatively little

sorghum land. They cultivated rice more intensively than almost any

other compound in the village. Several of the women's husbands in this

compound did not have their own sorghum fields and worked instead on the

fields of their wives. In four other cases, the junior household in the

compound was headed by a cousin of the compound head. The more distant

relationship between the two heads of households may have contributed to

the relative independence of the two households. Or alternatively, rice

cultivation may be responsible for the formation of multi-household

compounds that would otherwise not have formed given the lack of a close

relationship.

These cases also suggest that another result of intensive rice

cultivation may be a reduction in the importance placed on the

collective sorghum field and on the ultimate responsibility of the

compound head for insuring that an adequate supply of sorghum is


-25-










available. Sorghum from the collective field is stored in the granary

belonging to the compound head. The sorghum is distributed to all the

households in the compound in the months immediately preceding the

sorghum harvest. However, rice production has diminished the need for

the distribution of sorghum from the collective field and hence the need

for the collective field since households which have deficits in sorghum

can usually retain enough paddy to cover their grain needs.

Furthermore, collective fields may be becoming less important

since junior men have less incentive to work on them now that they have

the opportunity to cultivate rice. Since rice cultivation provides men

with cash that they use to purchase cattle for their own bridewealth

payments, a junior man might not be forced to work (or would work less)

on the collective field of the compound head in return for the compound

head's assistance with the payment of bridewealth. Thus there may be a

breakdown in the reciprocal obligations between junior and senior men:

the junior has less of an obligation to work on the collective field of

the senior because he is no longer as dependent on the senior for

bridewealth.

Except for the work done on the collective field, most of the

work done by a compound member is on his or her own sorghum field. As

Table 3.1 shows, the overwhelming proportion of the time women spent

cultivating sorghum is on their own fields. Women from rice-cultivating

households in Widigue and Vele spent proportionately less time on their

own fields than women from Zebe or from nonrice-cultivating households

in Widigue because the labor allocation data for the latter do not

include sorghum harvesting activities. Only about half the time Vele


-26-









and Widigue rice-cultivating women spent harvesting was on their own

fields, which lowered the percentage of time overall that they spent

cultivating their own fields. Since there are few other agricultural

activities competing for women's labor during the period of sorghum

harvesting, women have more time available to "help out" other compound

and family members.




Table 3.1 Percentage of Time Women Spend Cultivating
Sorghum According to Field Proprietorship

a a
Proprietor of-Field Zebe Widigue Widigue Vele
on which Woman Worked nonrice rice
cultivators cultivators



1. Woman's own field 94% 95% 86% 86%

2. Collective/husband 6% 5% 9% 9%

3. Other compound <1% 0% 2% 1$

4. Woman's family <1% <1% 2% 4%

5. Other noncompound <1% 0% 1% <1%


Labor allocation data were collected in Zebe and'Widigue (nonrice) only
until the end of the sorghum weeding period. Thus, percentages for those
villages do not take account of harvesting time.



Except for the one day one woman participated in a work party,

all the work that women did on the fields of other compound members was

considered to be goutna, which refers to labor which is freely given

without any expectation of return, or at least immediate return,

especially in cash. For the work that a woman does on her own family's


-27- -









fields, she often receives some grain at the end of the harvest.

Occasionally a person asks one of his or her friends or

kinsmen to organize a work party if there is a particularly urgent task

to be done which for reasons of illness, for example, he or she has not

been able to accomplish. No money is paid for this sort of work but at

the end of the work day the proprietor is expected to slaughter an

animal and provide a big meal. This kind of aid is called depma.

Villagers report that this practice is becoming much less common since

people would rather work for cash, a practice which has become

widespread with the advent of rice cultivation. Women as well as men

can call or participate in work parties, but out of the entire sample of

women only one worked on a friend's sorghum field for a single day as

depma.

Several instances were encountered where labor was hired for a

cash wage for sorghum cultivation. Labor which is remunerated in cash,

usually at the end of the work day, is called kerena. People hire

kerena labor only when there is pressing work to be done. Several of

the older widows in Zebe reported hiring kerena for planting or weeding

their sorghum fields, but none of the women surveyed worked as kerena on

sorghum.

Capital. There is little capital investment in sorghum

production. Biological and chemical fertilizers are hardly ever

purchased, though fertilizer sold to farmers for cotton production is

reportedly diverted to sorghum fields on occasion. The tools employed

are rudimentary. Seed is usually saved from the previous harvest. A

few households own or rent animal traction to plow their cotton fields,


-28-










and more rarely, their sorghum fields. In any event, the use of animal

traction is minimal at the present time.



3.2. The Organization of Rice Production

The installation of SEMRY. Rice cultivation for industrial

purposes began in the Yagoua area in the early 1950s under the direction

of the SAP (la Societe Africaine de la Prevoyance). SAP was given

responsibility for the production, milling and marketing of rice grown

in the area. However, when more ambitious plans were formulated which

called for a large capital investment in irrigation works, the

responsibility for rice production was vested in a project authority,

SEMRY, le Secteur Experimental de Modernisation de la Riziculture de

Yagoua. Under SEMRY's auspices, a dike was constructed along the Logone

River to control flooding and about 2000 hectares of land were developed

for gravity-fed irrigated rice production by the late 1950s.

Developing new land and irrigation works and rehabilitating the

old was an easy task for SEMRY compared to convincing the inhabitants of

the region to grow rice. The Massa had little interest in cultivating

rice. The village chiefs, who were installed by colonial authorities

following the pacification of the region in the early 1900s, were

charged with assuring that every taxable man cultivated a "piquet," a

half hectare of rice, and that he cultivated it acceptably. The chiefs

seemed to have benefitted most from rice cultivation; de Garine states

that in 1958 the chief in Doreissou (the village immediately to the

north of Vele) had 27 piquets which villages were obliged to cultivate

(1964:94). Farmers were also fined for failing to cultivate with


-29-










and more rarely, their sorghum fields. In any event, the use of animal

traction is minimal at the present time.



3.2. The Organization of Rice Production

The installation of SEMRY. Rice cultivation for industrial

purposes began in the Yagoua area in the early 1950s under the direction

of the SAP (la Societe Africaine de la Prevoyance). SAP was given

responsibility for the production, milling and marketing of rice grown

in the area. However, when more ambitious plans were formulated which

called for a large capital investment in irrigation works, the

responsibility for rice production was vested in a project authority,

SEMRY, le Secteur Experimental de Modernisation de la Riziculture de

Yagoua. Under SEMRY's auspices, a dike was constructed along the Logone

River to control flooding and about 2000 hectares of land were developed

for gravity-fed irrigated rice production by the late 1950s.

Developing new land and irrigation works and rehabilitating the

old was an easy task for SEMRY compared to convincing the inhabitants of

the region to grow rice. The Massa had little interest in cultivating

rice. The village chiefs, who were installed by colonial authorities

following the pacification of the region in the early 1900s, were

charged with assuring that every taxable man cultivated a "piquet," a

half hectare of rice, and that he cultivated it acceptably. The chiefs

seemed to have benefitted most from rice cultivation; de Garine states

that in 1958 the chief in Doreissou (the village immediately to the

north of Vele) had 27 piquets which villages were obliged to cultivate

(1964:94). Farmers were also fined for failing to cultivate with


-29-










appropriate dispatch. De Garine (1964:109), on the other hand,

attributes the lack of interest manifested by the Massa in rice

cultivation to the fact that they had sources of income, such as

fishing, which were more remunerative and far less demanding than rice

cultivation, and furthermore were beyond the purview of the village

chiefs.

SEMRY's current problems. Twenty-five years later, however,

SEMRY is still having trouble attracting farmers despite an extensive

investment program and a good record of high yields. With a loan from

the IBRD, a major program of construction of new irrigation works,

development of new fields and rehabilitation of the old was undertaken

in the early 1970s. In its present incarnation SEMRY (reconstituted in

1971 as the Societe d'Expansion et Modernisation de la Riziculture de

Yagoua) is a semi-autonomous government corporation which oversees the

management of about 5400 hectares of pump-irrigated rice fields.

Farmers' yields have increased from about 1 t/ha in the 1960s to the

present level of about 4.3 t/ha in the rainy season and 5.5 t/ha in the

dry season. The increase is due to better control over the supply of

water, the adoption of transplanting instead of broadcasting and

farmers' increased interest in rice cultivation.

Despite these advances, every year many fields go uncultivated

for lack of farmer interest. The best rainy season to date was in 1977

when 3925 hectares were cultivated out of a possible 5019. In the 1981

rainy season, however, only 3228 hectares were cultivated, despite the

increase in the producer price in 1980.

To attract more farmers, the producer price of paddy was raised


-30-










by 45% in late 1980, although the fixed charges were raised by 57%. The
2
net gain for the producer was positive, however. SEMRY has little room

to manuever in raising the producer price any further. Assuming that

the consumer price of milled rice remained fixed, an increase in the

producer price would reduce the-profit margin on milled rice. It is

not clear that the higher producer price would generate enough of a

supply response, given producer response to date, to compensate for the

smaller profit margin. SEMRY is constrained from raising the consumer

price of rice. Even at present its price is not competitive with that

of imported rice in the southern part of Cameroon. SEMRY rice competes

with imported rice in the North only because it is protected by the high

cost of transporting imported rice from the port of Douala. Despite

its small price advantage in the North, SEMRY has had trouble in the

past clearing the milled rice out of its warehouse. Recently, however,

there has been an increase in consumer demand from Chad (due to

dislocation caused by the war) and from Nigeria (due to exchange rate

fluctuations between the two countries).

At present, SEMRY's revenues are not sufficient to cover both

operating costs and amortization. A recent study commissioned by SEMRY

(S.E.M.R.Y., 1981) concluded that if SEMRY is to become profitable over

the long term it must not only maintain its record of high yields, but

that it must also put all the land which has been developed under

cultivation in the rainy season. This implies a 55% increase in the

area transplanted from the 1981 rainy season level. The study

recommends the formation of mutual guarantee groups which, over the long

run, would progressively be given responsibility for planting the


-31-









nurseries, distributing fertilizer and sacks, assuring payment of the

group's fixed charges and delivery of paddy, and maintaining the small

irrigation works. Mutual guarantee groups would presumably prevent

disputes over water rights from reaching serious levels (in 1980 there

was a dispute over water rights which ended in the death of one of the

disputants) and would encourage farmers to work harder. Evidently, if

SEMRY can make such groups responsible for the fixed charges of all its

members, and in addition take over some of the task which SEMRY

presently pays farmers to do, SEMRY would stand to profit. Such

measures would clearly be in SEMRY's interest, but it is not made

explicit what forms of persuasion will be employed to convince farmers

that organizing collective work groups and taking on the responsibility

for planting seedbeds, for example, is in their own interest. Even if

these measures were successfully instituted, it is questionable whether

SEMRY can ever be financially self-sufficient, given the record of other

rice-producing projects in West Africa.

SEMRY is also considering how to better enforce the contract

between cultivator and SEMRY as another means of increasing profits. At

present, cultivators sign up for fields every season with SEMRY

(although they retain the same fields year after year). They agree to

reimburse SEMRY in kind at the time of sale of their paddy for the

services that SEMRY provides, namely, mechanized plowing and the

provision of seedlings, fertilizer, insecticide, water, extension

services and sacks. The charge for these services is 55,000 CFA per

piquet which farmers pay in paddy at the end of the harvest. At the

current producer price of 55 CFA/kg this amounts to a fixed charge of


-32-










1000 kg of paddy, almost half of the average yield per piquet. Farmers

agree to deliver all but 10% of their total production to SEMRY, which

they are permitted to keep for home consumption. In reality households

retain about 17% of their paddy. The current paddy producer price is

55 CFA/kg, so a farmer whose piquet yields 4.3 t/ha would net about

55,000 CFA per piquet (assuming hired labor and paddy transport charges

of about 15,000 CFA/ha). In comparison, a bicycle costs 40,000 CFA, a

head of cattle in the range of 40,000 CFA and up, the least expensive

six-yard piece of cloth about 2,000 CFA, and a kilogram of sorghum about

75 CFA in the post-harvest period in 1981.

One of SEMRY's major concerns is collecting the fixed charge it

levies on farmers.5 A farmer whose harvest is poor will frequently not

sell his or her paddy directly to SEMRY. Instead the paddy will be

given to a friend to be sold to SEMRY along with the friend's. By not

delivering the paddy directly to SEMRY, the farmer avoids having the

fixed charge deducted for SEMRY's services. The farmer usually receives

the established producer price of 55 CFA/kg from his or her friend--the

friend does not materially benefit from the transaction. Alternatively,

a farmer whose harvest is poor will retain all of his or her paddy and

sell most of it on the parallel market, where paddy prices are about

10-20% greater than the official SEMRY producer price. SEMRY attempts

to stop illegal sales; at harvest time in 1981, as in other years, the

police were mobilized at roadblocks to check whether paddy was being

illegally transported.

Farmers who do not pay their fixed charges are then indebted to

SEMRY the following year. To avoid paying their debts, they will


-33-









relinquish control over their piquets and cultivate a friend's or

relative's, or they will get their wife or son to sign up for a new

piquet. Some farmers reportedly change their names and sign up for new

piquets. Improving the system of farmer identification and registration

of piquets is one area to which SEMRY is currently paying considerable

attention.

Present organization of rice production. At present, SEMRY

controls the dates during which seedlings can be transplanted and paddy,

harvested, which varieties are transplanted, the levels of fertilizer

and insecticide use, the allocation of piquets, the water supply and the

producer price of paddy. About the only aspect of rice cultivation

SEMRY does not control, somewhat to its dismay, is the decision farmers

make of how much and whose labor to allocate to rice production.

In contrast to sorghum fields which are for the most part

cultivated individually, most rice piquets are cultivated jointly by

members of the conjugal household, irrespective of whose name the piquet

is registered in. Women are permitted to and in fact do register for

piquets in their own names; in Vele about 20% of the married women had a

piquet in their own names. A fair number of these piquets were

registered in women's names, I think, for reasons of past indebtedness

of other household members or as a means of risk-spreading.

Even if a piquet is registered in a woman's name and she

receives the money from the sale of her paddy to SEMRY, her husband

expects her to turn over all the income from her field to him. He then

returns part of the income to her (see Section 4). It is difficult for

her to hide from him how much she receives; the paddy is weighed in


-34-










public and the producer is given a slip which records the quantity sold

and the price received. Besides, farmers know approximately how much a

sack of paddy is worth. A woman's husband has the right to any income

that she earns above and beyond the small sums she is allowed to retain

for her daily food purchases. Rice cultivation is analogous to women's

cultivation of tobacco, in that a woman must turn over the proceeds from

the sale of her tobacco if they are greater than the small sums she

needs for daily food expenditures.

In the few cases where men do cultivate tobacco (observed in

Zebe where tobacco is avidly pursued as a cash crop), they cultivate

separately from their wives. This reflects two factors. First, tobacco

fields are quite small (about .01 or .02 ha) and are cultivated quite

intensively. Second, tobacco production has traditionally been

exclusively a woman's crop.

In contrast, rice cultivation is a joint household activity.

Even in cases where a household cultivates more than one piquet a

husband and wife usually work the piquets together. One piquet may not

be ready quite as soon as another for transplanting, for example, so it

makes sense to finish transplanting one before beginning the next.

Also, should one spouse fall ill before work starts on the second

piquet, then the second piquet could be abandoned and the household

would not be liable for the fixed charges. The fact that rice

cultivation is a joint household activity may in part be a reflection of

the scale on which it is carried out; the basic field size is one

piquet (0.5 ha), although it is possible to cultivate half-piquets.

Unlike sorghum cultivation, no cases were encountered of piquets


-35-









worked collectively by a multi-household compound. Although

occasionally one compound member will come to the aid of another, there

is no expectation that all compound members should work on a collective

piquet under the direction of the compound head. The compound head has

rights to the labor of his married sons and their wives only for the

cultivation of the collective sorghum field which produces grain that is

distributed to the entire compound. He has no right to the income

otherwise produced by his married sons or their wives. Since rice

production is largely a commercial rather than a subsistence activity,

there would be no precedent for establishing a collective rice field.

Thus, the rice fields of compound heads are not functionally equivalent

to the collective sorghum field in that they do not produce grain which

is eaten as a last resort when supplies from individual fields are

exhausted. Nor are they symbolically equivalent in the sense that the

land surrounding the compound, unlike the land on which rice is

cultivated, is part of the compound's patrimony which cannot be

alienated and under the ritual control of the bum nagata (head of the

lineage's land).

Most of a household member's time is spent, therefore,

cultivating the piquets belonging to his or her household. Table 3.2

indicates the percentage of time women spent cultivating rice broken

down by the proprietorship of the piquet. In contrast to sorghum

production, a greater percentage of women's rice cultivation was spent

cultivating the fields of people unrelated to them.


-36-












Table 3.2 Allocation of Women's Rice Labor
According to Field Proprietorship


Proprietor of Field Vele
on which Woman Worked


Household


Other Compound


Woman's Family


Other Noncompound


88%


1%


1%


10%


-37-


Widigue


93%


2%


1%


4%









However, a substantial part of the time women spent cultivating

the rice fields of people unrelated to them was remunerated in cash

(kerena). Table 3.3 shows how women were remunerated for the labor they

contributed to fields not belonging to their households. Unlike

sorghum, in a number of cases women actually worked as kerena on other

compound members' fields or even on fields of their own families,

although the majority of the labor to their natal families was goutna.


Table 3.3 Remuneration of Women's Nonhousehold Rice Labor


Proprietor of Field Type of Vele Widigue
on which Woman Worked Remuneration


A. Other compound goutna 87% 94%
depma 3% 0%
kerena 10% 6%

B. Woman's Family goutna 81% 100%
depma 0% 0%
kerena 19% 0%

C. Other noncompound goutna 32% 9%
depma 2% 0%
kerena 66% 91%


Note: Goutna is aid freely given, sometimes remunerated in rice,
depma is a work party, remunerated in food, and
kerena is hired labor remunerated in cash.



A substantial percentage of the work women did on noncompound

fields was kerena. In particular, most of the transplanting and all of

the weeding that they did on noncompound fields was kerena, whereas

about half of the noncompound harvesting and threshing labor they


-38-










performed was goutna. The percentages are given in Table 3.4. The

reason that a much higher percentage of labor at harvesting and

threshing time is goutna and not kerena is that women who thresh as

goutna receive several kilos at the end of the work day. It is not

considered kerena labor, however, because they do not receive cash, and

also because there is no exact expectation as.to the amount of grain

which should be received for a day's labor.




Table 3.4 Type of Remuneration of Vele Women's Other
Noncompound Rice Labor According to Activity


Type of Labor transplanting* weeding harvesting
and
threshing


goutna 9% 0% 49%

depma 2% 0% 3%

kerena 88% 100% 48%


Does not sum to 100% due to rounding errors.



Summary of the major differences between sorghum and rice

production. The major difference between sorghum and rice production not

directly attributable to SEMRY's control over rice production is the

unit of production: in most cases rice fields are cultivated by the

conjugual household while sorghum is cultivated individually and

collectively by the compound. Furthermore, in contrast to sorghum

production, women spend more time working on the fields of noncompound

members and are remunerated in cash for a substantial part of their


-39-











noncompound labor. The differences in the pattern of labor organization

reflect the difference in the use to which the two crops are put.

Most sorghum is destined for subsistence, rather than as a means

of accumulating wealth. Even though sorghum is primarily produced by

compound members working on their individual fields, they do not have

complete control over the disposition of the sorghum harvest. There is

a complex set of norms about how sorghum is stored, whose sorghum is

eaten first, and rights to dispose of surplus production (see Section

4.2). Most sorghum production is individually produced because there is

very little to be gained by one household member mobilizing the labor of

another. As De Garine points out, even before rice production was

introduced, the privileged position of a polygynist was not based on his

capacity to mobilize his wives' labor for surplus sorghum production, but

rather on his ability to mobilize the surplus they generated from

activities such as beer-brewing and tobacco cultivation (1964:131).6

The mobilization of women's labor by.their husbands for rice production

is an extension, therefore, of a husband's customary right to the

surplus generated by his wife's non-subsistence crop labor. As one

might expect, then, the conflict between husband and wife is not over

his right to mobilize her labor per se, but rather over the disposition

of the surplus from rice production, that is, the income remaining once

subsistence needs have been met.



3.3. The Agricultural Calendar

SEMRY determines the date on which transplanting can begin by

when it makes seedlings available to farmers. God or the rains, for in


-40-










the Massa language they are one and the same, determines when sorghum

planting begins. In 1981, the rains were quite erratic at the beginning

of the season. A promising beginning in early May was followed by a

disappointing five-week drought. It was not until the last week of June

that the rains finally began with any regularity. A comparison of the

1981 rainfall and the thirty-year mean rainfall presented in Table 3.5

shows that in fact June was unusually dry.



Table 3.5 Rainfall in the Yagoua Area, 1981


Mar. Apr. May June July Aug. Sep. Oct. Total


Yagoua station 82 37 163 199 204 6 692

Average monthly
rainfall 1934-65 4 19 59 117 181 272 131 31 814


The late arrival of the rains effectively limited sorghum

planting to a three week period from about June 24 to July 14. However,

the rice seedlings were also available for transplanting at the end of

June. Nevertheless, most people chose to finish planting their sorghum

fields before beginning transplanting. Thus, transplanting did not get

seriously underway until about the middle of July. The major conflict

between rice and sorghum occurred in late July and early August, when

farmers were forced to choose between weeding their sorghum fields and

transplanting their rice piquets. Rice weeding, attaching of sorghum

(tying several stalks of sorghum together to protect them from wind

damage and lodging) and, for those households with access to dongolonga

fields (primarily rice-cultivating households in Widigue), field


-41-










preparation and transplanting of dongolonga overlapped in September and

early October. Sorghum was harvested in October and rice was harvested

in November and December.

The rainy season agricultural cycle has been divided up into

seven periods, each one encompassing the great majority of time spent on

a sorghum or rice activity. The 1981 labor calendar is summarized in

Table 3.6. The figures which follow each activity refer to the

percentage of the total number of days that women spent on that activity

that fell in the particular period under question. For example, from

July 15 to July 31, 36% of the rice transplanting done by women in Vele

was accomplished. There is only one major difference in the labor

allocation pattern anong the three groups of women. Vele women

apparently weeded a greater percentage of their sorghum crop earlier

than the women in Widigue. In part, this is due to the fact that they

spent less time weeding overall. However, they also spent more total

hours weeding in late June and early July period, possibly because they

finished planting earlier and did some weeding before starting

transplanting. Part of the difference may also be attributable to

coding errors or to errors in responses.

Although the sequencing of activities was quite similar in the

two rice-cultivating villages, rice-cultivating women in Widigue and

Vele allocated their time quite differently between sorghum and rice as

a comparison of Table 3.7 and Table 3.8 shows. The most pronounced

difference is that Vele women spent almost twice as much time weeding

their rice fields as Widigue women. Since the number of piquets per

worker cultivated by the households to which the women belonged was


-42-












Table 3.6 Cropping Calendar


Period Activity Percentage of Activity
Accomplished in Period


Vele Widigue Widigue
Rice Nonrice
Farmers Farmers


May 15-June 22



June 23-July 14




July 15-July 31





Aug. 1-Aug. 27




Aug. 28-Oct. 10





Oct. 11-Nov. 2



Nov. 3-Dec. 31


harvest of dry
season rice crop
sorghum planting

sorghum planting
1st sorghum weeding
rice transplanting


1st sorghum weeding
2nd sorghum weeding
rice transplanting



1st sorghum weeding
2nd sorghum weeding
rice transplanting


rice weeding
sorghum attaching
dongolonga field
prep. & trans.


sorghum harvest
dongolonga field
prep & trans.

rice harvesting
and threshing


-43-


100%
29%


68%
2%
16%


58%
6%
38%



39%
88%
41%


38%

58%
5%



58%
4%




37%
96%


100%
37%

61%
26%
12%


54%
8%
36%



21%
91%
47%


83%
100%

51%


76%

49%


93%


76%
100%

63%


86%

37%


96%











Table 3.7 Days Worked by Vele Women, 1981


5/15 6/23 7/15 8/1 8/28 10/3 11/3 Total
All Fields to to to to to to to
6/22 7/14 7/31 8/27 10/2 11/2 12/31


DRY SEASON RICE 11.9 11.9


SORGHUM
planting 5.1 8.4 .2 13.7
weeding 1.8 4.1 4.9 10.9
attach & harv. 2.1 3.6 5.7


RAINY SEASON RICE
transplant 2.2 6.7 8.6 1.0 18.6
weeding <.1 2.1 11.3 .2 13.5
harv. & thresh 1.8 25.8 27.6


DONGOLONGA 1.8 1.7 .3 3.8




TOTAL SORGUM 5.1 10.2 4.4 4.9 2.1 3.6 30.3


TOTAL RAINY
SEASON RICE 2.2 6.8 10.7 12.2 2.0 25.8 59.7



TOTAL 16.9 12.5 11.1 15.6 16.2 7.5 26.1 106.0
(hours/day) 4.3 5.7 6.5 5.8 4.5 2.4 4.4 4.6


-44-











Table 3.8 Days Worked by Widigue Rice-Cultivating Women, 1981


5/15 6/23 7/15 8/1 8/28 10/3 11/3 Total
All Fields to to to to to to to
6/22 7/14 7/31 8/27 10/2 11/2 12/31


DRY SEASON RICE 5.9 5.9


SORGHUM
planting 4.7 11.1 .6 16.4
weeding .1 .2 5.3 7.0 12.6
attach & harv. 2.2 3.2 .5 6.0


RAINY SEASON RICE
transplant 3.0 7.2 7.7 1.0 18.8
weeding .1 1.6 5.4 .1 7.1
harv. & thresh 1.1 24.4 25.5


DONGOLONGA .1 12.6 7.3 2.2 22.2




TOTAL SORGUM 4.8 11.3 5.9 7.0 2.2 3.2 .5 34.8


TOTAL RAINY
SEASON RICE 3.0 7.2 9.3 6.4 1.2 24.4 51.5



TOTAL 10.8 14.3 13.2 16.4 21.3 11.6 27.4 114.9
(hours/day) 2.8 6.5 7.8 6.1 5.9 3.7 4.6 5.0










virtually the same, the difference in the amount of time spent weeding

suggests that Widigue households weeded their rice less intensively than

Vele households. Instead.of weeding rice, however, they were

transplanting dongolonga. In fact, Widigue women spent 500% more time

than Vele women cultivating dongolonga. Not all households in Vele have

access to dongolonga land and the fields of those which do tend to be

smaller on average.

Widigue rice-cultivating women also spent several more days

planting and weeding their sorghum fields than Vele women, although

women in both villages spent the same amount of time transplanting rice.

Because sorghum land is readily available, in Widigue and no other

equally remunerative income-generating activities exist, Widigue women

spent several more days than Vele women on agricultural activities

during the period of peak labor demand. They earn most of their income

from sales of surplus grain. In contrast, Vele women, whose access to

sorghum land is more limited, have better opportunities to work as hired

labor on rice production throughout the year and a bigger market for

their sorghum beer (made in virtually all cases from purchased sorghum).

Thus, they are less dependent on grain production as a means of

generating cash for purchases of sauce ingredients.

Trade-offs between sorghum and rice cultivation. Since no

baseline data on women's labor allocation are available, one cannot

determine what activities women actually forwent in order to take on

rice cultivation. It would be useful to know what combination of

childcare, food preparation, leisure, farming, other income generating

activities, etc. women forwent, since the welfare consequences of


-46-









reducing the amount of time spent on one or more of these activities for

a woman and her household depends on the particular activity which is

actually forgone. In the absence of baseline data, inferences about

changes in women's labor allocation must be based on qualitative

evidence as well as cross-sectional data. The labor allocation patterns

of women from rice and nonrice-cultivating households in the sane

village, Widigue, are compared. The data were collected during the

period of peak labor demand, from the middle of May when the rains

started to the end of August when women finished weeding their sorghum

fields. They are summarized in Table 3.9.

A comparison of the total number of hours spent on agricultural

work by Widigue nonrice and rice-cultivating women during the period of

rice transplanting, from the end of June until the end of August, shows

that rice cultivators did not work significantly more time in total than

nonrice cultivators-about two days more over a nine and a half week

period. This suggests that in the period of peak labor demand rice

cultivation has substituted for, not added to, women's other

agricultural activities. Indeed, some of the rice cultivators in

Widigue reported that they had given up cultivating one of their sorghum

fields in order to take up rice cultivation. The following regression,

based on the labor inputs of all Widigue and Vele women (n=72), shows

that women make a one-to-one tradeoff between the number of days they

work on rice and sorghum cultivation during the peak transplanting

period (7/15-8/27):


(3.1) DAYS RICE = 28.57 1.04 (DAYS SORGHUM) R2 .77
(t-ratios) (26.94) (15.20) F=230.93


-47-











Table 3.9 Days Worked by Women from Widigue Rice
and Nonrice-Cultivating Households, 1981


5/15- 6/23- 7/15- 8/1- Total
6/22 7/14 7/31 8/27


ALL CROPS (all fields)
Rice 10.8 14.3 13.2 16.4 54.6
Nonrice 10.3 14.5 12.3 15.2 52.3
(.72) (.73) (.15) (.36) (.45)

SORGHUM (all fields)
Rice 4.0 11.3 5.9 7.0 29.2
Nonrice .8.2 14.2 11.8 13.4 48.8
(.00) (.01) (.00) (.00) (.00)

RICE (all fields)
Rice 5.9 3.0 7.2 9.3 25.4
Nonrice 1.4 .1 .1 .3 1.9
(.01) (.00) (.00) (.00 (.00)


SORGHUM PLANTING
(own field)
Rice 4.0 10.2 .5 14.7
Nonrice 8.0 13.3 .9 22.2
(.00) (.00) (.36) (.00)

SORGHUM WEEDING
(own field)
Rice .0 .2 4.8 6.8 11.8
Nonrice .2 .9 10.4 13.1 24.5
(.31) (.04) (.00) (.12) (.00)

SORGHUM 1ST WEEDING
(own field)
Rice .0 .2 4.6 3.3 8.0
Nonrice .2 .9 10.1 5.8 16.9
(.31) (.05) (.00) (.12) (.00)

SORGHUM 2ND WEEDING
(own field)
Rice .3 3.4 3.8
Nonrice .3 7.3 7.6
(.40) (.00) (.00)


Note: The figures in parentheses are the probabilities associated with
the two-tailed t-test used to test the hypothesis that the means of the
two groups are equal.


-48-











It is not surprising that women would make a tradeoff between

transplanting rice and weeding sorghum, since the agricultural workload

of women who cultivate only sorghum is already quite heavy. On average,

they spent six to seven hours a day, seven days a week from.the end of

June to the end of August cultivating. In addition, they spent about

another three hours a day on domestic labor activities, threshing grain,

grinding it, finding ingredients for the sauce, preparing.meals,

fetching water, bathing children, etc., all of which are essential to

the maintainence of their households.

Thus, women who undertake rice cultivation have no choice but to

give up some of the time they spend cultivating sorghum. It is

important to note, however, that rice cultivation has increased women's

agricultural workload only in terms of the number of days worked

annually and not in terms of the number of hours worked per day on

average during the period of peak of labor demand. The increase in the

amount of time women spend on agricultural activities is not necessarily

to their detriment, provided that they receive a corresponding increase

in real income. This issue is taken up in the next section.




-49-











SECTION 4. THE ALLOCATION OF WOMEN'S LABOR TO RICE PRODUCTION



This section examines the relationship between the amount of

compensation women receive from their husbands for their labor on rice

production and how they allocate their labor between rice and sorghum

production. Section 4.1 establishes that the amount of compensation

women receive is related to the number of days they work. It then

considers whether the amount of compensation women receive for their

work on rice production is greater than the income they forgo. Section

4.2 examines whether, with the introduction of irrigated rice

cultivation, the share of food women are expected to provide has

increased. If so, then the real value of the compensation they receive

from their husbands is less than the nominal value. However, the

evidence indicates that rice cultivation has increased women's real

income.

Given this apparent incentive, Section 4.3 examines the

allocation of women's labor between sorghum and rice production. It

compares the labor allocation pattern of women who control the disposi-

tion of rice income with the labor allocation pattern of women whose

husbands control the disposition of income from rice production. Even

though women in the latter group receive more income from rice

production than they forgo on account of rice cultivation, they do not

allocate their labor efficiently relative to the former. This

conclusion is the basis for an alternative approach to intrahousehold

resource allocation and income distribution based on bargaining theory.

This approach is outlined in the fourth section. From the perspective


-50-










of a bargaining model of household decision-making, Section 4.4

speculates on the factors that differentiate the allocatively efficient

married households from the allocatively inefficient ones.



4.1. The Remuneration of Women's Rice Cultivation Labor

The money and paddy which women receive from their husbands

after the rice harvest is perceived by them to be compensation for the

work which they do on their husbands' fields. It is given to them "in

return for their sweat." A woman receives about 7,700 CFA in cash and

about 9,200 CFA worth of paddy from her husband after the harvest, or

about 16,900 CFA in total. This is less than a quarter of the net

returns from rice production--about 70,000 CFA. Valued at the market

wage rates (see below), a woman's labor contribution is worth about

31,200 CFA, so her husband makes a profit of about 14,300 CFA from her

labor.

If a woman receives what she considers to be an insultingly

small sum of money, or no money at all, she is likely to become quite

angry with her husband and most unenthusiastic about participating in

rice production the following year. Husbands are quite aware that their

wives' continued participation depends on their own generosity. Of the

thirty-five married women in the sample from Vele and Widigue whose

husbands controlled the distribution of income from rice production,

only three women did not receive any cash at all.

In one case, the husband received no cash from the sale of paddy

since the harvest had been very poor. In the other two cases, however,

the women were very angry at not receiving any cash. One woman's


-51-










husband offered her 1,000 CFA which she refused to take. A fight

ensued. I visited the other woman several days after the rice harvest.

At that time she was waiting for her husband to decide how much money he

was going to give her. I went back to see her about two weeks later and

found out that she was at her parent's compound where she was

recuperating from a severe beating by her husband. Her co-wife

explained that their husband had beaten her because he was angry that

she hadn't prepared food for him for two days. This is one of the most

effective ways women have of making their displeasure with their

husbands known. The fact that almost all women did receive cash (except

in the case where there were extenuating circumstances) and that there

was considerable conflict in the two cases where they did not indicate

that women do have some claim on the money which is earned from rice

cultivation.

The following regression is estimated to establish that there is

a significant relationship between the amount of compensation women

receive from their husbands and the number of days they worked on their

husbands rice fields:2


(4.1) COMPENSATION = -1922 + 358 (DAYS) R2=.70
(t-ratios) (.82) (8.21) F=67.36


The rate at which they are compensated by their husbands, 358 CFA per

day, is significantly less than the the average returns to labor from

rice cultivation which are about 600 CFA/day and is also significantly

less than the average wage of 600 CFA/day which women would have

received had they been compensated for their labor on rice production at

the market wage rates. The average wage rates paid to Vele women for


-52-










transplanting, weeding and harvesting were 805 CFA/day (n=55), 523

CFA/day (n=26), and 501 CFA/day (n=67) respectively. Households

generally hire labor to replace ill household members, when they want to

finish a task quickly or when they find they cannot complete an

activity. However, as one would expect in a land surplus area, hired

labor comprises only about 10% of the total labor input. About 90% of

the hired labor is female.

One might wonder why women continue to work for their husbands

if the rate at which they are compensated .is so low relative to what

they could earn working as hired labor. In principle, married women are

expected to work on their husband's fields if they are not working on

their own. If they refuse to work on their husbands' fields, they risk

a beating. Thus, most women work as hired labor no more than several

days out of the entire agricultural season, most often when their need

for cash (usually to purchase food) is urgent. In order to profit from

their wives' labor, husbands must restrict their wives' opportunities to

work as hired labor.

However, several of the married women worked more than several

days as hired labor. Five of the fifteen married women in Vele who

worked as hired labor accounted for 67% of the total number of days

worked by the group of twenty-four married women whose husbands

controlled the disposition of the income from rice production. All five

women received less than the average rate of compensation from their

husbands. In fact, the woman who worked the most days as hired labor

(22% of the total days worked) was the one who received no cash from her

husband at the end of the harvest and was later severely beaten by him.


-53-









It is possible that husbands reduce the amount of income they

give to their wives after the rice harvest to express their displeasure

with their wives' choice to work as hired labor instead of working for

them. However, it is also possible that the causality also goes in the

opposite direction: women who spend more than several days working as

hired labor may do so in order to express their displeasure with the

amount of compensation they have received in the past from their

husbands. In fact, several women told me that if a woman is not

adequately compensated for her labor by her husband, then she will spend

more time working as hired labor the following year. Even so, women may

stagger the days that they work as hired labor or do their work in the

harvesting season so as not to unduly provoke their husbands. The woman

who worked the most days as hired labor only spent several days working

as hired labor during the transplanting and weeding period. She spent

very little time transplanting and weeding her husband's rice field.

Most of the days that she worked as hired labor were during the

harvesting period, since harvesting was quickly accomplished on her

household rice field because it had been only partially transplanted.

Opportunity cost of women's labor. Thus, unless they are

willing to provoke a serious conflict, women have little choice but to

work for their husbands in order to profit from the higher returns to

labor afforded by rice cultivation. The issue is whether they earn more

income from working for their husbands than they could earn from their

own income-generating activities. To determine if rice cultivation has

increased their incomes, the rate at which they are compensated by their

husbands needs to be compared to the opportunity cost of their labor.


-54-










However, the opportunity cost of their labor varies throughout the

agricultural season.

The discussion of the agricultural calendar in Section 3.3

indicates that rice transplanting competes with sorghum planting and

weeding labor. The average returns to labor and the hired wage rates

can be used to establish the opportunity cost of sorghum planting and

weeding labor. Labor is occasionally hired for sorghum planting or

weeding in villages that are located at some distance from the rice

fields. People from those villages have little incentive to take

advantage of the higher wage rate paid for transplanting because of the

time required to walk to the fields. The wage rate paid for sorghum

planting and first weeding labor are in the range of 450-550 CFA/day.5

The returns to sorghum labor also fall in this range.

However, rice weeding and harvesting do not compete with sorghum

for labor. Thus, if women did not weed or harvest their husbands' rice

fields, they would otherwise most likely be earning income from

beer-brewing, fabrication of clay pots, petty commerce, etc. The survey

of women's earnings in various periods throughout the year indicates

that women rarely earn more than the equivalent of about 100 CFA/day

even in relatively slack agricultural periods. This is true even in

Zebe, where women's earnings from sales of tobacco and fish are

considerable. Thus, 100 CFA/day can safely be taken as the upper limit
7
on the opportunity cost of women's nonagricultural labor.

To determine how much additional income Vele women actually

earned by working on their husbands' rice fields instead of pursuing

their own activities, the number of days women spent transplanting,


-55-









weeding and harvesting their husbands' rice fields, 16.6, 13.3 and 21.8

days respectively, are multiplied by the opportunity cost of that labor,

550, 100 and 100 CFA/day, to obtain 12,640 CFA. Vele women received on

average 16,900 CFA from their husbands. Thus, women received 4,260 CFA

more by working on their husbands rice fields than if they had engaged
8
in their own income-generating activities. This should be contrasted

to the difference of 14,300 CFA between the market value of women's

household rice labor and the amount of cash they receive from their

husbands. Thus, even though women have captured less than a quarter of

the net increase in household income generated by their labor on rice

cultivation, it is, nonetheless, a net increase in their income.



4.2. The Intrahousehold Responsibility for Food Provision

The last section established that women are compensated for

their labor on rice production at a rate which is greater than the

opportunity cost of their time. If, however, they are expected to use

the cash they receive to purchase food which their husbands otherwise

would have provided or purchased before rice cultivation was adopted,

then the real value of their compensation is reduced. It has been

alleged (see Section 5) that cash crop production puts a greater burden

on women to provide food. Essentially, the argument is that men

withdraw their labor from food production and use the income they

receive from cash crop production for nonfood expenditures. This

section considers the question of whether rice cultivation has altered

the division of responsibility for providing food. It begins by

describing the Massa diet, which is based primarily on a cereal dish


-56-










accompanied by a sauce which often contains a small quantity of fish.

It then considers whether rice cultivation has resulted in a decreased

supply of grain available for home consumption. Finally, it considers

whether there have been any shifts in the division of responsibility for

purchasing sauce ingredients.

The data on food expenditures were collected during five

different periods of the year to take account of the seasonal variation

in food expenditures. The survey was conducted 1) in early May, before

the dry season rice harvest, 2) in early July, after the dry season rice

harvest, 3) in late August, before the sorghum harvest, 4) in November,

after the sorghum harvest, and 5) in early January, after the rainy

season rice harvest. Women were interviewed every other day for two

weeks during each of these periods about the amount they and other

members of their households spent on grain and ingredients for the

sauce. In addition, the amount of income women earned., as well as its

source, were also ascertained.

In the first two rounds of the expenditure survey, women were

asked how much they and other members of their households spent on food.

Expenditures were broken down into expenditures on sauce ingredients,

sorghum and rice. However, women were not asked if they had earned the

money they used to purchase food or if the money was a transfer payment

from someone else. After the second round of the survey, however, the

questionnaire was revised and women were asked for the source of the

income they used to buy food. If a husband gave his wife money for the

express purpose of buying food, for example, then he was considered to

have made the purchase. However, the lump cash sum a woman received


-57-










from her husband for her labor on rice cultivation was considered to be

her own earned income, since the money was given in return for her labor

without stipulation on how it should be spent. Thus, in the first two

survey rounds purchases were differentiated according to the transactor

of the purchase, whereas in the last three survey rounds purchases were

differentiated not only by transactor, but also by the household member

who provided the income for the purchase.

The Massa diet. The typical meal consists of funa, a cereal

dish prepared by grinding sorghum (and even rice) between two stones

into a flour which is stirred into boiling water until it is of a

dough-like consistency. Pieces of funa are dipped into a sauce, dafna.

Sauces often contain some fish, preferably fresh, but often dried, and

are usually based on some kind of vegetable, such as okra or leaves,

which are gathered by women. Only on rare occasions is meat purchased

or are animals slaughtered, usually for a celebration or to feed a work

party. Fish and okra are the most common items purchased for the sauce,

although they are also frequently home-supplied. Condiments, such as

onions, garlic, and red pepper, are rarely purchased or even employed.

Even oil for the sauce or sugar for the porridges (which are made out of

sorghum or rice flour, leftover pieces of funa, or the crusts which

stick to the side of the cooking pot after the preparation of funa) are

considered luxuries. No purchases of tea or coffee were ever recorded,

though people do buy kola nuts. Thus, most of the diet is home-

produced.

The majority of calories are provided by funa. A food

consumption survey carried out in the late 1950s (before irrigated rice


-58-


* f











production was introduced) found that the Massa consumed about 700 grams

of sorghum a day except during the period immediately preceding the

sorghum harvest when the average cereal consumption dropped to about 500

grams per day per adult (de Garine, 1964:9 and 1977:45) In addition,

they consumed about 100 grams of fish and milk a day. The adult caloric

intake averaged about 3000 cal/day.

Unfortunately, it was not possible to undertake a food

consumption survey to determine how the food consumption pattern has

altered over the past twenty-five years. However, some educated guesses

can be made based a number of interviews that were conducted. People

frequently remarked that the fish population in the Logone River has

declined over the last ten to fifteen years. It is not clear what

effect, if any, pumping water out of the Logone to irrigate the rice

fields has had on the fish population. In any event, the decline in the

fish population makes it more difficult for people to catch fish using

their traditional traps. Thus, they are forced to rely more on

purchased fish caught by men who fish on an extended basis using nylon

nets. This has increased the need for cash to purchase fish and

perhaps resulted in a decrease in the amount of fish consumed. In

addition, some people also stated that grazing land was appropriated for

rice production. As a result, people in villages along the Logone,

particularly towards Yagoua, keep part of their cattle herd with kin who

live in villages where grazing land is more readily available. However,

it is not clear to what extent this has reduced the number of cattle

kept in rice-cultivating villages, and therefore the supply of milk.

The other major change in the food consumption pattern is that rice has


-59-










been substituted to some extent for sorghum. However, rice cultivation

does not seem to have reduced the quantity of grain available for

household consumption. The issue is taken up in the next subsection.

Responsibility for providing grain. Only in the best of years

would the Massa who lived along the floodplain of the Logone have

produced enough sorghum to meet household subsistence needs before the

introduction of rice cultivation (de Garine, 1964:87). Given the

frequent flooding of the Logone before the dike was built, sorghum

harvests were often poor. Deficits were made up with income from sales

of fish, tobacco, or livestock. When there was a serious shortage, the

compound head sold cattle.

The first sorghum which is consumed by the household is usually

that which a woman harvests from her own field. Most married women have

their own granary. However, in monogamous households, it is not

uncommon for husband and wife to store their grain together. If a man

has more than one wife, he often divides up the grain between them. In

a multi-household compound, the compound head stores the grain from the

collective sorghum field in a central granary. This grain is saved

until the hungry season, when it is distributed among the women of the

compound to tide the compound over until the harvest. The fact that men

and women often store their grain together, particularly in monogamous

households in villages which do not regularly produce surpluses of

sorghum, indicates that husband and wife share the responsibility to

produce and contribute grain for family subsistence needs.

On occasion, both men and women sell small quantities of grain

when there is a surplus. Before a household member sells a significant


-60-










amount of grain (say, more than 20 or 30 kilos), he or she discusses the

matter with his or her spouse. As one man said, he would never sell

grain to buy a calf without his wife's approval because the calf would

not be happy in the compound under those circumstances. This is

indicative of the tension that a unilateral decision to sell grain would

engender. Thus, women appear to have a strong voice as far as the

disposition of the sorghum harvest is concerned.

The introduction of rice production does not seem to have

resulted in men abdicating their responsibility to provide grain for

their households. Households in the project area retain as much rice as

the sorghum production which is forgone on account of cultivating rice.

Ideally, one would like to have time series data, but they do not exist.

Thus, cross-sectional data from one village, Widigue, are used to

determine what impact rice cultivation has had on the availability of

grain for household consumption.

The quantity of sorghum forgone by rice-cultivating women in

Widigue can be approximated by comparing the difference in the size of

rice-cultivating and nonrice-cultivating women's sorghum fields. Since

rice and nonrice cultivators planted and weeded their sorghum fields at

approximately the same intensity (planting: rice cultivators 64 days/ha,

nonrice cultivators 50 days/ha, t=.93; weeding: rice cultivators 53

days/ha, nonrice cultivators 54 days/ha, t=.00), the difference between

the two groups is primarily extensive rather than intensive. Nonrice-

cultivating women planted .54 ha in sorghum, and rice cultivators

planted .32 ha. However, 38% of the time nonrice cultivators spent

planting and 29% of the time rice cultivators spent planting was during


-61-









the dry season rice harvest. Assuming a linear relationship between

planting time and field size, nonrice cultivators would have planted .34

ha of sorghum in the rainy season rice transplanting period and rice

cultivators .23 ha. Thus, nonrice-cultivating women planted .11 ha more

of sorghum in the rice transplanting period. Assuming a yield of 1.0

t/ha, Widigue rice cultivating women therefore forwent about 111 kilos

of sorghum, valued at 8,325 CFA at the post-rainy season rice harvest

price of sorghum. Widigue rice-cultivating household retained about

17,500 CFA worth of paddy which is about 160 kg of paddy or about 106 kg

of hand-pounded rice per active household member.

Another method of calculating the amount of sorghum women

forwent on account of rice cultivation is to value their rice labor by

the average returns to sorghum labor. Assuming that the only trade-off

rice cultivators are forced to make between sorghum and rainy season

rice cultivation is during the period of rice transplanting, the

opportunity cost of women's transplanting labor can be valued by the

wage rate for sorghum planting and weeding labor, 500 CFA/day. Since

rice-cultivating women spent 17.3 days transplanting the rice fields of

their households in the sorghum planting and weeding period, they would

have forgone about 8,650 worth of sorghum, about 115 kg. This

calculation is virtually identical to the one based on the area planted.

Thus, women's share of paddy, assumed to be half of the paddy

retained by a monogamous household, is equal to the sorghum production

she forgoes. The question is whether men's share (the other half) is

equal to the sorghum production which men forwent that would have been

available to their wives for household consumption purposes. Data from


-62-










farm management survey indicates that rice-cultivating compounds

cultivate about .27 ha per active compound member, compared to the .32

ha per woman calculated on the basis of the women in my sample (Bikoi

1982). The similarity between these figures, particularly since they

are based on two different random samples, indicates that men's fields

were roughly the same size as women's.9 Thus, men's share of the paddy

which was retained almost certainly corresponds to the sorghum

production they forwent on account of rice.

The issue of who buys grain to make up any deficits in household

production is moot in Widigue, since virtually all households, rice-

cultivating and nonrice-cultivating alike, produce enough grain to meet

their annual subsistence needs. When both groups of women in Widigue

were interviewed at the end of the hungry season immediately prior to

the sorghum harvest, they reported that they had had a sufficient amount

of sorghum to meet household grain needs and had not had to buy grain.

Indeed, Widigue women consistently sold grain over the course of year,

reflecting large supluses. Thus, women did not have to use any of the

cash they received for their rice labor to purchase grain. In fact,

husbands in Widigue could have retained less paddy without compromising

household food supply. However, this would have reduced the amount of

surplus grain women had available for sale and women, therefore, would

have had little incentive to cultivate rice. Furthermore, men have an

interest in ensuring that women have a certain quantity of grain

available for sale, since income from the sale of grain is primarily

used to purchase ingredients for the sauce consumed by both men and

women.


-63-










Unfortunately, it is not possible to compare the amount of grain

retained by rice and nonrice cultivators in Vele, since virtually all

households in Vele cultivate rice. There is some historical evidence,

however, that provides an indication of the extent to which rice

cultivation reduced the area under sorghum cultivation in the villages

bordering on the Logone River. In 1958 de Garine surveyed a small

sample of compounds in Doreissou, the village to the north of Vele, and

found that compounds cultivated on average 0.2 ha of sorghum per active

compound worker (1964:85). In 1981, the average sorghum area under

cultivation was .13 ha per active worker (Bikoi, 1982).10 If this is

any indication of the magnitude of the reduction in the size of

household sorghum fields, then the amount of sorghum production forgone

on account of rice cultivation would certainly be less than the 106 kg.

of hand-pounded rice that Vele households retained per active household

worker.

Furthermore, if husbands are selling off their paddy to acquire

cash to the detriment of their households' subsistence needs, one would

expect that households in which women control the disposition of paddy

would retain more paddy than male-headed households on a consumer

equivalent basis. However, this is not the case. 'Households in which

women controlled the disposition of rice income retained only 167 kg of

paddy per adult compared to the 165 kg of paddy retained by households

in which the disposition was determined by men.

This leaves the question of how rice cultivation has altered the

responsibility for purchasing grain when the home-produced supply is

exhausted. With what they retain from rainy season rice production,


-64-











most households in Vele cannot meet their subsistence requirements from

rainy season sorghum and rice production alone. The average sorghum

area cultivated per consumer equivalent is .09 ha. which at 1 t/ha would

yield about 90 kg of sorghum.1 The additional 70 kg of hand-pounded

rice retained would total 160 kg of grain per consumer equivalent.12

This amounts to amounts to 438 g/day of grain per consumer equivalent,

about 15% less than the minimum level of 500 g/day cited by de Garine.

(If the average sorghum yield is assumed to be 1.6 t/ha, the ration per

consumer equivalent would be 586 g/day).

However, households also depend on their dry season rice

production. Those households whose fields are not irrigated in the dry

season work as hired labor or work for their kin in exchange for a sack

or two of paddy at the end of the dry season rice harvest. In addition,

47% of the households in Vele also have dongolonga fields. These dry

season activities significantly reduce the potential grain deficit faced

by most households.

Nonetheless, about a third of the households surveyed in Vele

(female and male-controlled alike) reported purchasing some grain during

the rainy season to tide them over until the sorghum harvest. In

August, the households in which women controlled the disposition of rice

production purchased 154 CFA of grain (about 2 kg of sorghum), and those

in which men controlled the disposition of income from rice production

purchased 133 CFA of grain over a two-week period, respectively.

However, in both cases average sales of grain outweighed purchases,

which indicates that some households were in surplus and others were in

deficit (in addition, some households were also speculating in grain).


-65-









In any event, the quantities of grain bought and sold were not large in

any case, particularly when compared to purchases of grain in Zebe,

where about two-thirds of the households ran out -of grain well before

August (again, same percentage of male and female headed households).

In May, Zebe households spent 1,114 CFA on grain (approx-imately 13 kg

of sorghum) over a two-week period; in July 253 CFA and in August 1,050
13
CFA. In Zebe, particularly during the hungry season, women bought the

majority of grain using income from the sale of their tobacco

production. This demonstrates that, where necessary, women have played

and continue to play a critical role in ensuring the adequacy of the

household grain supply through their (nonsorghum) income-generating

activities. This casts further doubt on the hypothesis that rice

cultivation has placed a bigger burden on women to supply grain to meet

household subsistence needs.

It is difficult to ascertain how responsibility for meeting

grain deficits would have been divided up between husband and wife in

he past in Vele. At present, however, men make a major contribution by

cultivating dry season rice and dongolonga. In addition men in several

households borrowed money or, more often, rice which they paid back

after the rice harvest with interest. Even in such cases women were

still compensated for their labor on rainy season rice production.

Thus, it seems that men continue to fulfill their traditional

responsibility to provide grain for their households.

Changes in the division of responsibility for sauce ingredients.

Unlike grain, most of the burden of cultivating and purchasing

ingredients for the sauce has traditionally fallen on women, though men


-66-










do provide cash for the purchase of sauce ingredients occasionally. In

August, I conducted an informal survey of'the married women in the

sample to determine what women's perceptions were of their husbands'

contributions. When asked if their husbands purchased ingredients for

the sauce, about two-thirds of the women replied affirmatively.

However, about half of these women then proceeded to qualify their

reponses by adding a proviso to the effect, "when he has the money," or

in several cases, "when he sees that I don't have any money." Some of

the women who said that their husbands did not provide money for food

explained that their husbands were sick or didn't do any work (i.e.

participate in any income-generating activities other than agricultural

work) and thus had no cash.

The impression gained from the informal survey, therefore, was

that men's contributions are neither sufficiently regular nor large to

have significantly lessened the year-round burden on women to provide

sauce ingredients. This was confirmed by the results of the last three

rounds of the expenditure survey, which asked for detailed information

on who actually purchased food and who provided the money which was used

to make the purchase. Women's perceptions of the regularity of their

husbands' contributions were reasonably accurate. Of the forty-one

women who said that their husbands purchased food, twenty-one received

money from their husbands (or their husbands actually purchased food) in

at least two out of the three survey rounds. Twenty women, however,

received money in only one of the survey periods, and one woman received

no money at all from her husband. On the twenty-two who said that their

husbands did not contribute, only four received money in least two out


-67-










of the three survey periods. The husbands of the other eighteen

contributed in only one of the three periods (six husbands) or not at

all (twelve husbands). The results are summarized for each village in

Table 4.1. As it confirms, there are a substantial number of husbands,

both rice and nonrice-cultivating alike, who contribute seldom if ever

to the purchase of sauce ingredients.

Table 4.2 shows the percentages of husbands' and wives'

contributions to food expenditures in August, November and January.

There is considerable variation between months and between villages.

Husbands' contributions are highest in Zebe, in general, because of the

extensive purchases of grain, which require both men's and women's

incomes. Expenditures on food in Widigue are low relative to the other

villages, and husbands' contributions are also generally quite low.

This reflects the greater availability of home-produced ingredients for

the,sauce in Widigue (because of the larger sorghum fields, the

production of intercropped sauce crops is greater), as well as the

relative scarcity of fish available for purchase. Vele husbands may

purchase more fish than Widigue husbands simply because fish is.more

readily available and because their wives incomes' are not sufficient to

cover additional food expenditures.


-68-













Table 4.1 Wives' Perceptions of Frequency of Husbands' Food Purchases
Compared to Observed Frequency of Husbands' Food Purchases


yes yes, no
qualified


All Villages
A. Number of women who responded: 25 16 22
B. Number of husbands who purchased
food in three or two survey periods: 13 8 4
C. Number of husbands who purchased
food in one or no survey period: 12 8 18


Vele
A. Number of women who responded: 11 10 7
B. Number of husbands who purchased
food in three or two survey periods: 7 6 2
C. Number of husbands who purchased
food in one or no survey period: 4 4 5


Zebe
A. Number of women who responded: 6 2 6
B. Number of husbands who purchased
food in three or two survey periods: 5 1 0
C. Number of husbands who purchased
food in one or no survey periods: 1 1 6


Widgue rice cultivators
A. Number of women who responded: 7 3 4
B. Number of husbands who purchased
food in three or two survey periods: 0 1 0
C. Number of husbands who purchased
food in one or no survey periods: 7 2 4


Widigue nonrice cultivators
A. Number of women who responded 1 1 5
B. Number of husbands who purchased
food in three or two survey periods: 1 0 2
C. Number of husbands who purchased
food in one or no survey period: 0 1 3


-69-











Table 4.2 Food Expenditures by Husbands and Wives


August November January

pre-sorghum post-sorghum post-rice
harvest harvest harvest


Vele
wife
husband





Zebe
wife
husband





Widigue
Rice Cultivators
wife
husband





Widigue
Nonrice Cultivators
wife
husband


1798
389
2187


1478
525
2003


839
82
921






463
88
551


(82%)
(18%)






(74%)
(26%)


655
358
1013





1703
146
1849


(90%)
(10%)


(84%)
(16%)


400
23
423


669
400
1069


(65%)
(35%)






(92%)
(8%)







(95%)
(5%)


1045
654
1699





1432
1334
2766






1003
115
1118


(63%)
(37%)


981
289
1270


(60%)
(40%)






(52%)
(48%)







(90%)
(10%)







(77%)
(23%)


-70-











Without baseline data, it is not possible to determine whether

rice cultivation has increased or decreased the magnitude of men's

contribution to sauce ingredients. However, there is evidence

indicating that women would not purchase more sauce ingredients if they

controlled more of the income from rice production. Table 4.3 compares

expenditures on sauce ingredients of households in which women

controlled the disposition of income from rice production with

households in which men controlled the disposition. The difference

inwomen's expenditures on sauce ingredients between the two groups is

especially striking in January, after the rice harvest. However, when

the contribution of husbands who controlled the disposition of income

from the rice harvest are factored in, the gap between the two sets of

households is dramatically reduced and becomes statistically

insignificant. This suggests that the two sets of households have

similar preferences regarding food expenditures. Thus, the gender of

the person who controls the disposition of income from rice production

does not have a significant effect on the amount of income which is

spent on sauce ingredients.4

Husbands' willingness to purchase ingredients for the sauce

essentially gives women latitude to spend their income on nonfood items

without compromising the desired level of food expenditure (relative to

what women who control the disposition of income from rice production

would spend). In fact, about two weeks after they had been given cash

by their husbands for their labor on rice production, women had already

spent 65% of the money on "big ticket" consumption items--mostly

cloth and shoes and in several cases, enamelware as well. This does


-71-












Table 4.3 Food Expenditures of Independent
and Married Women's Households


August November January


pre-sorghum post-sorghum post-rice
harvest harvest harvest


Women's Expenditures
on Sauce Ingredients
Independent 1729 1042 1788
Married 1699, 633 937
(.92) (.21) (.01)


Household Expenditures
on Sauce Ingredients
Independent 1810 1158 1813
Married 2143 .1042 1678
(.32) (.77) (.71)


Household Expenditures
on Sauce Ingredients ,,
(per consumer equivalent)
Independent 775 468 843
Married 799 448 622
(.87) (.90) (.34)



Note: Figures in parentheses are the probabilities associated with
two-tailed t-tests. See note to Table 3.9.

** See note 11 for a definition of consumer equivalent.


-72-










not include what they spent on their children's clothing. Only a few

women reported having spent a thousand or so CFA, out of the seven or

eight thousand CFA they received, on sauce ingredients. Indeed, as

women themselves say, rice production has provided them with the

opportunity to purchase such items. Thus, it appears that the cash

women receive from their husbands represents a real increase in their

income, in that they have more money available to spend on consumer

goods without forgoing expenditures on food. While it is possible that

women may carry a bigger burden of the responsibility for purchasing

sauce ingredients, their real incomes have nevertheless increased.



4.3. Allocation of Labor Between Sorghum and Rice Production

Comparison of married and independent women's labor allocation.

It would seem, therefore, that once some minimum quantity of sorghum is

produced, women would have sufficient incentive to allocate the

remainder of their labor time to their husbands' rice production.

However, the following comparison of the labor allocation pattern of

married women who cultivate for their husbands with that of women who

cultivate for themselves suggests otherwise. Out of the sample of

forty-two women from Vele, thirty-six were included in the comparison.

One woman was excluded because she actually worked on the piquet of her

co-wife (their husband was unable to cultivate), and the other four were

dropped because they were older widows who were only able to help their

sons cultivate their piquets (contrary to what they had stated when the

census was taken). The fifth was not included because she decided not

to cultivate rainy season rice after her husband fell ill.


-73-










not include what they spent on their children's clothing. Only a few

women reported having spent a thousand or so CFA, out of the seven or

eight thousand CFA they received, on sauce ingredients. Indeed, as

women themselves say, rice production has provided them with the

opportunity to purchase such items. Thus, it appears that the cash

women receive from their husbands represents a real increase in their

income, in that they have more money available to spend on consumer

goods without forgoing expenditures on food. While it is possible that

women may carry a bigger burden of the responsibility for purchasing

sauce ingredients, their real incomes have nevertheless increased.



4.3. Allocation of Labor Between Sorghum and Rice Production

Comparison of married and independent women's labor allocation.

It would seem, therefore, that once some minimum quantity of sorghum is

produced, women would have sufficient incentive to allocate the

remainder of their labor time to their husbands' rice production.

However, the following comparison of the labor allocation pattern of

married women who cultivate for their husbands with that of women who

cultivate for themselves suggests otherwise. Out of the sample of

forty-two women from Vele, thirty-six were included in the comparison.

One woman was excluded because she actually worked on the piquet of her

co-wife (their husband was unable to cultivate), and the other four were

dropped because they were older widows who were only able to help their

sons cultivate their piquets (contrary to what they had stated when the

census was taken). The fifth was not included because she decided not

to cultivate rainy season rice after her husband fell ill.


-73-










Of the thirty-six remaining women, twelve controlled the

distribution of the income from the sale of their paddy. Seven of these

were widows who cultivated their own piquets. The other five were women

who had their own piquets and whose husbands were too sick or too old to

cultivate rice. These women hired their own labor and controlled the

distribution of the proceeds received from the sale of their paddy. The

other twenty-four were married women whose husbands, all of whom

cultivated rice, controlled the distribution of the income derived from

the sale of the paddy, which was cultivated jointly by the household.

For the sake of rhetorical simplicity, these two groups of women will be

referred to as independent and married women, although the group of

independent women does contain several married women.

Table 4.4 compares the labor inputs of married and independent

women to sorghum and rice cultivation throughout the rainy season. As

Table 4.4 shows, there is a major difference in the amount of time they

allocate to sorghum and rice production. The difference becomes quite

pronounced in August during the latter half of the transplanting period,

when married women spent far less time transplanting rice and far more

time weeding sorghum than independent women. Married women continued to

spend less time on rice and more time on sorghum than independent women

throughout the rest of the season.

Several factors might account for the difference in the labor

allocation patterns of the two groups. The number of children a woman

has to feed might influence how much time she allocates to sorghum and

rice. Since sorghum is the preferred cereal, a woman might decide to

cultivate as much sorghum land as is necessary (or to cultivate that


-74-









Table 4.4: Days Worked by Vele Independent Women and Married Women, 1981


All Fields 5/15 6/23 7/15 8/1 8/28 10/3 11/3 Total
to to to to to to to
6/22 7/14 7/31 8/27 10/2 11/2 12/31


SORGHUM
Planting
Independent 4.6 7.7 12.3
Married 5.8 8.6 .2 14.6
(.55) (.39) (.22) (.36)
Weeding
Independent 1.8 3.6 2.2 7.6
Married 1.9 4.0 5.1 11.0
(.87) (.70) (.04) (.19)
Attach & Harv.
Independent 1.3 2.1 3.4
Married 2.1 3.8 5.9
(.21) (.04) (.04)
Total Sorghum
Independent 4.6 9.5 3.6 2.2 1.3 2.1 23.4
Married 5.8 10.5 4.2 5.1 2.1 3.8 31.5
(.55) (.47) (.58) (.04) (.21) (.04) (.14)


RAINY SEASON RICE
Transplanting
Independent 2.9 8.4 13.4 .8 25.5
Married 2.4 7.0 8.1 .7 18.2
(.57) (.31) (.01) (.84) (.03)
Weeding
Independent 2.0 15.5 .9 18.4
Married 2.3 11.8 .2 14.3
(.75) -(.08) (.17) (.04)
Harv. & Thresh
Independent 1.8 29.0 30.8
Married 2.0 25.0 27.0
(.83) (.06) (.08)
Total R. S. Rice
Independent 2.9 8.4 15.4 16.3 2.7 29.0 74.7
Married 2.4 7.0 10.4 12.5 2.1 25.0 59.4
(.57) (.32) (.01) (.08) (.40) (.06) (.01)


TOTAL ALL CROPS
Independent 17.1 12.4 12.0 17.8 19.1 7.5 29.1 114.9
Married 16.7 13.0 11.2 15.5 15.9 7.4 25.3 104.8
(.78) (.70) (.44) (.08) (.08) (.92) (.06) (.06)

Note: Figures in parentheses are the probabilities associated with two-
tailed t-tests. See note to Table 3.9.


-75-









land which is readily accessible to her as intensively as is necessary)

to produce enough sorghum to meet her household's basic subsistence

needs, no matter how great the returns to her labor from rice

cultivation might be. Thus, the more children a woman has, the more

time she might be inclined to spend on sorghum production. On the other

hand, since rice is also home-consumed, a woman with more children might

decide to forgo cultivating some of her sorghum land (or cultivate it

less intensively), if rice cultivation is more profitable and/or less

risky than sorghum cultivation to reduce the risk of not having enough

grain to meet household food needs.

The presence of young children might also influence how a woman

chooses to allocate her labor between sorghum and rice production to

some degree. If her sorghum field is closer to the compound, a woman

might prefer to work nearby her children if they are left in the

compound. However, it should be noted that it is not unusual for women

in Vele to have a sorghum field located at some distance from their

compounds. Furthermore, the presence of very young children probably

has a greater influence on the total amount of time she spends on

agricultural activities. Women with very young infants do not usually

cultivate either rice or sorghum. However, these women were

deliberately excluded from the sample in order to focus on the economic

determinants of women's labor allocation to rice production.

Another factor which might influence the amount of time a woman

allocates to sorghum production is the amount of sorghum land to which

she has ready access. The rate at which the returns to sorghum planting

or weeding labor diminish depends not only on the amount of time which


-76-











is spent planting or weeding, but also on the size of the field. The

smaller the field, the more quickly the point of rapidly diminishing

marginal returns is reached. Thus, a woman with a very small sorghum

field might take up rice transplanting sooner than a woman with a larger

field unless, of course, the desire to produce a certain amount of

sorghum causes the former to compensate for the small size of her field

such that she spends even more time cultivating sorghum than the latter.

The above factors might account for some of the variation that

is observed in women's labor allocation to sorghum and rice. However,

they would not account for the difference which is observed between the

groups of married and independent women unless, for example, one group

had more children or greater access to sorghum land. Although on

average the group of independent women had more children present in

their households than the group of married women, 1.6 and 1.4 children

respectively, the difference was not significant (t=.34). Furthermore,

the difference in the amount of sorghum land they cultivated was not

significant: the group of married women cultivated on average .19 ha

and the group of independent women cultivated .18 ha (t=.23).

Nevertheless, as an additional check one can control for these factors

to determine whether the difference in the amount of time they spent on

sorghum weeeding and rice transplanting is significant.

Controlling then for the number of children, the following

regression shows that there is not a significant difference in the

number of days which the two groups spent preparing and sowing their

sorghum fields in the rice transplanting period, 6/23 8/27 (where

GROUP is a dummy variable equal to one for the independent group of


-77-









women, CHILD is the number of children in the household, and the figures

in parentheses are the t-ratios):


(4.2) SORG PLANTING = 9.7 0.7 (CHILD) 0.8 (11.00) (1.51) (.72) F=1.30


As one would expect, the inclusion of a variable for the size of the

sorghum field which was actually planted significantly increases the

explanatory power of the regression:


(4.3) SORG PLANTING = 6.9 0.6 (CHILD) 0.6 (GROUP) R2=.32
(6.04) (1.46) (.66) F=4.79

+ 14.3 (LAND)
(3.31)


The reason that the addition of a variable for sorghum field size does

not explain even more of the variance in planting time is that some of

the field preparation and sowing had to be repeated after the drought.

The insignificant difference in the amount of time the two groups spent

planting is consistent with the similarity in the average size of their

sorghum fields.

However, the difference in the amount of time which the two

groups spent weeding sorghum in the rice transplanting period (6/23 -

8/27), even after field size and the number of children are controlled

for, is more significant:


(4.4) SORG WEEDING = 9.6 2.3 (CHILD) 2.9 (GROUP) + 24.5 (LAND)
(3.35) (2.17) (1.28) (2.27)
R2=.28
F=4.03


As Table 4.4 indicates, married women spent approximately 50% more time


-78-










weeding sorghum than independent women in this period. The difference is

most pronounced in the second half of the rice transplanting period

(8/1-8/27).15 The following regresssions also indicate a similar

pattern:


(4.5) SORG WEEDING = 4.1 0.9 (CHILD) 0.1 (GROUP) + 17.0 (LAND)
(6/23 7/31) (2.47) (1.54) (.05) (2.76)
R= .26
F=3.57


(4.6) SORG WEEDING = 5.6 1.4 (CHILD) 2.9 (GROUP) + 7.5 (LAND)
(8/1 8/27) (3.35) (2.23) (2.16) (1.20)
R= .26
F=3.65


The GROUP dummy variable is not significant in the regression for the

first half of the weeding period, when the amount of time a woman spent

weeding was related to the size of her field and the number of her

children. In the second half of the transplanting period, there is a

significant difference in the amount of time expended by the two groups

on weeding. As Table 4.4 shows, married women spent more than twice as

much time weeding as independent women in August.

Conversely, independent women allocated more labor than married

women to rice transplanting, as the following regression shows:

(4.7) RICE TRANSPLANTING = 12.9 + 2.2 (CHILD) + 9.5 (GROUP) R2= .28
(6/23 8/27) (5.25) (1.64) (3.27) F=6.24


The group of independent women spent 24.7 days transplanting their

household rice fields compared to married women's 16.4 days (t=2.64).

The difference is reflected in the area transplanted. Independent

women's households cultivated almost one piquet per adult worker (.47


-79-










ha) while married women's households transplanted on average slightly

more than one half-piquet (.31 ha) per adult worker.

As with sorghum weeding, however, the difference in the amount of

time the two groups spent transplanting is highly significant only in

August, as the following regressions show:


(4.8) RICE TRANSPLANTING = 8.3 + 0.2 (CHILD) + 3.1 (GROUP) R2 .08
(6/23 7/31) (5.17) (.20) (1.66) F=1.38


(4.9) RICE TRANSPLANTING = 4.6 + 2.0 (CHILD) + 6.4 (GROUP) R2= .33
(8/1 8/27) (2.90) (2.33) (3.38) F=7.74


Independent women spent 13.0 days transplanting their household rice

fields in August, while married women spent only 7.4 days (t=2.95).

The difference in the amount of time they spent transplanting was

reflected in the differences in the amount of time they spent weeding and

harvesting. Independent women spent 17.7 days weeding and 27.7 days

harvesting their household rice fields, compared to married women, who

spent 13.1 days weeding and 21.8 days harvesting. Yelds of independent

and married women's rice fields were not significantly different:

independent women 4270 kg/ha and married women 4336 kg/ha, (t=.16).

Allocative Inefficiency. Cn the basis of the above comparison,

married women's households can be said to be less allocatively efficient

than independent women's households. Both groups of women spent

approximately the same amount of time planting and doing the first

weeding of their sorghum fields. The small amount of sorghum production

that married women would have sacrificed by not doing the second weeding

of their sorghum fields could have easily been made up by retaining a

slightly greater amount of paddy. In the rice transplanting period,


-80-


*












independent women spent 3.3 days less than married women, who spent 9.1
16
days, weeding their sorghum fields. Valued at 550 CFA per day, this

labor is worth 1,815 CFA, or about 33 kg of paddy. To accomplish their

transplanting independent women also spent 3.1 days more on agricultural

labor than married women in the peak transplanting period; thus, to have

taken on more transplanting, married women would have had to forgo some

of their nonagricultural activities. The difference in the total amount

of time spent on agricultural production cannot be attributed to

different childcare demands, since the the difference in the amount of

time they spend on agricultural production is significant even when the

number and age of children are accounted for.17

Married women's underallocation of time to transplanting

(relative to the labor allocation of independent women) cost their

households a significant amount of incomt of income. One method of

calculating the loss in income from women's underallocation.of labor to

rice production is to take the difference between the returns to the

additional labor independent women spent on rice cultivation and the

opportunity cost of that time for married women, assuming that rice

cultivation is not an option. Independent women spent 7.3 days

transplanting, 4.1 days weeding and 4.1 days harvesting more than married

women; in total they spent 15.5 days more cultivating rice (on all

fields) than married women.18 If the labor is valued at the hired wage

rate, then independent women earned about 10,075 CFA more than married

women (or rather, their households). However, married women spent 5.9

days more cultivating sorghum in the rice cultivation period for a return

of 3245 CFA, if each day is valued at 550 CFA.


-81-









In total, then, independent women spent 9.6 days more on

agricultural labor than married women. However, the possibility that

married women were engaging in nonagricultural income-generating

activities during these 9.6 days cannot be ignored. If they did so, they

would have earned 960 CFA, assuming that the returns to their labor were

100 CFA/day. In total, then, independent women earned 10,075 (3,245 +

960) = 5870 CFA more than married women (or, more accurately, their

households). This is not a trivial loss of income--it represents about

6% of household income from rainy season sorghum and rice production, or

about 12% of the returns to women's rainy season agricultural labor.19

What needs to be explained, therefore, is why married women's

households are less allocatively efficient than independent women's

households. One hypothesis is suggested by dividing the married women

into two groups on the basis of whether their households cultivated as

much rice land per active household worker as the group of independent

women's households. As the following regression shows, married women

whose households cultivated as much rice land as the group of widows (and

spent the same amount of time cultivating rice) received a higher rate of

compensation from their husbands than those whose households cultivated
20
half as much rice land on average:2


(4.11) COMPENSATION = -1758 + 262 (DAYS) + 65 (DUMMY) R .74
(t-ratios) (.62) (4.27) (2.11) F=39.91


(where DUMMY is a dummy variable for the number of days worked by women

whose household cultivated as much rice land as the widowed groups). The

difference between their rates of compensation suggests that if married

women received a higher rate of compensation, they would allocate more time


-82-











to rice production. Thus, while the majority of married women are

compensated at a rate greater than the opportunity cost of their labor, it

is apparently not a sufficient inducement for them to take on the

cultivation of an additional rice field.



4.4. A Bargaining Approach to Household Resource Allocation

As became apparent through interviews with men and women and

observation, there is frequent and sometimes pronounced conflict between

men and women over the division of income from rice production. Men have

traditionally had the right to appropriate any income earned by their

wives. On one hand, men have an interest in using their wives' income,

especially to purchase livestock for bridewealth payments. Ch the other,

their interest conflicts at the margin with women's interests in using the

income to purchase consumer goods which have become increasingly more

available and socially necessary since the advent of rice cultivation. The

conflict between husband and wife over the amount of remuneration women

receive for their labor on their husbands' crops is not unique to the SEMRY

area (see Section Five).

Thus, in such situations, it seems most appropriate to model the

intrahousehold conflict over the division of income as the outcome of a

bargaining process. In many areas of Africa women do not have independent

access to the resources necessary for cash crop production. Although they

have independent access to other income-generating resources, the returns

to their labor are often lower than the returns to labor from cash-

cropping. Both husbands and wives can benefit, therefore, if women

cooperate with their husbands on cash crop production as long as the share


-83-











of the profits women receive is greater than the opportunity cost of their

labor and less than the returns to their labor from cash crop production.

In short, under such circumstances the agricultural household can be

conceptualized as a bilateral monopoly in which the indeterminancy problem

associated with the division of the gains is resolved by bargaining.

The traditional model of the household assumes that household

members do not have conflicting interests over the allocation of time and

income (or if they do, that such conflicts are resolved by the imposition

of one household member's preferences). This assumption implies that joint

rationality will always prevail when the household is presented with new
21
economic opportunities.21 A bargaining model, on the other hand, predicts

that resources will be efficiently allocated only when household members

cannot rationally (i.e, according to the mutually observed bargaining rule)

expect other members to make further concessions. A bargaining perspective

suggests that the majority of married rice-cultivating Massa women may be

allocating their labor inefficiently because they are holding out--

"striking" as it were-for a higher rate of compensation.

However, some households have apparently been able to compromise on

a sufficiently high rate of compensation that allocative efficiency

obtains. I can only speculate at this point about some of the factors

that may be responsible for inducing certain households, or women in those

households, to cultivate additional piquets of rice. Men may not wish to

take on the cultivation of an additional piquet because of the

opportunities to earn income from other sources, for example, fishing. In

addition, men who have a number of cattle may prefer to spend their time

managing their cattle affairs (children do the actual herding).22


-84-











Obviously, it is important to understand the factors which influence the

labor allocation pattern of men, but that is an issue beyond the scope of

this study.23

Junior married men in multi-household compounds appear to be

overrepresented in the more extensive group of household cultivators. It

is unlikely that these men have significant cattle herds; most likely any

cattle which they had managed to accumulate would have been used for

bridewealth payments, and they would not likely have inherited -any cattle

given their junior status in the compound. Conflict between junior men and

the compound head over the compound head's obligation to provide cattle for

the bridewealth payment of the junior men (particularly if there are still

cattle outstanding) may in part be responsible for junior men's greater

effort to accumulate cattle. Disputes between the compound head and junior

men over whether the compound head uses his cattle to acquire another wife

or whether he aids the junior man to obtain his first wife are often the

root cause of the most serious of intrahousehold conflicts. A husband of

one of the women in the sample moved out of his father's compound (with his

mother and his younger brother) and set up his own compound alongside his

father's because of a dispute over his father's refusal to complete his

bridewealth payment. If a man still owes cattle to his wife's family, he

would be under considerable pressure to work as hard as possible to

discharge the debt quickly. In such a case, he can ill-afford to dispute

his wife's right to compensation, since he needs the additional income that

he receives from his wife's labor on a second rice field.

However, even if a man is not interested in cultivating a second

piquet, his wife could still take on the cultivation of an additional


-85-










half-piquet of her-own. Very few married women do so, however. Several

examples were encountered of married women, all first wives, who had their

own piquets which they cultivated independently of the rest of their

households. Their husbands cultivated rice with their junior co-wives.

These households were the exceptions to the usual practice of joint

household rice cultivation. In some cases women were permitted to keep all

the income from their field; in other cases their husbands doubled the

amount of compensation they received for their labor relative to what their

co-wives received. Further research is required to determine the factors

that influence whether a woman is permitted to have her own piquet and the

degree of control over the income from that piquet. It may be that a

senior wife whose many years of labor have aided her husband to accumulate

enough cattle to marry two of three other women is permitted to accumulate

wealth in her own right to a greater extent. Other variables which may be

important are: whether she has an adult son in the compound (since she

could leave with him if he decided to establish his own compound); whether

she has a daughter of a marriageable age (in which case her family would

probably not be required to return the bridewealth payment in the event of

divorce); and whether she is inherited. Inherited wives appear to enjoy

somewhat greater economic autonomy. For example, one of the married women

in Widigue sample who controlled the income from her own piquet was

inherited by her husband. Another inherited wife in Vele did not even live

in her husband's compound; instead, she lived with her deceased husband's

older brother in his compound and cultivated her own piquet independently

of him. Other inherited wives cultivate rice with their adult sons rather

than with their husbands.


-86-










Thus, there may be structural factors which essentially place some

women in a better bargaining position. In general, however, a husband may

be reluctant to have his wife take on the cultivation of her piquet because

it might mean that she would be in a stronger position to challenge his

right to her income. In addition, there is the possibility that women

themselves may be reluctant to take on additional piquets,..either with or

independently of their husbands. If a woman is opposed to her husband

taking a second wife, as some (but not all) women are, then she would have

little incentive to help him accumulate cattle--unless he made it worth her

while in the short run.

If households are increasing the number of piquets they cultivate

.as they gain greater familiarity with rice cultivation and knowledge of how

the returns to their labor from rice compare to sorghum, then it is

possible that in the future farmers will allocate more labor to rice

production with little prompting from SEMRY. But if the small number of

households which cultivate additional piquets do so because of structural

factors such as the ones outlined above, then farmer participation in rice

cultivation will probably not increase dramatically, ceterbis paribus.

Since SEMRY's financial viability depends on increasing the area which

farmers transplant in the rainy season by 50%, it behooves SEMRY to

understand why women allocate as much--or as little--labor as they do to

rice production and to consider ways of inducing women to participate to a

greater degree. For example, SEMRY could invest in grain mills and pumps

to reduce the amount of time women spend on domestic labor, but in the

presence of ongoing conflict between husband and wife over the division of

the proceeds, it is unclear whether additional female labor for rice


-87-









production would be forthcoming. If the intrahousehold conflict over

income is a significant factor in depressing the amount of labor allocated

to rice production, then SEMRY's long-term financial position depends on a

quick resolution of the conflict.


-88-










SECTION 5. A COMPARATIVE PERSPECTIVE


Section 4 examined how the introduction of the SEMRY irrigated

rice production project altered women's labor allocation and the

household's food provision pattern. It reached the following

conclusions. Even though women's husbands do not allow them access to

rice land on their own account, women have nevertheless managed to

secure a share of the income from rice production by working for their

husbands. .Furthermore, the compensation they receive is greater than

the income they forgo on account of rice cultivation. This increase in

income is real; rice cultivation does not appear to have increased the

share of household subsistence needs for which women are responsible.

Rice production has offered women increased opportunities for employment

at higher rates of remuneration. It substitutes for income-generating

activities which are less remunerative than the rate at which women are

compensated for their labor on rice production. As was pointed out, the

intrahousehold rate of compensation is below he market wage rate. If

women could cultivate rice in their own right, they would earn more

income. Nonetheless, it is important not to lose sight of the fact that

rice cultivation has improved their material position.

As was pointed out, the rate at which they are compensated does

not necessarily indicate the benefits they receive from rice

cultivation; women may derive benefit from their husbands' expenditures

made as well as from their own. However, the comparison of married and

independent women's labor allocation patterns suggests that there is a

conflict at the margin over the use of income from rice production.


-89-









Most married women allocate their labor inefficiently relative to

independent women. The small percentage of married women who allocated

'hei' labor as efficiently as independent women were compensated at a

higher rate by their husbands. A bargaining approach explains why

women, even those who efficiently allocate their labor, receive a

relatively small portion of the net returns to their labor from rice

production. It also suggests that the majority of married women,

despite the fact that they are compensated at a rate greater than the

opportunity cost of their labor, are allocating their labor

inefficiently because of the conflict with their husbands over the

intrahousehold distribution of rice income. The intrahousehold

conflict, therefore, depresses total rice production, threatening the

financial viability of SEMRY.

This chapter draws on comparative material from elsewhere in

Africa to place these findings in perspective. Its principal concern is

to show how the structural position of women within their households and

the wider economy mediates their access to resources and control over

the disposition of household income. The extent to which women control

critical resources, the allocation of their labor, and the income from

that labor has an important bearing on their ability to protect their

interests, which do not always coincide with their husbands'. In the

context of the bargaining approach outlined in the last chapter, this

chapter addresses three issues: 1) women's control over

income-generating assets; 2) given, in many cases, their relatively weak

control over productive resources, the terms under which their labor is

mobilized by their husbands and 3) changes in the intrahousehold


-90-










allocation of responsibility for household maintenance expenditures.

One could also consider shifts in the assignment of responsibility for

domestic Labor tasks (childcare, cooking, laundry, etc.) but this seems

to be the area least open to negotiation at present.

This approach is taken in deliberate contrast to the

perspective presented by Boserup (1970) in Women's Role in Economic

Development. Boserup argues that the deterioration of the status in

African women is due to the uninformed, culturally biased policies of

agricultural development authorities who extended labor-saving

agricultural technology and "modern commercial agricultural practices"

to men (1970:54). In Boserup's view, the monopolization of labor-saving

technologies by men creates a labor productivity gap between male and

female labor which in turn increases men's status relative to women's:

"the corollary of the relative decline in women's labour productivity is

a decline in their relative status within agriculture, and, as a further

result, women will want either to abandon cultivation and retire to

domestic life, or leave for the town" (1970:53). Thus, to Boserup,

technology is the "key" variable (Quinn, 1978:182) that determines

women's relative status, irrespective of the particular socioeconomic

structure in which it is embedded (Beneria and Sen, 1981).

As Tosh (1981) points out, the "cash crop revolution" in Africa

depended in relatively few instances on the introduction of labor-saving

technology or "modern commercial agricultural practices." Rather, it

was most often a result of a reallocation and intensification of labor

in response to new cropping opportunities or the demands of the colonial

state. Thus, the decline in African women's status (where it has


-91-









declined) cannot be attributed to the monopolization of labor-saving

technologies by men as a result of the gender-biased policies of the

colonial state.

This is not to imply, however, that women have had access to the

resources necessary for cash crop production, be they land, labor

draft-powered technology, improved seeds, fertilizer, etc., on the same

terms as men, or that the policies of the colonial state were gender-

neutral, e.g. with respect to land ownership. However, it must be

recognized that men were often structurally better placed to take

advantage of the policies and actions of the colonial state (or to make

women bear a greater burden of the costs of those policies) even when

the policies were nominally gender-neutral. The critical question is

how men were able to exploit the opportunities afforded by cash crop

production to strengthen their control over critical resources and the

income from those resources to pursue their own strategies of

accumulation (Guyer, 1983). In turn, the greater control that men were

often able to exercise over the resources necessary for cash crop

production created, in many instances, the conditions for the

mobilization of women's labor for cash crop production or the

intensification of their labor in subsistence production.


-92-










5.1. Control over Income-Generating Resources

To deliver inputs to women, it is necessary to understand how

the social organization of production influences their access to

resources. As the analysis of the SEMRY project showed, women are not

discriminated against by project authorities; rather it is the social

organization of production and distribution which denies them

independent access to rice fields and control over the product of their

labor.

Similar phenomena have been documented elsewhere in Africa. The

establishment of cocoa farms in Nigeria and Cameroon was undertaken by

farmers largely on their own initiative with little direct involvement

of colonial authorities. Thus it is difficult to attribute men's

control over cocoa farming, as Guyer says, "entirely to the machinations

of the colonial government" (1980b:364). She observes that in the case

of the Yoruba, cocoa farming was an extension of men's predominant role

in food farming, and that among the Beti, cocoa production was

controlled by men because

its technical characteristics as a crop meant that it was
planted in newly cleared forest fields using men's
agricultural techniques. It also constituted a form of
permanent occupation of the land in societies in which
land ownership rights, as opposed to limited rights of
usufruct, were vested in men. (1980b:364).

In addition, as Berry (1975) and Hill (1963a) have pointed out, in many

areas cocoa farms were established primarily by migrants who not only had

to have sufficient capital to acquire land where necessary and mobilize

laborers to help them clear the forest land but also had to maintain their

households until the cocoa farms matured. Migrant cocoa farmers also

relied on indigenous social and economic institutions to provide them with


-93-




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs