Front Cover
 Title Page

Group Title: Conference on Gender Issues in Farming Systems Research and Extenion, University of Florida, February 26 to March 1, 1986
Title: Gender factor and technology options for Zambia's subsistence farming systems
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00081678/00001
 Material Information
Title: Gender factor and technology options for Zambia's subsistence farming systems
Series Title: Conference on Gender Issues in Farming Systems Research and Extenion, University of Florida, February 26 to March 1, 1986
Physical Description: Book
Language: English
Creator: Sutherland, Alistair J.
Publisher: University of Florida
Publication Date: 1986
Subject: Farming   ( lcsh )
Agriculture   ( lcsh )
Farm life   ( lcsh )
University of Florida.   ( lcsh )
Spatial Coverage: Africa -- Zambia
North America -- United States of America -- Florida
 Record Information
Bibliographic ID: UF00081678
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved, Board of Trustees of the University of Florida

Table of Contents
    Front Cover
        Front cover
    Title Page
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
Full Text

.... ... ...... .- .- .........
W---- iA-- --- --- I-.. -^-- IY ~--L-----

-at the University o floncd- -

Conference on

The Gender Factor

and Technology Options for

Zambia's Subsistence Farming Systems

by Alistair J. Sutherland

Paper presented at conference on "GENDER ISSUES IN FARMING



UNIVERSITY OF FLORIDA: February 26 March 1, 1986.



Recent reviews of the role of women in farming systems have

pointed to the variability of women' contributions across

the developing world (Ferguson and Horn, 1985; Lele, 1985;

Rockefeller and ISNAR, 1985). In the Southern African con-

text, some writers have noted that rural women do not com-

prise a uniform and undifferentiated group, and that economic

differentation between households may be a more significant

variable than gender of household head in accounting for

variation in farming (Brown, 1981; Fortmann, 1984; Keller

1984.; Safilios-Rothchild, 1985). In the case of Zambia, this

recognition represents a progression of interest and

awareness, as the development effort relating to women moves

increasingly from campaigning at a general level to

implementation in specific situations (Keller, 1984;

Safilios-Rothschild, 1984; ZARD, 1985). Effective implement-

ation requires that empirical relations at the field level

are properly understood in advance, and such understanding

implies detailed and thorough socio-economic studies of

women' roles (Keller, op. cit.).

In Zambia, along with a growing recognition of variation in

gender roles, comes an awareness of the need for the fuller

integration of women's interest within the larger conto:t of

rural development. This awareness is all important in view of

the large number of projects recently inventoried (Keller,

n.d.) which are supported by government and a range of donors

and which aim to increase women' participation in rural

development. Agricultural projects which have been targeted

specifically at women's groups have proved relatively in-

effective and have often failed to reach the poorer house-

holds (Chilivumbo and Kanyengwa, n.d. and Keller, 1984 ).

Farming systems research, when institutionalized at the

national level, is an approach with the potential for a

particularly full integration of women's interests into the

agricultural development process (CIMMYT, 1985). The time is

ripe for focused empirical studies giving the data needed to

implement a research and development approach which takes

women' role and contribution to agriculture properly into

account (Ferguson and Horn op. cit.). Moreover, the

treatment of gender-as a social variable needs to be made a

priority for social scientists working in FSR programmes CIP,

1985). It should '-also -be a priority to make systematic

comparisons within a country of women' roles in agriculture.

Such comparison need to be systems specific and done in

conjunction with the identification of recommendation do-

mains, and the subsequent setting of research priorities. If

a comparative and systematic approach is not adopted, the

danger is that a well documented case from a particular

country may be taken to be typical of that country (or larger

region in which that country is included), when in fact it

is not (Sharpe, 1984). Thus FSR provides the framework for

country based comparative studies which can be used Lo pro-

ceed to one of the objectives of this conference; comparison

between countries, sub-continents and continents.

My paper begins a country description for Zambia. I set the

stage with a brief summary of Zambia's farming systems re-

search programme, before proceeding to present brief case

studies which address the issue of gender and its

relationship to recommendation domains and research

priorities in three of Zambia's eight provinces.


Zambia's farming systems research programme was established

around 1980, following a demonstration of farming systems

methods by CIMMYT's East Africa programme in 1978/9 (Kean,

1985). A separate farming systems section known as the

Adaptive Research Planning Team (ARPT) was established within

the Research Branch of the Department of Agriculture,

Ministry of Agriculture and Water Development. Each of

Zambia's eight provinces will have its own ARPT team com-

prised of a farming systems agronomist and economist. To

date six provincial teams have been established. I joined

ARPT as a consultant rural sociologist to advise on the need

for, and nature of a professional sociological input into the

programme (Kean and Sutherland, 1984). Research extension

liaison officers and a consultant nutritionist were also

added to ARPT.

Gender issues received little attention in the initial years

of data collection in Zambia. ARPT team members were trained

through CIMMYT training programmes which, while emphasising a

'user perspective', did not specifically emphasis the

importance of gender issues (CIMMYT, 1985). Greater

attention was brought to bear on the issue after the socio-

logist and nutritionist joined the team. The case studies

presented below have arisen out my involvement as consultant

sociologist in the ARPT programme.


Luapula Province lies in the northern high rainfall belt of

Zambia. The province is characterized by subsistence cropping

throughout, and by fishing (commercial and subsistence) in

areas close to the main rivers swamps and lakes. Cassava is

the principal starch staple supplemented with finger millet

(the traditional staple) where chitemene shifting cultivation

is still practiced, and with maize where cultivation has

become semi-permanent. Both rice and maize are grown on a

limited scale as cash crops, largely in areas where

institutional support and incentives are provided by donor

projects. Annual rainfall averages 950 1200mm and as in

the rest of Zambia is concentrated between the months of

October and April, and there is no frost. The soils are of

two types, upland and 'dambo' (low-lying areas prone to

waterlogging), with nearly all crops apart from rice being

grown on upland soils. Most soils have a sandy/gravelly

texture and poor fertility due to a low Ph caused by exces-

sive leaching. Owing to disease problems and tsetse fly

cattle are very scarce and hardly ever used for draught

purposes. The traditional method of chjitemene using an axe

has given way to the mounding and ridging of fields using a

hand hoe. Population and pressure on land adjacent to

settlement is growing because the population is concentrated

in large villages along the major roads, rivers and lake-

shores of the province. Female headed households (FHHs)

comprise from 27% to 38% by district of all households

according to the 1980 census (Safilios-Rothschild, 1984).

In view of the high proportion of FHHs and the prominent role

women play in subsistence farming, in 1983 a study was

conducted by the Zambian counterpart sociologist in order to

establish the role of women in farming in Luapula, and to see

if FHHs comprised a separate target group for on-farm re-

search (Haalubono, 1984). Two different areas representing

two recommendation- domains, 'traditional recommendation

domains 7 & 4' (TRD7 & TRD4) were chosen for the study;

Mukunta Village loc-ated in a fishing area of TRD7 on the

shores of Lake Mweru, and Mabumba Village a plateau village

of TRD4 which relied on cropping for subsistence. These

villages were the two primary target areas for surveys and

on-farm research in the province. In each village, a sample

of ten FHHs was compared with ten male headed households

(MHHs) in relation to:- division of labour in cropping;

choice of crops; cash earning opportunities; land tenure, and

decision making. Data collected through unstructured inter-

views was supplemented by participant observation.

Certain features were common to both areas. The size of semi-

permanent land holdings is limited by the amount a family

(often with hired assistance) can prepare using a hand

Land tenure is based on customary rights allocated by

local chiefs and secured through clearing of bush




custom of

payment of

her kin (bu


but there




The kinship system

is matrilineal

wi th

residence giving way slowly to virilocal

Along with this changing trend in residence, the

bride service is dying out, being replaced with

brideprice to facilitate removal of a woman from

it not giving the husbands' people custody over the

Each household is an independent production unit

is an obligation to share between relatives and

and on death the larger matrilineal kinship group

the property: spouses and a man's children are

not allowed to inherit.

The division of labour in cultivation is organised and al-

located according to gender. In semi-permanent fields men

are responsible for the preparation of mounds for cassava and

ridges for maize, while women are responsible for planting,

weeding and harvesting. In chitemene fields men (preferably

sons in law) are responsible for lopping branches, while

women are responsible for piling and burning branches, sowing

seed and harvesting. In terms of decision making, women are

the ones who are primarily responsible for food crops, while

men tend to dominate cash cropping decisions. However, as has

been reported elsewhere in Zambia, in most cases spouses

consult each other before making decisions (Due et. al. 1984).

the comparison of MHHs and FHHs revealed




In Mukunta,

surprisingly few differences. The two types of household

enjoyed the same type of access to land and the same choice

of crops (except for one male household which had a small

pineapple and sugarcane plot). The main difference was in

the household size and composition; FHHs were smaller with

less male adults, which made for a difference in access to

male labour in land preparation. This obstacle was overcome

by hiring labour for cash which was easy as cash earning

opportunities for women in the area were plentiful, and men

from Zaire were at hand as casual labourers. In Mukunta women

participated in rice growing as a cash crop; a factor

facilitated by the recruitment policy of the donor project

along with a difficulty in recruiting men for agricultural

credit due to the higher returns from fishing in the area.

While women were excluded from fishing they were the major

fish traders in the area, and both married and unmarried

women had access to this trade. A nutritutional survey in

Mukunta showed seasonal deficiencies in high energy foods

for children and in the calorie intake during the rainy


In Mabumba, there were more differences between MHHs and

FHHS than in Mukunta, but again these were not marked. No

major crops were exclusive to one or other type of household.

The most notable difference in cropping was that each FHH

tended to have a groundnut field, but this was rare in NHHs.

Moreover, growing maize and rice as a cash crop was confined

mainly to MHHS. This may have been due to a combination of

two factors; a different donor being involved in the

administration of credit and lack of fishing opportunities for

men in the area. This notion is supported by the fact that

some FHHs in the sample were active in fish trading as a

source of cash income. A nutritional survey carried out in

Mabumba showed a similar pattern to Mukunta, but with a

bigger problem of high energy foods and calorie intake during

the land preparation season. This was due to the virtual

absense of fish in the diet and the greater distance from

residence to fields, giving mothers less time for preparing

food for their young children.

With regard to recommendation domains, the conclusion drawn

from the findings is that it is not necessary to treat FHHs

in TDR4 or TDR7 as separate target groups. However, research

priorities have been adjusted. Ways and means of introducing

more high energy foods into the cropping system (such as

groundnuts and sunflower) and of ways of reducing the time

and labour in cassava drying, storage and processing to give

women more time for cooking and child-care have been more

seriously considered and technologies addressing some of

these area are being tested. While it not make sense to

treat FHHs as a separate target group, it does to ensure

that at least one third of cooperating farmers in the area

are FHH and that wives are actively involved in trial plan-

ning, site selection and trial assessment.


Lusaka Province occupies the rural area adjacent to the

national capital. It is characterized by a mix of cash

cropping and subsistence farming. The Lusaka ARPT has divided

the province into two traditional recommendation domains;

TRD1, the tsetse fly infested area to the east and south-east

of the capital where subsistence hand-hoe cultivation

prevails, and TRD2, the tsetse fly free areas of communal

lands closer to Lusaka where ox-cultivation and a mix of

subsistence and cash cropping is common. The climate is much

drier than in Luapula Province, with more extremities in

temperature, rainfall, and altitude. The soils are very

variable, ranging from deep clay loams of high fertility to

less fertile shallow gravelly soils on higher ground. Clay

dambos are cultivated in the higher rainfall area, and

alluvial soils are important for cropping in the parts of

TRD1 adjacent to the Luangwa and Zambezi Rivers which are

prone to drought. Maize and sorghum are the main staples,

Settlement is more scattered than in Luapula: villages are

smaller, but more compact groups of households are found.

FHHs are frequent, especially in the more remote district

where they approach 45% of all households. TRD1 is habited

by a mix of ethnic groups. The Sengaand Kunda are matri-

lineal with a tradition of uxurilocal residence. Their Soli

neighbours are also matrilineal, but with virilocal resid-

ence, while the Gova are patrilineal with patrilocal

residence. Household income in TRD1 is supplemented by off-

farm activities such as beer brewing, handicrafts and

gathering of wild produce for sale, and by money sent from

relatives working in town.

In TRD1 a survey was carried out covering 88 households (67

MHHs and 21 FHHs) in which I ensured that gender was included

as a variable. Unfortunately the survey did not include the

full proportion of FHHs (24% in an area where the figure was

about 40%), and the large size of households reported in some

cases suggested that unmarried daughters with children were

included in the households of their parents. While the

results have yet to be fully analysed, some observations are

possible on the basis of data analysis to date.

In terms of access to land there was very little difference

between MHHs and FHHs. In spite of some variation in field

types, the two types of household have similar access to all

kinds of field. The two types of households had roughly the

same proportion of children in school and also offspring in

permanent employment. MHHs were on average about 20% larger

than female households and their reported area cropped to

staples was, on average, larger by a similar proportion. The

methods of planting and land preparation and the main labour

constraints in crop production were almost identical for the

two types of household. The most significant difference was

in the cropping pattern. FHHs were more than twice as likely

as MHHs to grow maize as a sole staple crop and less than

half as likely to grow sorghum as a sole staple. The other

difference was that no FHHs grew cash crops but 10% of MHHs

did so.

In TDR2 the rainfall (800 900mm per annum) is higher and

more regular than TDR1, and the higher altitude makes for

lower temperatures and a greater incidence of frost. Soils

are similar, but with the notable absense of alluvial sands.

Staple crops are the same, but maize is much more important

relative to sorghum, and cash crops cotton, sunflower, and

soyabeans are quite commonly grown. Cash earning

opportunities from off-farm activities are more numerous,

with charcoal burning and casual labour and petty trade being

particularly important. Settlement tends to be scattered with

the fragmentation of family groups into more isolated

independent households, particularly in areas closer to the

city where more cash cropping is found. The domain is

comprised of a mix of Lenje, Soli, Tonga and peoples with a

considerable sprinkling of other incoming groups. Kinship

tends towards matriliny with virilocal residence.

In TRD2 a less extensive rapid survey of 34 households (16

FHHs and 18 MHHs) for the purposes of farmer selection revealed

large differences between the two types of household. Four

out of five FHHS grew maize only as their staple crop

compared with only one out of three MHHS in the sample. MHHS

were more than twice as likely as FHHs to grow a cash crop in

addition to maize. The biggest difference was in access to

draught power. Three in four MHHs were ox owners compared to

one in four FHHs. There were also differences in household

size, and size of land holdings; MHHs had more members and

tended to have larger holdings. Because of the concentration

of o;xen in MHHs it was difficult to separate oxen ownership

from gender as a variable influencing cropping practices.

However MHHs which lacked oxen had similar cropping patterns

to FHHs suggesting that access to draft power was the more

critical of the two variables.

This comparison within Lusaka Province reveals more

differences between recommendation domains than the

comparison in Luapula. In TRD1 the initial findings suggest

that the extent of difference between households does not

justify treating FHHs as a separate target group. However,

research priorities need to be reconsidered in relation to

the differences noted. Research on maize needs to be

targeted more at FHHs. Perhaps more importantly women' need

for more labour saving technologies need to be addressed,

particularly in relation to bird scaring on sorghum and

weeding sorghum and maize (weeding was noted as the major

bottleneck by all types of household). As fewer FHHs grew

sorghum, one reason being that it requires more weeding and

bird scaring than maize, such technologies would make this

useful drought resistant staple more available to FHHs, and

help in reducing risk of crop failure.

In TDR2 big differences between the two household types in

combination with the similarities between FHHs and cattleless

MHHs suggests the need for a subdivision of the domain into

two target groups. Although further research is required to

verify findings, the most obvious suggestion is to sub-divide

the domain into ox owners and non owners; to use access to

draught power rather than gender as the critical variable.

As in TRD2 women's activities need to receive higher priority

in the search for labour saving devices, particularly those

relating to weeding.


Western Province is inhabited by the Lozi'and related peoples.

Like Luapula, Western Province is characterized by sub-

sistence agriculture and by fishing in areas adjacent the

major rivers and swamps. It differs in that livestock,

particularly cattle, are important for most households in the

province. Other differences are much greater variation in

ecology, more ethnic groups and resulting farming systems,

and a greater dependence on imported staple foods. The

latter. feature is due to a combination of a more risk prone

system of cropping susceptible to both drought and flooding

on less fertile soils.

The case reported below arises from research in Senanga West

District in the south west of the province during the

planting season of 1983 (Sutherland, 1984). The research did

not set out specifically to compare female and male headed

households, but included type of household as a variable in a

study of cropping patterns, land tenure, and access to

draught power. As no previous surveys by the provincial ARPT

had included gender of household head as a variable, the

study was the first attempt in the province to look

systematically at gender in relation to recommendation

domains and research priorities. While the .definition of

recommendation domains in the province had included gender of

household head as a variable, responses from extension

workers indicated that only 10% of household heads were

females. It was only after the study was complete that a

much higher proportion (38%) was found in the study area and

examination of the 1980 national census data revealed a much

higher figure across the province (31% to 39% in different


Senanga West is a drought prone area with a subsistence

economy which combines crops with cattle keeping and some

fishing,-hunting and gathering. Its 18,000 square kilometers

are predominantly sandy flood-plain studded with forested

ridges and termite mounds. However, along the banks of the

main rivers of the district, especially the upper Zambezi,

more extensive belts of forest with sandy loam soils are

favoured for settlement and cultivation. A population

density of 3.2 per square kilometer means that grazing is

relatively abundant. Cattle are numerous, and are kept

primarily for local subsistence uses; draught power, milk,

manure, and sale in emergency. They also have a social role

in local institutions such as brideprice, debt payments,

funeral rites and ajistment. In the flood plain areas

particularly, arable land is quite limited. Land tenure is

administered through the office of the Lozi paramount chief

by local indunas and village headmen and rights are secured

by residence and by use. The division of labour in

agriculture makes males responsible for livestock, and ox

ploughing and manuring. Adults of both sexes plant, while

women weed and harvest.

Ecological variation gives rise to three main soil and field

types: mutemwa fields on sandy loams found on forested

ridges; mazulu fields on wooded termite mounds with sandy

clay loam soils; and sitapa.fields on seepage soils adjacent

to rivers. In addition, many hamlets have manured gardens

close by where cattle are penned at night. Settlement is

scattered, people living in hamlets containing between 3 to

15 households: a household being defined as the owner (plus

dependents) of a house and granary with independent rights to

arable land. A hamlet is a discrete cluster of households

related by kinship and usually sharing a cattle pen. A

collection of hamlets of different sizes makes up a locality

or "neighbourhood". This area is inhabited by three main

ethnic groups; Lozi, Shanjo and Mbunda. Different ethnic

groups do not share hamlets but often share the same neigh-

bourhood. Lozi and Mbunda hamlets are situated on the banks

of the main rivers and adjacent upland forest and along

tributaries while Shanjo are confined to the drier flood


Farming in Senanga West is typically subsistence. Cash crop-

ping is almost unknown, and the district has been the

recipient of famine relief due to three years of drought and

an influx of refugees from neighboring Angola and Namibia.

Farmers reduce the risk of crop failure by using a range of

drought resistant staples spread over different field types

and planted at different times. While all households have

independent rights to arable land and its product, very few

are truly independent in production. Rights in land and

assistance in production are secured by virtue of membership

in a hamlet and a neighbourhood. Nearly everyone is active in

farming in spite of unfavourable soils and rainfall, frequent

crop failure, and often pitiful yields. Relations of

production are largely uncommercialised and the substitution

of cash for other scarce factors is very infrequent. Kinship

varies somewhat between groups. Lozi and Shanjo have

cognatic kinship with virilocal residence, while Mbunda have

matrilineal kinship with a mix of uxurilocal and virilocal


The survey of 190 households revealed differences in farming

practices and resource endowments which corresponded with two

social variables in particular, ethnic group and sex of

household head (polygamous households comprised 9% of the

sample and were classified as male headed). There was no

significant difference between the three main ethnic groups

in the proportions of FHHs (38% 40% for all groups).

Lozi and Shanjo had, on average, significantly more fields,

and more types of field, scattered over a wider geographical

area than Mbunda. In Lozi and Shanjo settlements, FHHs had

significantly fewer fields per household and also fewer types

of field. With Lozi, FHH'S relied more on mutemwa fields

while MHH's relied more on mazulu sitapa and manured fields.

Shanjo FHH'S relied more on mazulu fields, while Shanjo MHH'S

relied more on manured fields and mutemwa fields. The dif-

ferences between household types suggest that FHHs are less

likely to spread risks in cropping, and this places them more

in a position of dependence on MHHs, especially in bad


With crop preferences and planting priorities, there were

more differences between the three ethnic groups than between

types of household in relation to preferences and planting

priorities for the three staple cereals; maize, pearl millet

and sorghum. Overall there was little difference between

types of households' and staple crop preferences. However,

with Shanjo and Lozi, FHHs using mazulu and sitapa were

significantly more likely than MHHs to plant pearl millet as

the first crop, while Mbunda and Lozi FHHs using mushitu

fields were more likely to plant maize. There was also some

variation between household types in relation to crop

preferences and cropping patterns. FHHs were proportionately

less likely to have planted early than MHHs. This tendency

was most pronounced in Lozi and Shanjo FHHs depending on

mazulu and sitaEa fields. In fact on mushitu fields FHHs

were proportionately more likely to have planted an early

crop than MHHs. This is because most Mbunda FHHs relying on

mushitu fields only, depend largely on hand hoe cultivation,

and so do not have to wait their turn for oxen.

With the ownership of oxen, ploughs and harness, gender of

household head was very significant. MHHs were six times as

likely as FHHs to own oxen. On average, only one household in

five owned oxen, but with significant differences between

ethnic groups. Ox ownership was confined to one in three

Shanjo, one in five Lozi and one in 17 Mbunda households.

With MHHs, polygamists were more likely to own oxen (47%) and

more likely to own more than two pairs of oxen (29%) than

were monogamists (29% and 6% respectively). This strongly

suggests that in cattle owning hamlets, cattle ownership and

polygamy are indicators of status and influence. The owner-

ship pattern of ploughs and harness was similar to that of

oxen; MHHs households being much more likely to own these.

With co-operation in ploughing, a highly significant finding

was that of the households which ploughed, on average only

one in seven was, in terms of oxen and equipment, independ-

ent (one in five Lozi, one in seven Shanjo and no Mbunda).

Only one in ten households hired for cash; the majority (72%)

co-operated on the basis of kinship and neighbourly relation-

ships. No FHHs were independent in ploughing. In Lozi and

Shanjo hamlets, most FHHs depended on assistance from male

relatives and neighbours, while in Mbunda villages they

mostly hired for cash or relied on hand-hoe cultivation. In

Lozi and Shanjo villages two in seven MHHs were independent

in ploughing; no Mbunda MHHs were independent.

The findings Senanga West raise a question mark regarding the

applicability of the concept of recommendation domains for

Western Province. The high degree of variation over a small

area and population make the notion of recommendation domains

unworkable. Thus while gender of household is a significant

variable in accounting for differences in farming practises

in Senanga West, the importance of ethnic and ecological

variation overlays that of gender, making for a complex

pattern. In place of recommendation domains, it is

appropriate to look for technical options which will benefit

the majority of households in the area, particularly the

resource poor households which includes FHHs. In this

context, there is a need for improved drought resistant

varieties of pearl millet and maize, for more labour saving

weeding technologies, and for ways of generating a cash

income to enable cattleless households access to oxen as

required. In the testing of new technologies cooperating

farmers need to be sub-divided into ox-owners and non-owners,

and about 40% of cooperators should be FHHs from all ethnic



The comparison above provides an initial indication of the

variability of women' roles in Zambia's subsistence farming

systems, as evaluated from the point of view of identifying

recommendation domains and setting research priorities.

Gender roles were significantly influenced by a range of

factors; cultural, economic, political, and ecological.

In order to begin a basis for comparison, these influences

are summarised in brief below:

a. The type of kinship systems influenced the degree of

autonomy of women: matrilineal systems gave women greater

autonomy from spouses and male relatives than did cognatic

systems. But kinship was overlayed by residence at marriage

and by the existence or not of cattle and ox ploughing.

b. Perhaps more important than the type of kinship system was

the type of residence at marriage: uxurilocal residence being

associated with both greater female autonomy and more

submission of younger MHHs to older FHHs than was the case

with virilocal residence, where married women and FHHs were

subject to more male authority.

c. The presence or absense of draught power was also a highly

significant factor. Where draught power was important, then

women generally, and FHHs in particular, were far more

dependent on men and MHHs for successful crop production.

d. The availability of scarcer types of arable land was also

a factor related to gender differences. This was most

pronounced in Western Province where access to draught power

and animal manure imposes limitations on the kinds of land

that can be easily cultivated. Significantly, in parts of

TRD1 in Lusaka Province where alluvial soils are highly

valued but not dependent on draught power for cultivation

available data suggested little difference between household

types in degree of access.

e. The local availability of cash and labour/draught power

also influenced gender differences. Where cash earning

opportunities for women were plentiful and labour/draught

power was easy to hire, FHHs are in a less disadvantaged

position than where these area in short supply.

f. When ecological conditions make cropping a high risk

activity and tillage is by ox ploughing FHHs and wives are

more dependent on males for food supply than where ecological

conditions favour a reliable cropping system which does not

depend on ox ploughing.

These tentative observations relate to farming systems which

are subsistence oriented. As a subsistence farming system

moves towards a cash cropping system, as is beginning to

happen in parts of Luapula and Lusaka Provinces, other

factors more closely connected with the larger political

economy influence gender differences (Chilivumbo, 1984). The

reliance on credit and extension advice for cash cropping

tends to favour males and the better off in spite of donor

efforts to reach the rural poor (Due et. al. 1984, Mungate,

1984). Moreover, the fragmentation of settlement and weaken-

ing of kinship and neighbourly ties which accompanies the

growth of cash cropping tends to place greater labour burdens

on women in the households of cash croppers and makes it more

difficult for FHHs to secure assistance from kin and

neighbours in critical farming operations.

At a more immediately applicable level, some tentative recom-

mendations regarding the treatment of gender issues in the

identification of recommendation domains and setting of re-

search priorities for FSR in Zambia conclude this paper.

1. The gender issue needs to be broadened beyond the

consideration of whether or not to treat female headed

households as a separate target group. From the data

available in Zambia and elsewhere, FHHs are often an

internally differentiated group, while married women often

face similar constraints to single women in the management of

their farms and households (Fortmann, 1984). This suggests

that when setting research priorities, it should be women's

interests as a group, whether married or not which should be


2. When oxen are introduced into farming at the subsistence

level, and cash earning opportunities are restricted mainly

to males, this tends to create bigger wealth differences

between MHHs and FHHs and thus relegate most FHHs to a level

of impoverishment and dependence on MHHs; a 'feminisation of

poverty' occurs (Brown, 1981:17). However, FHHs are not the

only households affected by this impoverishment and it

probably makes more sense to identify recommendation domains

on the basis of a household's access to draught power than on

the basis of the gender of its head.

3. When setting research priorities and conducting trials on

farmers fields in Zambia there is a tendency not to consider

the gender aspect. Instead, returns to cash invested or

labour input (measured as a genderless input) is the main

yardstick of evaluation. There is, moreover, a real shortage

of technical scientific expertise to develop and test

appropriate technology to reduce women' burden in household

work not directly connected with crop production (Keller,

1984). This is a problem which cannot be tackled only by

improving the gender sensitivity in setting research

priorities and testing new technologies. There is a need to

lobby for national commitment to developing and testing

appropriate household technology as one of the activities of

the Research Branch of the Ministry of Agriculture.

4. While there is a place for thorough studies of gender

roles, it is quicker and cheaper to ensure that gender is

included as a variable during survey work. Yet the results of

such surveys often do not tell the whole story, and too much

time can be spent on detailed studies of gender roles in

agriculture with little visible benefit to research. It is

necessary to continue to explore the importance of gender in

a particular system by ensuring that women are actively

involved in problem diagnosis, pre-screening, testing, and

evaluation of new technologies. To effect this it is

necessary to sensitise farming system practitioners to the

importance of involving women farmers at all stages of the

farming systems research and extension process, and to set

out clear guidelines as to how this can be achieved.

5. In establishing the significance of gender through

application of the diagnostic sequence put forward by CIMMYT

(Collinson, 1981) great care is required during the

delineation of recommendation domains. The experience from

Zambia is that extension workers, local leaders (and often

local social scientists also) underestimate the proportion of

female headed households. Data from reliable sources such as

census data should be used to supplement information, and a

sociologist may be needed to carry out rapid estimates in the

field. When carrying out surveys social scientists on the

FSR team must ensure that a representative proportion of FHHs

are included in the interview sample. Moreover, steps must

be taken to ensure that women are involved in the interview

if necessary by identifying specific questions to be answered

by women only. The opinion of women farmers should be sought

when setting research priorities and designing on-farm

trials. Moreover, a representative proportion of farmer

cooperators should be FHHs.

The above suggestions should serve to make FSR in Zambia more

sensitive to the importance of gender and more effective in

incorporating a gender perspective into the research

programme. In order to facilitate this, rural sociologists

working with ARPT have a gender component written into their

terms of reference. A sample of these is contained in

Appendix A.



Brown, Barbarra
1981 The Imeact of Male Labor MigEration on Women in
Botswana Paper presented at the Annual
Meeting of the African Studies Association,
Bloomington, Indiana

Chilivumbo, Alufeyo
1984 "Small-Scale Farmer Resource Endowment;
Evaluation, Research and Development Issues"
Paper presented at ARPT/CIMMYT Workshop on
'Role of Rural Sociology (and Anthropology)
in Technology Generation and Adoption',
27 29 November, Ridgeway Hotel, Lusaka.

Chilivumbo, Alufeyo and Kanyengwa, Joyce





Women's Participation in Rural Developgment
Programmes: the case of SIDA-LIMA Programme.
Mimeograph: Rural Development Studies Bureau,
University of Zambia.

"CIMMYT'S Experience with the User's
Perspective in Technology Development" in
Rockefeller Foundation and ISNAR (eds)
Women and Agricultural Technologqy Relevance
for Research Vol. II Experiences in Inter-
national and National Research. The Hague,
Netherlands pp. 13 26

"Women and Potatoes in Developing Country Food
Systems: The CIP Experience" in Rockefeller
Foundation and ISNAR (eds) op. cit. pp. 27 34

Collinson, Mike
1981 "A Low Cost Approach to Understanding Small
Farmers" Agricultural Administration 8:
433 450.

Due, Jean and Mudenda, Timothy with Patricia Miller and
Marcia White
1984 Women's Contribution Made Visible: of Farm and
Market Women to Farming Systems and Household
Incomes in Zambi a 1982. Illinois Agricultural
Economics Staff Paper No. 84 E-285: University
of Illinois at Urbana-Champaign.

Ferguson, Anne E. and Horn Nancy
1985 "Situating Agricultural Research in a Class
and Gender Context: the Bean/Cowpea
Collaborative Research Support Programme."
Culture and Agriculture_ 26, Spring 1985,
Special Issue: Women in Farming Systems:
Cameroon, Botswana, Zambia, Tanzania, Taiwan

Fortmann, Louise
1984 "Economic Status and Women's Participation in
Agriculture: A Botswana Case Study" Rural
Socigolgy 49(3):452-464

Haalubono, Albertina
1984 Causes and Characteristics of Female Farming
in Mansa and Nchelenqe Districts. Luapula
Province (Draft) Adaptive Research Planning
Team Special Study Report, Ministry of
Agriculture and Water Development, Lusaka

Kean, Stuart

Institutionalising Zambia's Farming SystemsL
Research Programme: the Case of the AdaptLive
Research Flanning Team Mimeograph, Research
Branch, Ministry of Agriculture and Water
Development, Lusaka.

Kean, Stuart and Sutherland, Alistair
1984 "Institutionalising Rural Sociology into
Agricultural Research The Zambian Case
Study" Paper presented at ARPT/CIMMYT Work-
shop bn 'Role of Rural Sociology (Anthropol-
ogy) in Technology Generation and Adoption'
27 29 Nov. Ridgeway Hotel, Lusaka.

Keller, Bonnie

Keller, Bonnie

Report on Current Efforts to Integrate Zambian
Women in Develoqment Mimeograph, University
of Zambia

An Inventory of Current Efforts to Inteqrate
Zambian Women in Develyoment Mimeograph, Dept.
of African Development Studies, University of
Zambia, Lusaka

Lele, Ulla

"Women and
cit. pp. 243

Structural Transformation"
Foundation and ISNAR (eds)
- 265

Mungate, Dennis

"WOMEN, The silent farm managers in the small
scale commercial areas of Zimbabwe" Zimbabwe
Agriculture Journal 80(6):245 249

Safilios-Rothschild, Constantina
1984 The Policy Im pliations of the Roles of Women
in Agricultural Research in Zambia. (preliminary
draft) Mimeograph, The Population Council and
National Council for Development Planning, Lusaka

Safilios-Rothschild, Constantina
1985 "The Persistence of Women's Invisibility in
Agriculture; Theoretical and Policy Lessons
from Lesothlo and Sierra Leone" Economic
Development and Cultural Cha~ge 33(2):
299 317


Sharpe, Barrie
1985 "Social Knowledge and Farming Systems
Research: Ethnicity, Power and the Invisible
Farmers of North-Central Nigeria" African
Social Research 38, Sutherland and Chileya
(eds) Special Issue on 'African Farming
Systems: Contributions of Anthropology and

Sutherland, Alistair
1984 Draft Power and other socio-economic aspects
of farming+ systems in Senanga West District
Western Province: A Preliminary ReQort
Mimeograph, Adaptive Research Planning Team,
Ministry of Agriculture and Water
Development, Lusaka.

The Rockefeller Foundation and International Service for
National Agricultural Research (ISNAR) (eds)
1985 Women and Agricultural Technolqogy Relevance
for Agricultural Research. Volume 1 Analyses
and Conclusions, The Hague, Netherlands

Zambia Association for Research and Development (ZARD)
1985 Womens Rights in Zambia Proceedings of the
"Second National Women's Rights Conference,
held at Mindolo Ecumenical Foundation, Kitwe,
22 24 March 1984.

Appendix A


ARPT Rural Sociologist for Region 1 (Northern and Luaeula


The rural sociologist for Region 1 will work within the
Adaptive Research Planning Team (ARPT) of the Research
Branch of the Department of Agriculture, MAWD. ARPT brings
together social and biological scientists to provide a socio-
economic and "farming systems" input into the national
agricultural research programme, with particular emphasis on
helping small farmers. The provincially based ARPT teams are
comprised of farming systems agronomists and economists who
carry out survey work and conduct on-farm trials to identify
problems and test technical solutions among targeted groups
of farmers in each province. The methodology used is based on
the CIMMYT .approach, but with the recognition that. this
approach needs to be broadened and modified to suit local

conditions in Zambia's nine provinces.

The rural sociologist has a key role to play in the
broadening and modification process, both by contributing to
the effectiveness of existing activities and by carrying out
specific activities to meet local needs not covered in the
CIMMYT sequence. Of particular importance in Zone 1 is the
fuller inclusion of women and gender differences into the
adaptive research process. In both Northern and Luapula
Provinces women play the major role in food production and
processing, and female headed households comprise a
significant minority of the.rural population (30-40%). The
involvement of women is not an isolated activity, but one
which can be effectively tackled by taking into account the
position of women within the local community and kinship
organisation. These three aspects women, community, and
kinship have been largely neglected in previous adaptive
research. But they are crucial when dealing with the
improvement of subsistence farming systems with technologies
which relate to storage and processing as well as production,
and which involve cash investment and/or cooperation between
households. Improved storage structures and methods, new food
crops, agroforestry, and the introduction of animal draft are
all technologies which require the consideration of local
gender and community arrangements and structures before they
can be successfully tested, evaluated and disseminated
through the local extension service in a way that benefits
poorer (including female headed) households.


1. To assist the ARPT in Luapula and Northern provinces with
zoning/target grouping. In zoning attention must be paid to
including female headed households within a community
perspective and ensuring maximum use is made of local
knowledge relating to agriculture and socio-economic factors.
Avoiding extension biases, particularly the neglect of female
headed households and married women, is another area
requiring sociological expertise.

2. To assist provincial teams with diagnostic survey work.
During informal surveys reduction of extension bias and a
full inclusion of women (both as wives and as household
heads) require particular attention. Attention should also
be given to ensuring the selected "target area" is
representative from a socio-economic point of view. During
formal surveys, sampling and framing of questionnaires
requires special attention to gender issues and the inclusion
of an indigenous knowledge perspective in surveys.
Assistance with training enumerators and analysing and
interpreting results will also be a duty.

3. Full participation in technology screening, with

particular emphasis on representing local household and
women' interests (eg. food security, labour implications,
food processing and storage) as distinct from technology
explicitly geared to cash croppping, and anticipating likely
social consequences of any technology under serious

4. Provide assistance in the design of on-farm experiments
and the selection of farmer co-ooperators to ensure the
target group farmers (especially women) are represented and
participate as fully as possible in the on-farm research

5. To sentisise other ARPT staff, and related staff, to the
importance of including and considering the interests of
women in all stages of the adaptive research and extension
process, and to the relevance of intra-household and
community structures to technology design and adoption.

6. In consultation with provincial teams and with the
national rural sociologist and national co-ordinator, design
and execute special studies to investigate social and
cultural issues, such as the appropriateness of technology
for women, having first established that these issues cannot
be adequately covered within the current methodology.

7. Provide assistance as required to the provincial RELOs
in:- the framing of technical recommendations to ensure
correspondence with local units of measurement; training
field extension staff in farmer classification; and
developing methods of advice delivery which ensure that that
female and resource poor farmers are reached by new messages
arising from ARPT on-farm research.

8. To serve as an informal liaison person between Northern
Province ARPT and Luapula ARPT, thus facilitating better
communication and co-operation in research effort.


Minimum of a Masters Degree in Rural Sociology or
Anthropology and at least three years experience in
agricultural development, preferably in Africa. Sensitivity
to the important role of women in agriculture is essential,
and a working knowledge of farming systems research is

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