EXAMPLES OF FARMING SYSTEMS RESEARCH AT ICARDA

By

David F. Nygaard and Andr~e Rassam

Paper Presented at a Conference on

Intra-Household Processes and Farming Systems Analysis

Be]lagio, Italy, Mach 5-9, 1984

David Nygaard is currently Vice President of the Agricaltural

Development Counril in New York and was Program Leader for the

Farming Systems Progam at ICARDA, Aleppo, Syria. Andre

Rassam is a researcher at ICARDA and a graduate student at the

University ot Vestern Ontario, Lonoon, Ontario, Canada.

I. INTRODUCTION

The concept of Farming Systems Research is seen as a new and different

approach to organizing agricultura! research in developing countries. As

such, it has received significant attention by researchers in agricultural

development activities and substantial support by organizations financing

agricultural development projects. That is surprising since there are still

too few examples of successful Farming Systems Research (FSR) projects to

merit that kind of attention and support; the time frame in which agricul

tural research is conducted is just too long and the corcept of FSR too

recent for it to be otherwise. Thus, one may argue: that it only demon

strates a lack of viable alternatives for solving the problems we face in

developing rural areas rather than attests to the viability of FSR itself.

FSR as it is used in the current literature covers a wide and extremely

diverse set of activities. This was one problem that we confronted at the

International Center For Agricultural Research in Dry Areas (ICARDA) in de

veloping the Farming Systems Program (FSP) until we stopped searching for a

universally applicable definition of FSR and focused instead on a research

program that was important for the region ICARDA serves and appropriate and

useful to the center. Realistic research cbJectives and a clear program

statement of purpose increased the stature of FSK within the center and gave

us the necessary support to carry out our research. Today, three years

later, while it is premature to say that the approach has been successful,

tLhere are results from FSP research projects that indicate we are making

progress in improving the farming systems we are studying. More importantly,

they are results that would not have been achieved had we not adopted a

Farming Systems Research framework.

 -2-

The major objective of this paper is to illustrate the value of FSR

based on the ICARDA experience %;th a cautionary note to keep FSR in per

spective, a perspective that is more narrowly defined than it would be by

many other FSR advocates. We hope to achieve this objective by attempting

to:

1) describe FSR as it currently functions at ICARDA and stress

its unusual features;

2) give examples of successful results to date of the research

program, and

3) outline elements of cuncern about the future of FSR-as a

development strategy.

In the following pages, we argue that FSR deserves the serious atten

tion it is receiving by development specialists, note that it requires

creative researchers and sustained efforts to achieve its full

potential,

but suggest it is not a panacea to development problems. In the next sec

tion, the major elements of FSR at ICARDA are presented. This description

is important in order to understand two examples of FSR which are given in

sections three and four. The final

section describes the limits of FSR and

our concerns that these

limits are seldom recognized by FSR practitioners.

 -3-

II. THE FARMING SYSTEMS PROGRAM AT ICARDA

ICARDA's research was set up in 1977 to have a farming systems orien

tat;on. rhe major ai~n was to develop appropriate technologies that could

be easily integrated into the existing farming systems in the region; thus

increase the quality and quantity of food, and in turn, improve the well

being of the population, particularly the small-holder and resource-poor

farmers. The Farming Systems Program was expected to help set research

priorities for the Center and assure that research findings of the com

modity programs were suitable for and acceptable to these farmers.

At first, the program 4as hesitant to accept this responsibility; we

began with no theoretical basis on which to build and wi-h few staff

trained in Farming Systems Kesearch. Initially, our research projects

were urganized along di-ciplinary lines, arid, while there was some excel

lent research being conducted within the program, there was little inter

disciplinary communication and consequently little focus on the important

interrelationships of the many components of the farming systems we were

studying.

In reorganizing the program, we benefited greatly from the experiences

of other FSR practioners. In particular, Norman (1980) gave us a structure

in which to visualize our research, and CIMMYT (197 ) and IRRI (1978,1981)

provided some methods and tools with which to proceed. The following

definition of FSR and statement of the goal and objectives of the FSP

guided the selection of research topics and the allocation of research

effort.

Definition

Farming Systems Research as it is perceived in the program is a process

that identifies problems limiting agricultural productivity and then searches

 for Solutions
to these problems.

FSR is ccrrprehensive
in that an effort
is

made to evaluate new technologies

in the light of all the components

of the

system including the complex interdependencies

of
these components. This

process recocnizes
the resources
and constraints of the
farming families

(who are
both producers and consumers)

and seeks solutions that

are rele
vant,
useful and acceptable
to these
families. Research
is undertaken by

multidisciplinary
teams of scientists
that
interact continually with the

farmers for whom the

research
is
intended.
This approach should ensure

that the
research produces appropriate
technologies which, therefore, will

be more easily and quickly adopted.

The FSP perceives its research as

a process that
passes
through four

stages (a' a Norman).

These stages are:
(1)
diagnostic,
(2) design or

experimental,
(3) testing,
and (4)
extension. FSR
is problem oriented;
a

clear diagnosis and definition of

the problem are of
paramount importance

if the effort to
find appropriate solutions
is
to succeed. This
will

determine the make-up of

the
team
and the allocation of research effort

to

the various stages.

Inceed, problem-oriented
research acts
to keep the

team together, the effort

focused
and on-schedule.

This process
is dynamic and iterative
since we
frequently return
to

previous stages
to clarify points
as
we gain knowledg-, confront problems

and consider
research alternatives.

In addition, the
distribution be.ween

stages is not sharply defined as

there is
much overlap, and we work at

several staces
simultaneously.
Finally,

it is flexible and adaptable to

many circumstances and different

problems.
By visualizing the
research

process this way,

we keep our work

in perspective vis-a-vis other scientists

and the farmer as well.

Program Goal
and Long Range Objectives

and

The program seeks to find strategies that will add stability

technical and

improve the farming systems

in the region by increasing the

Particular emphasis is placed

economic efficiency of limited resources.

in crop and live

on
soil and water resources, combined with improvements

stock husbandry. To achieve

this goal, the program has two overriding

agricul
objectives which allows the design of
strategies for increasing

tural production.

The first objective is to develop and

refine methods and tools that

An agricultural system is determined by its

are required to conduct FSR.

natural resources, human resources,

historical development and the current

social and economic environment. Because the ICARDA region

is large and

diverse, these factors

are found in many combinations, and consequently,

the systems
in our region are numerous. Therefore, w? do not aim to de

velop a new system or technology that has wide applicability or adapability,

and

but rather a process

that can be used to improve a particular system

than be repeated elsewhere.

The second objective is to

promote the use of FSR as an efficient

approach to solving agricultural problems. FSR is a relatively new

approach to agricultural research, and

few people in the region are fami

Jiar with it. Thus, exposing scientists to this approach, and training

them to use it, is a high program priority and will take a long time to

achieve satisfactorily.

Within this FSR framework, the FSP developed five research projects

for 1981/82 cropping year, all centered in Aleppo, Syria.

In 1983 we

added a sixth project which was our first major outreach activity--an

FSR project in Tunisia. The current projects

are:

 -6-

Project I

The Productivity of Cere l Crops under Rainfed Mediterranean

Environmental Conditions.

Proiect II

Nitrogen F;xation, Productivity and Water Use of Grain and

Forage Legumcs under Rainfed Mediterranean Enviror.mental

Conditions.

Project III

Crop Rotations and Cropping Systems.

Project IV

The Role of Animals in the Farming Systems of the ICARDA Region.

Project V

Environmental Zoning.

Project VI

Farming Systems Research in Tunisia: A test for FSR methodolcgy.

Few FSR practitioners would have difficulty with the preceding descrip

tion of FSP, the program goals and objectives or the six research projects.

Before discussing examples of

some results of our program however, it may

be useful to mention characteristics of the program that have made it more

successful;then it would have otherwise been; they deal

with disciplinary

balance, research emphasis and project accountability, and these features

are somew.hat unusual in comparison with othe- FSR projects.

The first one is the strong contribution :f biologicai scientists to

our research. The ten senior positions in the program

include two agronomists,

two soil scientists, a crop physiologist, a microbiologist, a livestock

scientist and three social

scientists (currently two agricultural economists

and one economist). Well over 80 percent of our research budget

is allocated

to support the biological scientists anJ their research.

We suggest that this

Is in stark contrast to many other FSR activities where social scientists

control and often dominate. Indeed, the fact that a very high percentage

of the people attending this workshop are social scientists illustrates

this point well. At ICARDA, the biological research is as central to the

research of the FSP as is the contribution of the social scientist.

Secondly, FSR projects at ICARDA include activities in all four of the

stages described earlier. For example, diagnostic work is only conducted

within a project where experimental research and on-farm trials are also

current activities. This is important because it assures a distribution of

research resources across stages. In the early years of the FSP, a major

portion of research funds are being spent at the experimental stage. This

can be contrasted with other FSR programs where very little emphasis is

given tc experimentation or contrasted with 'so-called'' FSR projects which

only include diagnostic research. We do not include the latter independent

type research in this discussion of FSR.

An emphasis on all stages of farming syste s research in our projects

is important for two other reasons. It keeps the research pointed toward

the ultimate objective (of increased productivity and improved welfare for

the target farmers). Otherwise researchers too often get lost in the inter

mediate staqes and lose sight of the longer range goals.

The ornanization of the FSP at ICARDA has shown the interrelationship

between the stages and the iterative nature of reLearch efforts. An example

of this is given in the next section, and we only mention here that therv is

a need to be working on several stages simultaneously in order to have the

flexibility to "get it right."

 -8-

Finally, FSR tends to be so inclusive that it is difficult to know

what
to expect from such projects and it is even more difficult to measure

research productivity. The current project orientation of

the FSP at ICARDA

has substantiall, alleviated these problems. A clear statement

of the objec

tives of each project and a stepwise progression toward reaching those objec

tives by which short term results can be evaluated were essential in achiev

ina this.

Two examples from the ICARDA experience should help to clarify these

ideas. The first describes a barley research project and stresses the impor

tance of trials on farmers' fields and differences among farms. The second

looks at the impact of a new technology on labor use within a family as

well

as within a village.

11. TRIALS ON FARMERS' FIELDS

For several years the Farming Systems Program has been engaged in a major

research effort on barley production in climatic environments typically found

in the Mediterranean basin. The research has focused on improved agronomic

management with particular emphasis given to moisture-use by the crop. This

experimental small-plot work has been conducted at five sites selected along

a steep rainfall gradient in Aleppo Province where average rainfall varies

from more than 500 mm/year to less than 250 mm/year.

The choice of agronomic practices to include in the research were par

tially based on surveys of the area where important differences were found

in acricultural practices as well as socio-economic conditions from one

village to another. For example, as one moves from wet to dry villages,

barley production increases in economic importance and farm size also in

creases. However, families in these drier areas are poorer and many require

off-farm income to maintain their livelihoods. Mechanization is common in

the dry areas for barley production but few chemical inputs are used and no

new cultivars are available. Finally, barley production has a very important

dual purpose role -- as a fodder for grazing animals or harvested for grain

and straw. The weather conditions for the particular year in question deter

mine the value and use of the crop. Finally, as we traverse the transect to

ward the drier villages, animal numbers per family increase and the percentage

of family income derived from sheep production increases. It is thought that

barley and sheep compliment each other in a way so as to add inter-seasonal

stability to the family income stream. This information was useful in helping

the agronomist develop priorities for experimental work.

 The experimentation focused on

nitrogen
and phosphorus fertilizer use

and seeding rate

of two cultivars (one old and one new) of
barley. The experi
mental methods and
results of these agronomic
trials discussed in greater de
tail elsewhere (Nygaard, 1982;
Farming Systems rrogram,
various annual
reports).

Important contributions were made by

the
soil physicist (moisture use),
the

soil chemist (soil chemical

analysis for available phosphorus
and nitrogen

status and uptake) and crop physiologist

(plant growth analysis) as well

as

the agronomist. One of

the findings that became clear at
the
initial stages

of this research was

the dramatic differences
between yields
on the experi
mental plots
(even under middle levels
of input use) and yields on nearby

farmers' fields.
Table I indicates
these dramatic differences

in the 1980/81

crop year. Note that the gap

in percentage terms
is more important
in the

drier areas.
This yield gap was maintained
in subsequent seasons,
and economic

analyses showed that

it would be economically
profitable
to close much of the

gap.

After three years of experimental wcrk,

we attempted to
replicate some

of these results on farmers' fields.

For
the reasons given above and due
to

ICARDA's special interest in the poorer

farmers in the region, the

research

concentrated on
barley production in the drier
areas. To keep the
trials

relatively simple and easy for

farmers
to
follow and understand, we decided

to have four treatments:

Plot 1. Local barley (Arabi Aswad) with other

inputs
controlled

by the farmer according

to his normal
practices.

Plot 2. Local barley plus 50 kg. of P 0 with

2 5
farmers' practices.

Plot 3.
A newer cultivar (Beecher) with
farmers' practices.

Plot 4. A newer cultivar plus 50 k/ns of P 0

2 5
with farmers'

practices.

Each plot was one-tenth of a hectare

in
size.

 -II-

TABLE I

Barley Yields 1980/81

Rainfall1 Potential

Yields on Yields on
Increase

Zone Farmers' Fields

FSP Trials (in percent)

1 3,150 4,100 30
2 1,211 3,610
198

3 1,106 2,400 117

4 674 1.710 254

I. Agricultural areas in Syria can be divided into four acri

cultural stability zones
which are most strongly influenced

by average rainfall. They are:

Zone I Over 350 mm of rain per year.

Zone 2 250-350 mm and not less than 250 in two or

three related years.

Zcne 3 Over 250 and not less than this in half

the years.

Zone 4 200-250 mm and not less than 200 in half

the years.

 These trials were the first joint (scientist-farmer) managed on-farm

trials the program conducted as part of its FSR strategy. The first
year of

this
research, 1981/82 was technically disappointing. However, the experi
ence was still educational and beneficial
for the multidisciplinary
team.

It is useful to describe that experience.

In Tabie II the results of the

first year's trials are presented. It

is clear from these data that there is such variability across sites
that

differences
in pair-wise comparisons are not statistically significant.

While the phosphorus effect is at least positive,

it is small in comparison

to the response we found in yields on the experimental plots. In addition,

the newer cultivar appears

to be actually less productive than the traditional

variety.

The team visited the farmers that were cooperating in these trials and

their families several times during

the growing season. Discussions during

these visits touched on several subjects, e.g., traditional technology for

barley production, the value and use

of the grain and by products,
complimen

tary animal activities and other agricultural and nonagricultural

tasks in

which household members are involved.

A number of problems or potential problems that would effect

the adoption

of new agronomic practices were revealed during these visits. An essential

one was the importance of

the cropping history of the plot in determining

barley yields. All of the small-plot experimental work was conducted on fields

which were fallowed in the preceding year. However, six of the eight' fields

in the on-farm trials were planted to barley in the previous year; we did not

pay enough attention to cropping history.

Note that two trials

were grazed by animals and not harvested; thus,
results

of only six fields are reported in Table II.

 TABLE II.

Grain and Straw Yields (kg/ha) of Barley On-Farm Trials, 1981/82

Arabi Aswad-P Arabi Aswad+P Beecher-P Beecher+P

Trial

Zone No. Grain Straw Grain Straw Grain Straw Grain Straw

3 1 7714 1191 61;7 715 508 671 971 975

3 516 594 431 476 584 605 536 532

4 796 1082 910 1236 595 967 799 1133

---------------------------------------------------------------------------------

Mean 696 956 663 809 562 748 769 880

5 46 222 217 298 159 443 163 309

6 170 295 144 215 70 215 120 335

7 831 970 973 1148 668 1269 771 1170

---------------------------------------------------------------------------

Mean 3h9 484 445 5514 299 642 352 605

 It would be difficult to argue that

we
would have missed the ;mpor
tance of croppinc history had this
research
not
been done with a FSR per
spective. and %-'e

do not do so. However,

a FSR approach increased our

under
standing of
cropping rotations and
farmers'
attitudes about
them. and

increased the likelihood of catching

the cropping history
issue. In the

second year, 1582/83, we chose clots that were

faliow in 1981/82. The result

was
a dramatic improvement in yields
in the second year
and greater differ
ences among treatments. Table III
allow us to compare the results of
the

two years. The two villages are the same in each

year
but some of the farmers

changed. Averages
are
given for grain yield only
since the straw yields

sho.' a similar trend.

These results in the

second year clarified several

points.
First, there

was a significant response

to phosphorus

for both the traditional as well

as

the newer cultivar; this was

true in both agricultural
zones. Second, Beecher

barley is not ti.e improved variety we thought

it was, and
subsequently the

barley breeder has developed other

new
lines that are superior to Beecher.

In addition, partially
as a result of
these on-farm trials,
the selection
pro
cess for
new barley cultivars is increasingly
being done
in harsher drier environ
ments in addition to
continued selection
at
ICARDA's research station which
is

located in a relatively higher rainfall

area.

The importance of
the barley/fallow rotation

lead us to search for

more information about

this and other
rotational issues.
Somel (1984) con
ducted a larger survey, of barley producers
and found that
only 50 pcrcent

of the farmers in Zone 3 practiced this rotation while 63 percent

of farmers

in Zone
4 did so (Table IV). As a result,
now much more emphasis is
being

given to other rotations

in the FSP
and some attention is being paid to
a

 TABLE III.

Grain Yields (kg/ha) of Bari y Plots

at Breda and Khanasser for Two Years

1981/821 1982/832

Zone 3

Arabi Aswad no P 696 786

Arabi Aswad + P 663 1026

Beecher no P 562 708

Beecher + P 769 1287

Zone 4

Arabi Aswad no P 349 933

Arabi Aswad + P 445 1474

Beecher no P 299 785

Beecher + P 352 1007

1. Averages based on 3 farmers in each village.

2. Averages based on 4 farmers in each village.

 TABLE IV.

Usual Crop Rotations on the Largest Barley Plot

(Percentage of Farms)

Zone 2 Zone 3 Zone 4

West NE West NE West NE

Barley-Fallow 67 33 50 19 63 8

Barley-Barley 13 29 29 41 20 83

Barley-Barley-Fallow -- 19 -- 33 -- 8

Barley-Legume-Fallow 10 --

Other 10 19 21 7 17 --

Source: Somel, Kutlu, "Barley Production in Syria," Discussion

Paper, ICARDA (forthcoming). Results of survey of 150
barley producers ir,Northern Syria.
 -17

continuous barley rotation which poorer, small

holder farmers claim they are

economically obliged to follow.

Other points were raised by the farmers and, as a result, the team de

veloped several follow-up studies:

1) Farmers thought that Beecher barley, grain and straw, was

a poorer

quality of feed than

their traditional variety. Therefore digestibility

studies were beaun at the experiment station by the livestock unit and,
in

fact, the animal component became a much more important research element of

the project. (Thomson,1983.)

2) There is a problem of phosphorous availability in the dry zones

of Syria. More information is needed and currently being sought to deter

mine total national supplies of phosphorous, alternative uses and pricing

and distribution oolicies of the government.

3) Agricultural credit would need to be reviewed as

well if an in

crease in phosphorous application

is going to be recommended in these areas.

4) The use
of barley for grazing forage, etc., was highlighted in

this research and several new studies were begun

to better understand the

farmers' decision making process. (Nordblom, 1983a, 1983b).

More on-farm trials need to be conducted before any recommendations are

made to Syrian farmers; in an

area with such high climate variability, two

years of experimentation is not enough. Nevertheless, recommendations to

farmers are one aim of this research and the process will not be complete until

this
is done. Hopefully, this example illustrates the value of a farming

systems approach for developing appropriate technologies for target farmers.

IV. THE IMPACT ON AND DISTRIBUTION OF LABOR USE

Perhaps the most exciting research result to date at ICARDA is the

development of new chickpea cultivars that are resistent to Ascochyta

blight. Chickpeas are grown in most of the Mediterranean basin as an

important legume due to the beneficial role it plays (a) in the cropping

rotation (it is a host for nitrogen fixing rhizobia) and (b) in human

diets (it is both high in protein content and has an amino acid balance

that compliments cereals).

In much of the ICARDA region, it is planted in March after the

major portion of winter precipitation has fallen. By planting late

farmers avoid Ascochyta blight which is highly likely to attack chick

peas in the cooler winter months of January and February. However, re

search has shown that if the disease can be avoided, inc reased mois

ture use by the plants can increase yields two to three fold. Thus a

major breakthrough occurred when resistant lines to this disease were

found after intensive research efforts of the Food Legume Improvement

Program at ICARDA.

These new lines are now being tested on farmers' fields and are

about to be released by some national programs in the ICARDA region,

Syria being one of them. Biological scientists at ICAPDA are aware

of indirect as well as direct effects of technological changes, and

they requested social scientists to look at the impact of a change in

chickpea production which basically entailed using a new cultivar and

advancing the date of planting by 2 or 3 months. Thus they hoped to

minimize any negative effects of the new technology and identify in

advance any socio-economic constraints to adapting the proposed change.

 -19-

The first obvious benefit of biological scientists initiating

social science research was that we avoided ''so what?'' questions

being asked. An earlier research study of several villages in

Aleppo Province drew such a response by our colleagues from the

biological sciences when the report became available. Whether or

not such Questions are valid is irrelevant to this discussion.

What is relevant is th3t social scientists were reacting and respond

ing to questions posed by biological scientists and at the same time were

gaining the interest and cooperation, of the biological scientists.

Studying the impact on agricultural labor use was one aspect of

this research. Previous research at ICARDA had shown that new mechani

cal technologies were rather quickly adopted in Syria. Land preparation,

for example, is almost completely done by tractor today and an increasing

percentage of harvesting, particularly cereal crops, is being mechanized.

We also know that machines are almost entirely operated by m2n.

We developed working hypotheses which suggested that the recommended

changes in chickpea production are likely to run into the following

problems:

1) labor bottlenecks at the earlier planting date,

2) an increase in weed populations,

3) different labor demands at harvest,

4) increased chickpea cultivation in drier areas, e.g.

Zone 2 in Aleppo Province, and

5) altered forces operating on chickpea markets.

With these issues in mind, Rassam developed a study of labor

use in chickpea production with a particular focus on the impact of

the new technology on

female labor. Her study aims to describe the

existing family agricultural labor patterns as

well as hired female

for

labor, to anticipate probable effects of

the new recommendations

minimize

chickpea production and to proscribe alternatives that will

the changes
on families if the new technology is adopted.

Rassam visited 46 farmers in four villages in Aleppo Province

during the Il82/83 cropping year. Two of the villages were "n Zone I

located

where chickpea production is widely practiced, while two were

in a slightly drier area

in Zone 2 where chickpeas are not currently

grown. We anticipate that the ne%.' cultivar can

be used to introduce

data set

chickpea production to these villages.

The analysis of this

is not complete and will e reported by Rassam in the

near future

(Rassam, in
process). How.ever, some indications from the research are

available and presented belo,._

In Tatle V, specialization o agricultural

tasks by family members

is shown. Land preparation and chemical weed control

is completely

done by men. Fertilizer application and seeding are generally mascu

line tasks. On the other hand, hand weeding and feeding and caring

for toe animals is mostly done by

women. Seed preparation and the

various steps
in the harvest process seem to be shared jointly by

within the

men and wo-ner although there is also specialization by sex

harvest process itself. Aside

from weed control (in the wetter zone

more chemical control

was used), there is little difference between zones.

Table VI breaks down hourly

labor input into two cropping groups.

Female labor is more

important in legume production (62 percent) than

cereal production (42 percent), and hired female

labor both locally

and from other villages is very important for legumes. A sizeable

 TABLE V.

Sex of Labor for Various Agricultural Tasks

(Numbers of Households)

Zone I Zone 2
Female Male Mixed Female Male Mixed

Land Preparation

Mechanized 0 23 0 0 20 3

Seed Preparation 2 0 19 2 0 19

Planting 0 17 5 3 12 8

Fertilizer
Application 0 16 6 0 15 4

Hand Weeding 11 0 3 8 0 11

Chemical Weed
Control 0 14 0 0 5 0

Harvesting 0 7 15 0 3 19

Threshing 0 8 14 1 6 13

Winnowing 0 5 14 1 2 15

Animal Husbandry 12 1 5 12 2 6
 TABLE VI.

Labor Use Differences in Cereal and Legume Production,

Aleppo Province, 1982/83

(Percentages of total hours of labor input)

Cereals Lecumes

Fema 1e

Family 37 16

Hired from Village 4 18

Hired Elsewhere 1 14

Sub-total 42 62

Male

Family 22 16

Hired from Village 17 6

Hired Elsewhere 15 5
Sub-total 54 27

Children (12 years

old or less)

Family 3 4

Hired from Village 1 2

Hi red E!sewhere 0 5
Sub-total 41

Totals 100 100

 -23

amount of the male hired

labor would be associated with rented
equip

ment which usually comes with an operator. In general the female

labor

contribution was higher than we

expected, and women
will
certainly be

influenced by any technological change, specifically

by
a change in

legume production.

The greatest labor requirement

is at harvest time, and harvesting

legumes, particularly lentils,

is extremely labor intensive
(Table VII).

In addition, Rassam found that

72 percent of the legume
labor require

ment for harvest was done by females. Since cereal

crops occupy about

50 percent of the cultivated area compared to 25 percent

for legumes the above

data should be weighted accordingly.

The households in the sample had a large amount of

variability

in size and structure. The analysis has not vet

been done to examine

how the family demographic structure

effects cropping
choice, hired

labor practices, and choice of

technology. Nevertheless, there is

enough information to anticipate problems and effects

of the
new

chickpea technology. The important ones

are:

1) A labor bottleneck at planting is apparently less

important than we hypothesized; planting is not a

labor intensive activity.

2) Weed control, however, will be very importan t.

Earlier planting will increase

the weed problems on

the chickpea plots. Hand weeding these plots will

be most important for women

in the family. The

fact that most weeding is currently done without

hiring labor indicates that there is no labor

bottleneck at this time of the cropping calendar,

but a greater weeding effort may overtax

a family's

female labor pool.

3) If chemical weed control

of chickpeas is introduced,

this task will be undert3ken by the men and this will

radically effect
the labor allocation to chickpea

production. Whether such a change would be

good or

bad for women depends on the alternate use they make

of

their time.

 TABLE VII.

Percent of Labor Hours Allocated to Various

Tasks in the Production of Cereals and Lequmes

Cereals Legumes

Land Preparation
10.1 3.7

Planting
5.l 1.5

Fertilizer Application
11.0 0.9

Hand Weeding
20.0 16.1

Chemical Weed Control

0.8 0.1

Pest Control
8.5 2.4

Harvesting
26.2 58.4

Transport from Field

12.0 5.3

Thiesthing
4.6 6.6

Winnowing
0.4 1.6

Other
1.3 3.4

100 100

 -25

4) The change in labor

use at harvest could
be profound. First,

there
is already a sizeable requirement
for
hired female labor.

If the harvest process is not mechanized,

this requirement will

most likely increase due

to greater yields.
Secondly, the

demand for mechanized chickpea harvesting

will
increase and

appropriate equipment could be developed

is already developing legume harvesters rather quickly. ICARbA

and
threshers. Thirdly,

land currently allocated to lentil production,

particularly in

Zone 2, will shift

to chickpea production

if the new cultivars

are successful. Hand harvesting chickpeas

is simpler than

lentils and a decrease

in the labor requirement,
mostly hired

female labor, could be

important.

5)
The price effects of a substantial
increase in the
supply of

chickpeas in local markets are unknown

and require urgent atten
tion.
The demand for chickpeas is certainly
price inelastic and

price changes could be

severe. Information
on marketing activi
ties by family members also deserves attention.

6) Finaliy, the
household processing of chickpeas
and the
impact of

the new chickpea cultivars on

family nutrition
This is being done in a complimentary needs attention.

study by
a nutritionist

on the FSP team.

Answers
to many of these points can be
found
in the on-farm trials by

carefully observing and monitoring the

responses of the
farming family to

farmer-managed chickpea trials on

his
land. Thus, we
return to the process

discussed
in the first example. Joint-managed
trials
are currently under way

in a collaborative effort between

the
Food Legume Improvement Program, the

Syrian Ministry of Agriculture and Agrarian

Reform and the
FSP; Farmer-managed

trials are
the next step. Note that
FSR is conducted by other programs
at

ICARDA in addition to the FSP.

It is
possible that the
FSP will cease to exist

at
ICARDA when the Center organizes all
of its research along the lines de

scribed above.

 -26-

CONCLUSION

There are
more and more examples of results
emerging from FSR projects

that indicate the advantages of this approach.

Due to
the large number of FSR

projects
thac have begun recently, an increasing
number of
case studies will

become available. The experience of FSR

at
ICARDA, although slort in duration,

can also contribute to this

literature.

We have argued in this paper that

the contribution
of the Farming System

Program at ICARDA is unusual due to the way

we
have organized our research.

Examples of on-farm barley trials and

labor
use
in chickpea production were

given to
support this contention and illustrate

our approach. Key elements

in

our program are multidisciplinary

teams with
strong biological as well as

social science components and which study

not
only the whol,2 farming system

but the complete process of developing new

agricultural technologies. The

benefits of FSR,
we believe, are more
in
organizing agricultural research

projects and less a total approach

to rural
development. Its practioners

must be well
trained in specific disciplines

rather than generalists; FSR

is certainly not a discipline itself.

Therefore, we are
concerned at the
large
number of projects that are

included under the rubric of FSR. Many of

these,
for example., are no more
than a way for social scientists to get financial
support to do their sur
veys. The results of some efforts are bound

to be disappointing. And if

the answer to the question "So what's new?''

is "Nothing!", FSR will

fall

into disfavor and we will

have missed an opportunity
to build new, and we

think more effective, agricultural research

units.

 BIBLIOGRAPHY

Byerlee, D., et al, Planning Technologies Appropriate

to Farmers: Concepts and Procedures, Mexico City,

CIMMYT, 181.

2. De Datta, S.K., K.A. Gomez, R.W. Herdt, and R. Barker,

A Handbook on the Methodology for an Integrated

Experiment-Survey on Rice Yield Constraints,

International Rice Research Institute,

Los Banos,

Philippines, 1978.

3. "Farming Systems Proaram," Various Annual Reports, ICARDA.

4. Gilbert, E.G.,
D.W. Norman, and F.W. Winch, ''Farminq

Systems Research: A Critical Appraisal," MSU Rural

Development Paper, No. 6, Michigan, 1980, pg. 17.

5. Nordblom, T.L., "Livestock-Crop Interaction: The Decision

to Harvest or Graze
Immature Grain Crops,'' Discussion

Paper 9, ICARDA, 1983.

6. _, Discussion Paper 10, ICARDA, 1983.

7. Norman, D.W., ''The Farminc Systems Approach: Relevancy for

the Small Farmer," MSU Rural Development Paper, No. 5,

Michigan, 1980.

8. Nygaard, David F., ''Tests on Farmers' Fields: The ICARDA

Experience,'' Proceedings of Farming Systems

in the

Field Symposium, Kansas State University, 1982.

9. Rassam, Andree, The Role of Women in Syrian Agriculture,

M.S. Degree Thesis, University of Western Ontario,

(In Process).

10. Somel, K. "Barley Production in Syria," Discussion Paper,

ICARDA (Forthcoming).

1H. Thomson, Euan, ''On-Farm Livestock Trials,'' Paper presented

at Farming Systems Workshop, Kansas State University,

1 83.

12. Zandstra, H.D., E.C. Price, J.A. Litsinger and R.A. Morris,

A Methodology for On-Farm Cropping Systems Research,

IRRI, 198l.