6 /, 47 0
Lescothho Farr,mi ng_Siste__s Research Pro.i'ec t (Q632-0065)
Technical assistance to the Lesotho Farming Systems Research
Project (LFSRP) was provided by the Consortium for International
Development (CID), with Washington State University (WSU) as lead
university. The project's technical assistance contract began in
March, 1979, providing nine technical assistance (TA) positions.
Members of the technical assistance team (TAT) began arriving in
country in July 1979, and were fully on board by August 1980.
The original five-year project time period was extended for two
years to March 31, 1986, with a subsequent agreement to extend
the project to July 31, 1986.
The LFSRP was evaluated three times: an interim evaluation
in 1981 (Martin, et al., 1981); a special evaluation in 1983
(Dunn, 1983); and a final evaluation in 1986 (Frolik and
3.1 2ConcEt What was the basic technical idea underlying the
Lesotho is exceptional in Africa in that it relies on off-
farm income opportunities, principally outside the country, for
its people's livelihood. It is estimated that only 17% of house-
hold income comes from on-farm, agricultural activity.
The Lesotho Farming Systems Research Project (LFSRP) was
conceived to assist the newly established Research Division (RD)
of the Ministry of Agriculture (MOA) in conducting agricultural
research on more productive farm enterprise mixes acceptable to
farmers, sensitive to farmers' management ability, appropriate to
resource availability, and protective of the land base. As
identified in the PP (p. 13), "the thrust" of the project was "to
develop effective means to reach farmers and gain their
understanding and acceptance of the practices recommended."
3.2 Design How was this basic technical idea translated into a
The project's design provided a standard series of inputs
(technical assistance, training, commodities, and construction)
to be used to produce a number of outputs (a farming systems
research unit, a farming systems program, strategies for reaching
farmers, trained Basotho personnel, and a research and informa-
Several shortcomings in the project's design were identified
in the first and second evaluations. First, the evaluations
raised a question concerning how the initial and ultimate
beneficiaries had been defined. The second evaluation noted that
the PP had envisaged the initial beneficiaries to be farmers
indicating both a willingness and ability to try improved farming
techniques, with the implication that this group would be
composed primarily of the relatively better-off farmers. The
ultimate target group was identified as "those farmers or farmer
groups who indicate a reluctance to improve traditional
agriculture due to a lack of resources, financial or physical, or
knowledge that change is possible." This definition of initial
and ultimate beneficiaries "tended to overlook the importance of
classifying farmers on the basis of resources and/or farming
systems practiced and the need to develop agricultural
recommendations for each group" (Dunn, 1983:38).
A second design shortcoming was the idea of having extension
agents -seconded to the Research Division. The second evaluation
team found that this idea had proven to be less than satisfactory
and that the extension service should have been integrated as a
full partner into the project rather than seconding a number of
agents to the project. At the time of the second evaluation, a
systematic means of liaison between research and-extension was
being implemented by the project in the form of monthly meetings
involving the two groups. However, in the opinion of the
evaluation team, the extension service regarded activities in the
project's prototype areas (PAs represented the lowlands, foot-
hills, and mountains) as part of the research program rather than
an integral part of the extension service. In the team's view,
the extension service should have played a major role in
planning, designing, staffing, implementing, and monitoring any
trials or demonstrations being placed on farmers' fields (Dunn,
1983:43). As the team noted, the project
neglected the development of the district and national
extension service. As an example, project funds were
unavailable for inservice training costs to hold workshops
1; ;with district level subject matter specialists and other- ..'. .
district agents to participate in routinely held workshops, -'
Field extension tworkers3 outside of the prototype areas':: :'
were not provided with a means of transportation, i.e.,
motorbikes. While the Research Division provided funds for
printing and distributing extension circulars, the
Agricultural Information Office was constrained financially
in the amount of extension materials produced for its farmer
audience (Dunn, 1983:46).
A third design shortcoming was the limitation of the FSR
effort to the PAs. The implication of this design was that only
a small portion (1.0 to 1.5 percent) of the 240,000 households in
Lesotho would be potential project beneficiaries. The second
evaluation team recommended that a second phase of the LFSRP be
designed to cover a greater number of administrative districts so
that a greater portion of extension resources could participate
with the RD in on-farm research trials and planning extension
demonstrations on farmers' fields. The evaluation team also
noted that this strategy would reduce the chances of the project
favoring one farmer group over another.
3.2 Im 2ele1ntation How was the project managed by the host-
country implementing agency, the technical assistance team,
The first evaluation of the LFSRP noted a host of problems
that the project encountered during its early implementation:
a slow start by the Technical Assistance Team (TAT)
(caused by team members being selected without the
involvement of the TAT Leader, lack of orientation to
the project before leaving the States, delays in the
arrival of team members in country, TAT members not
arriving in the sequence planned, lack of orientation
assistance of AID when TAT members arrived in country,
inadequate introduction of TAT members and the project
itself to GOL agencies and other entities with which
they were expected to work, and delays in housing and
inability of project management (MOA, TAT, and AID) to
provide a unified approach to direct and guide planning
and implementing activities at the national and PA
lack of short- and long-term agricultural research
policy and strategy in the Research Division (RD);
limited number of skilled MOA professionals assigned to
delays in assigning counterparts to be trained to
replace TAT members (some counterparts were not
assigned until almost six months to a year after the
arrival of TAT members);
delays in selecting and processing participants for
academic training (this increased the likelihood that
there would not be sufficient overlap with TAT members
to provide on-the-job training;
selection of PAs that did not have access to inputs and
delays in assigning research extension assistants to
the project; and
-- minimal previous research on which the TAT could draw.
Another constraint became apparent, namely, that the drop
off of GOL budget support to the project hampered the implementa-
tion of trials and the provision of required follow-up. Yet,
despite budgetary limitations, high expectations were held for
the project. At the GOL and donor levels, there were expecta-
tions that the project would rapidly develop enterprise mixes
which could be used in small farmer development programs in the
country. And farmers in the PAs held expectations that the
project would provide inputs and services typically provided by
other development projects. RD personnel sought to reduce these
expectations through repeated explanations that the project was
not an area development project but rather action-oriented
research that would be slow in yielding benefits.
The Technical Assistance Team (TAT) provided the mix of
technical skills in agricultural production and supporting
services outlined in the PP. However, TAT effectiveness was
hampered by "uneven arrival" of team members in country and "the
absence of an ongoing agricultural research program and organi-
zational base" (Martin, et al., 1981:8). While the TAT assisted
in strengthening the foundation of the RD as a newly formed
research institution, the evaluation team recommended that the
TAT needed to play "a stronger role in the management and plan-
ning areas...to provide a sharper focus on reaching the specific
objectives of conducting relevant research and...transferring
technology to small holders" (Martin, et al., 1986:8). .
In the view of the first evaluation team, the project's : 7
designers had been unrealistic in thinking that a FSR Unit could
be established in the RD as a newly created research institution.
Indeed, the team found "a divergence Eof3 thought on the...extent
to which a Farming Systems Research Unit is being or should be
established within the Research Division" (Martin, et al.,
1986:8). Many RD professionals felt that the TAT should support
the building of the entire RD. Indeed, at the time, the organi-
zational chart assigned Contractor personnel to several sections
within the RD (Martin, et al., 1986:8).
Two problems, not anticipated in the PP, were encountered in
implementing the project: (1) the absence of a standard,
published set of crop production recommendations; and (2) the
extremely limited availability, particularly in the PAs, of
inputs (fertilizer, seed, chemicals and even simple oxen equip-
ment). Further, the evaluation team found that the project and
the RD were not implementing "a program of action specifically
designed to follow through on selected alternatives" (Martin, et
al., 1986:12). An absence of collaboration between farm
management and agronomy to identify constraints specific to each
alternative also was noted.
The first evaluation team felt the project's "greatest
chance...for...short run impact (probably its only chance)" was
to focus on the production of food crops (Martin, et al.,
1986:12). The team recommended that the project concentrate on
food crop production, that the research program not be restricted
to the three PAs, and that an agronomist be assigned to each
ecological zone of the country, with the responsibility of
attending the PA within the zone but also working as needed
outside the PA.
The team also pointed out that the project could address the
input problem by reporting and analyzing data on the severity of
the problem and the potential profitability of inputs under good
farming technology. The project also could assist in developing
"a self-sustaining solution of the problem--not a short-run
subsidized easing of the problem that cannot be sustained"
(Martin, et al., 1986:15). Further, the team suggested, farm
management efforts should focus on helping to develop improved
technology, by identifying the farmer constraints specific to
practices or systems being considered. This, as the team pointed
out, involves careful monitoring of the economics of the
technologies being tested.
The team noted the importance of the project assisting the
RD in developing communication links with the MOA Information
Section. Examples of such links could include assisting in the
preparation of technical publications and extension training
,-. materials, providing research reports to other MOA subject matter..
divisions, and conducting seminars for technical division chiefs
and district agricultural officers to explain the nature of the
agricultural research program, the production problems receiving
attention, and the results of station and farm-level adaptive
research trials which have provided information to disseminate to
farmers. These measures to strengthen communication lirks
between the project and other MOA divisions and offices would,
the team noted, aid the government's efforts to unify an approach
to agricultural development and expand the project's impact to a
wider audience than the farmers in the PAs.
3.4 Evaluation How was the project's performance measured or
At the time of the project's first evaluation, the TAT had
been assisting in project implementation for nearly two years.
At that point, the first evaluation team found that
there was no evidence that farmers are adopting...improved
farming practices developed under project-initiated
activities. The agronomic, range management and livestock
research activities already underway are at the beginning
stage of an applied research program. These research
activities will need to be carried on for a number of years
before a proven technology exists which can be disseminated
on a broad basis to the farming community. Accordingly, it
is uncertain whether or not the Project will reach the
stated objective of reaching five percent of the households
in the project area with enterprise mixes (Martin, et al.,
Consequently, in the team's view, "the normal start up period of
settling in and getting organized to do agricultural research
work" had impeded achievement of project outputs. In the team's
view, it was too early in the research process to determine how
farmers would accept and utilize new practices of relevant
technology (Martin, et al., 1981:21).
The evaluation team found that the project design team had
made a basic assumption that there was a considerable amount of
relevant data available that the TAT and the RD could collect,
analyze, and use to develop a research program without having to
"start from scratch." However, the TAT discovered that the
actual data base related to FSR/E was weak and spotty. While TAT
members collected and analyzed existing data, their efforts
produced mixed results, and there was no attempt to coordinate
and synthesize the data collected by individual team members.
For example, none of the materials collected had been used to
- develop a profile of existing farming systems in.-Lesotho. :-..',
'' Such a profile, the team noted, would -e useful in assisting
the RD and TAT "to reach a consensus on what type of farmers,
what extension strategies and what production technologies should
receive priority attention" (Martin, et al., 1986:19). "Lack of
consensus 20 months after initiation of the project as to who
precisely, the target population is and what types of innovations
are most likely to improve his/her farm enterprise is a
significant liability" (Martin, et al., 1981:28).
One of the difficulties in reaching consensus was the
existing split in policy orientation on agriculture. While donor
projects were aimed at the Lesotho small holder, the first
evaluation team found that the GOL was "engaged in a substantial
program of large-scale mechanized farming to make the country
self sufficient in food grains by using modern technology and
inputs in a...commercial operation" (Martin, et al., 1981:31).
Even within the project, the evaluation team found a split
between those who felt the project should aim at improving the
level of subsistence versus those who felt the project should aim
at developing a viable small-scale commercial agriculture. "The
project itself is divided on this issue" (Martin, et al.,
Despite splits in policy orientation, a baseline survey of
households in the PAs had been initiated and was nearing the
analysis stage at the time of the first evaluation. However, the
evaluation team felt that the formal baseline survey approach was
not an-efficient or relevant use of project staff.
A focus on more rapid methods of conducting farming systems
research (e.g., following the "Sondeo" method developed in
Guatemala or that developed by...CIMMYT for use in East and
Southern Africa) would have been a more appropriate approach
assuming the availability of local staff to carry out such
rapid assessment surveys (Martin, et al., 1986:19).
In a second evaluation of the project (Dunn, 1983:6), the
evaluation team found that the contribution of data that were to
be provided by the baseline study to the RD had been less than
desirable, and that this information should have been provided in
year two of the project at the latest.
Farmer records were also being developed by the project at
the time of the first evaluation. While the evaluation team felt
that this information would be useful in problem identification,
the team cautioned that the data largely described what farmers
do and needed to be supplemented with information on why farmers
do what they do. A subsequent evaluation (Dunn, 1983) found that
the project had made progress in identifying and classifying farm
households according to availability of resources and agricul-
tural production. However, the team recommended that greater
attention be given to matching trials with potential adopter
In terms of strategies for reaching farmers, the project
gathered information from farmers through meetings, a baseline
survey, farm records, and informal, individual contacts. The
project established Village Agricultural Committees to
communicate to and receive feedback from farmers. Work had been
completed and was continuing on producing and distributing
"Cropping Guidelines" and other technical publications. While
crop demonstration plots and/or communal gardens had been
established in each of the PAs, the project had taken no steps to
monitor and evaluate the effectiveness of alternative strategies
for reaching farmers.
A final conclusion of the first evaluation team with respect
to project evaluation was that the GOL, TAT, and AID should be
more systematic in their monitoring and evaluation of the
The three should plan a truly collaborative evaluation at
least once a year and should formulate specific benchmarks
directly related to agreed plans of action which can be
monitored by all parties on a more frequent basis (Martin,
et al., 1981:32).
By the time of the second evaluation, the team' found that
the farming systems approach had been integrated into the RD and
that farmers were being directly involved in field testing and
demonstration (Dunn, 1983:4). Further, training sessions were
being held for extension field workers as well as for Village
Agricultural Committees. The team also noted that the total work
time spent by the TAT in the field had increased measurably over
the years, with some professionals now devoting up to 75 percent
of their time in the field. More broadly, the project was
continuing efforts to establish an Agricultural Research and
Planning Coordinating Council and to assist the RD in drafting an
agricultural research policy paper.
The increased amount of information on the project available
by the time of the second as compared with the first evaluation
made it possible for the second team to focus on what the PP
referred to as objectively verifiable indicators (OVIs).
Evaluation of the project in this respect is now reviewed in
terms of each indicator and the project's progress on the
indicator at the time of the second evaluation.
FarminGgSstems Researh Unit .
OVIl. Research priorities are determined through the use of" .
social and economic benefit/cost techniques by -:12/79--.
-The evaluation team found no evidence that either technique
was ever applied to selection of research priorities.
OVI2. FSR Unit results are being published and disseminated
to all relevant GOL divisions and other donor project
activities by 12/79.
A system for reporting research and trial results was
established and a number of publications were prepared and
OVI3. The FSR Unit is benefitting from improved professional
relationships with worldwide research institutions by
The project had initiated, maintained, and strengthened ties
with international research institutes (CIMMYT), research sta-
tions in the Republic of South Africa, and universities in the
U.S. such as WSU and Utah State University.
OVI4. The FSR Unit is pursuing or considering a program for
replicating FSR/E after the project ends.
Inclusion of the TAT within the RD as a support group for
Division activities provided a foundation for institutionalizing
FSR/E in the RD; however, the first evaluation team recommended
that the concept of having a separate FSR Unit within the RD be
Farming Syst ems FPr og ram
OVI5. Three systems using alternative technologies developed
and tested in three physical environments by 8/80.
The first evaluation noted the lack of a reliable set of
crop production recommendations for Lesotho. The second
evaluation found the number of on-farm trials in place to be a
vast improvement over the findings of the first evaluation.
However, the team also found
a lack of agreement among RD staff and units as to...what is
the FSR methodology being employed by the RD.... . The
evaluation team feels some concern over the many concepts of
FSR held by either. WSU or Basotho staff in the RD. While we
are very pleased with the effort to develop the Lesotho
model of FSR, the fact remains that all station-generated
and imported technology must be verified on a representative
sample of Lesothoan farms selected from homoegeneous agro-
climatic regions before such technology is ready to.
demonstrate (Dunn, 1983:27-28). ..-
The first evaluation proposed steps to strengthen and expand
on-farm trials. One step was to give the Deputy Director full
responsibility for coordinating farm trials, to facilitate an
orderly transition of farm research responsibility from the TAT
to Basotho staff. While the second evaluation voiced "concern
over the many concepts of FSR held by either WSU or Basotho
staff," the evaluation proposed its own "FSR Methodology" in its
report (Dunn, 1983:52-61). The need for clarification on the
FSR/E approach to be followed was again echoed by the third .
evaluation's recommendation that "the FSR interpretation (there
are many) for Lesotho" be spelled out in writing, with copies...
made available to all concerned (Folik and Thompson, 1986:iv).
StategieLs for Reachin Farmers
OVI6. Alternative strategies for MOA farmer communication and
education developed and tested by 8/80.
The project initiated Village Agricultural Committees as an
experimental approach to reaching farmers in the PAs. Also, a
group approach was being used on communal vegetable fields and
grazing schemes. The team recommended follow-up on these two
approaches to assess adoption rates of recommended technologies.
The team also recommended that the project consider testing a
facilitator approach to communicating with farmers.
Trai ned Baisotho Personnel
OVI7. Basotho personnel trained and assigned to 26 positions
in FSR Unit of RD by 3/84.
While the short-term training had progressed well, tardiness
in obtaining qualified participants for long-term training during
the early years of the project had resulted in delays in parti-
cipants completing training and returning to the RD.
Research and Inforrmation Data ase
OVI8. Not stated in Dunn (1983:32-35).
The first evaluation had recommended that the TAT, working
jointly with RD staff and AID, should:
-- ...a) analyze and synthesize the available data related
to Farming Systems...; b) identify and classify Farming
Systems types; c) identify the immediate beneficiaries
of the Project (based on GOL policy and USAID growth-
with-equity considerations); and d) establish which
farming systems and which potential beneficiaries will
receive priority in research activities.
Identify and disseminate a few proven technologies as
soon as possible to give the farming systems approach
more credibility (Martin, et al., 1981:58-59). : "
By the time of the second evaluation, the project had
prepared some annotated bibliographies and demonstrated some
technologies. However, the team found "confusion" about which
technologies needed to be tried and validated on farmers' fields
and those "proven" ready for demonstration and dissemination.
Since the first evaluation progress had been made on identifying
and classifying households in a PA in terms of the physical
resources which influence farming practices and on conducting
trials and demonstrations representing a range of complexities
and requisite resources. However, the team found that some RD
researchers needed to better understand how each research station
trial and each farm-level trial or demonstration is related to a
potential adopter group.
Various RD units (Range Mangement, Farm Management, Rural
Sociology, Marketing, Extension/Communication) had collaborated
in collecting data. However, the second evaluation cautioned
that "it is crucial that the data collected be analyzed and taken
into consideration when determining priorities for crop, live-
stock, and range trials and demonstrations" (Dunn, 1983:33).
Further, while the first evaluation pointed to the need for the
project to understand why farmers do what they do, the second
evaluation noted that little attention had been given to:
the reasons for the practices followed by the farmers: a
sufficient amount of information exists on what farmers do
but not why. . Collection of information on the Whys
requires a very well designed research effort...(Dunn,
Accordingly, the second evaluation recommended that the
Give top priority to research aimed at understanding
the farmers' rationale for specific crop and livestock
practices and intra-household decision-making related
to key variables.
Continue work on classifying farmers and adapting
recommendations to the physical resources of each
Conduct an economic analysis based on data from
farmers' fields prior to classifying a technology as
ready for demonstration and dissemination.
-- Give greater attention to monitoring, to assess
adoption rates. :
End of_ Prij_ctStatus lEOPS) ... .
The EOPS was that at least five percent of the farmers
(about 146 farm households) in the project's PAs will be using
technologies developed by the project. Despite progress made
with on-farm trials, the second evaluation team cautioned that:
There is...a difference between on-farm trials and adoption
of improved farm technology. ... In the case of all
agronomic trials observed..., significant adoption probably
cannot be expected to occur before the 1984-85 or the 1985-
86 cropping seasons. Again, verification and demonstration
must dccur before adoption can be expected (Dunn, 1983:36).
However, by the time of the third evaluation (Frolik and
Thompson, 1968), sufficient data had become available to enable
tne evaluation team to conclude that the project design target of
reaching at least five percent of the farmers in the PAs had been
been achieved. A factor identified as a major contributor to
achievement of the design target was the role of the Village
Agricultural Committees (VACs). The use of the VAC had proven to
be "an excellent way of getting farmer and community involvement
in technology testing, transfer, and adoption" (Frolik and
Thompson, 1986:ii). For example, VAC members assisted in the
choice of research problems and farmers for on-farm trials.
A study by the LFSRP concluded that VAC members had been
"effective disseminators of agricultural information and
diffusers of innovations" (cited in Frolik and Thompson,
1986:37), with each VAC member influencing an average of 8.8
persons through a combination of telling, showing, and
facilitating the observation of agricultural innovations. Based
on an extrapolation from a sample of 54 of the 234 VAC members,
the team concluded that "it is likely that farmer contact group
members have diffused innovations deriving from farming systems
research to nearly 2000 persons" from 1979 to 1984 (cited in
Frolik and Thompson, 1986:37). The effectiveness of the VACs in
the three PAs resulted in the Extension Division of the
Department of Field Services adopting this model for all 10
extension districts of the country.
3.5 Institutionalization How did the project provide for the
implementing agency to develop a sustainable capability to
continue to perform the types of activities supported by the
The first evaluation found that the project design had not
adequately addressed the Research Division's manpower and
organizational needs. As the team noted, there were not enough
trained Basotho agriculturalists to work with the TAT as co-
workers and to leave the country for training. Further, existing
training plans did not permit allow sufficient time to recruit
and train national staff in functions that would continue after
project termination. Also, as earlier noted, the project
experienced delays in selecting and processing participants for
training. Only three participants had been sent for long-term
training in the U.S. as of the date of the first evaluation
report. The first of these, sent in 1978, returned with a M.Sc.
degree and became the Director of the RD. Overall, delays were
also encountered in programming short-term external and short-
term internal training programs.
To accelerate training and staff development, the first
evaluation team recommended that the project assist the RD in
preparing a manpower development plan to increase the total
number of Basotho receiving specialized training in agriculture.
Training could be accelerated by intensive courses and on-the-job
training in the RD as well as short-term training at the IARCs.
By the end of the project, the third evaluation teari found that
good progress has been made in degree-level training of RD
personnel. However, the process is a slow and costly one
with many participants entering U.S. universities at the
beginning bachelers level. There has been some, but not
extensive, use of non-degree level training at international
agricultural research centers and the U.S. there has been
an active program of short courses and in-service training
with counterparts. Nevertheless, with the departure of the
WSU team, the RD is not a viable research institution in
terms of the adaptive research goals set forth in its policy
statement (Frolik and Thompson, 1986:28).
Accordingly, the team recommended that AID continue su.pc~ort for
training RD personnel.
The team~ also recoinriiended that the proj-;-ct red"-i: "its
visibility as a Fariing Systermis Fro.je--t (tMr-ti n, et a-.,
1981:23) and that the TAT identify rore closely with tth- RD, by
orienting the project "to the devel opritent of the F'-seaErc
Division as a National Institution." The team noted that the RD
needs...to incorporate two fundamental criteria of...Farming
Systems Resarch. One of these is a firm knowledge of the
farmer and his system of farming and a sound ur-derstanding
of why that system. The second fundamental criteria is the
inclusi-on of adaptive or on-farm research, i.e., the testing
under farm conditions of technology before it is promoted on
a large scale for farmer adoption (Martin, et al., 1981:23).
A second constraint to institutionalizing FSR/E, noted by
the team, was the project's "confinement" to the PAs. These
areas, the team felt, had been made too small and restrictive,
and that work in each had been so intense that the project
appeared to be tab:ing on an area de-eli.ent rather t-i a
techn '.lggy irtnovation focus. At the sa li,, ti;,:e, ..,i- .:iin, i ;
the, FA~, the TAT hadc not facilitated thtr c---;-pi...,.- art
efft-ct ive workir'ig relationship between t!he F'r arni t!.- L strici.
Agricultural Office structure.
Adaptive on-farm research is only a very small step away
from result demrionstrations--one of the most effective
extension tools and district personnel, in their own
interests, not as a favor to research, may participate in
trials and be able to move new technology to farmers. ...we
see a need for the Research Division and the contractor to
initiate more collaborative research/extension activities
with the District Agriculture Offices. The district level
subject matter specialists could be tapped to assist in the
conducting and monitoring of adaptive research trials. This
joint collaboration at the district level will aid in
strengthening the professional skills of subject matter
specialists and provide a background for training the
extension offices (Martin, et al., 1981:23-24).
Accordingly, the first evaluation team recommended that the
project abandon establishing a Farming Systems Research Unit and
instead focus project resources on institutionalizing an
effective agricultural research and extension capacity in the
MOA. While the LFSRP could make progress toward developing this
capacity, the evaluation team felt that:
The development of a research/extension project must be
considered long term with a planning horizon of ten to
twenty years. Given the current state of research in
Lesotho it is not realistic to expect that enough can be
achieved in five years in developing institutional capacity
(Martin, et al., 1981:31).
The first evaluation team also noted that the PP had made "no
mention...of a longer horizon (15-20 years) which is always
needed to develop a purposeful agricultural research institution"
(Martin, et al., 1981:1).
Although the project output of a FSR Unit had not been
officially changed by the time of the second evaluation, all
parties (GOL, TAT, and AID) agreed that the project should
strengthen the overall RD program rather than establish a FSR
Unit. However, while the expansion of the project to work with
the entire RD was good for research, the allocation of a greater
amount of project staff and counterpart time on non-FSR:,
activities was partially responsible for a delay in implementing
farm-level trials. Such a dilution of effort was not necessarily
bad. However, the reorientation of the project should be taken
into account in evaluating expectations regarding what the
reoriented project could reasonably be expected to accomplish as
a FSR/E initiative, especially in view of the project having
abandoned the output of establishing a FSR Unit and having
adopted the output of strengthening the research/extension
capacity of the RD.
In terms of institutionalizing a methodology for FSR, the
second evaluation report noted that:
the "complete how to" of FSR, from the initial stages of
problem diagnosis and farm-level testing to the final stages
of demonstration and subsequent adoption, has yet to be
developed for Lesotho. Since the 1982-83 crop year
represents the first attempt at systematic on-farm trials,
much of the planning necessary for subsequent phases of FSR
will fall on the RD between the upcoming harvest and the
1983-84 planting season. This evolving methodology, when
finished, will allow extension of FSR to other areas of
Lesotho. Also, by relying on the many Basotho researchers,
extension agents and farm record managers, the Lesotho
method of FSR will be developed jointly between the
contractor and the local staff. Such a joint development
means that the skills to extend FSR to other areas of the
country will be left with Basotho researchers in the RD and
the extension division (Dunn, 1983:21).
The second evaluation team recommended that the project make
a greater effort to involve CIMMYT FSR outreach staff and ICRISAT
staff in planning on-farm trials in future cropping seasons.
This recommendation implies that the FSR/E expertise required for
planning on-farm trials may have gone beyond that of the TAT.
Progress toward institutionalizing FSR/E in Lesotho by the
end of the LFSRP was summarized in the project's final evaluation
report (Frolik and Thompson, 1986). While the RD has a strong
orientation to farmers' problems, excellent links to farmer and
community groups, and adaptive research in farm management,
marketing, rural sociology, and extension, "with the departure of
the WSU team, the RD is not a viable research institution in
terms of the adaptive research goals set forth in its policy
statement" (Frolik and Thompson, 1986:28). Further,
the RD does not yet have the institutional capacity to carry
out an effective adaptive research program without
continuing technical assistance. The critical mass of
personnel is lacking in all sections and collectively. Some
disciplines received little, if any, support from the FSR
project. Capacity to plan, lead, and implement an
effective, well-balanced, adaptive research program is a
critical need (Frolik and Thompson, 1986:iii).
The team felt that the chief problem was the lack of enough
adequately trained and experienced staff members to provide the
But some significant progress had been made. The TAT and
the RD had successfully oriented the RD to conducting FSR closely
tied to farmers and farm problems. However, while the project had
made progress in working with farmers, similar progress had not
been made in "building the production research capability of the
RD including the Station and substations" (Frolik and Thompson,
1986:33). Accordingly, the key area identified in the third
evaluation as needing strengthening was the "research station
base of adaptive research in the production disciplines and a
clear understanding of the need for a balanced program of
research stations and substations and/or PA headquarters
experimentation, and on-farm trials, tests, and demonstrations"
(Frolik and Thompson, 1986:ii).
The evaluation team called for assistance to the RD to
continue as a component of the follow-on Lesotho Agricultural
Production and Institutional Support (LAPIS) project. Also, the
team recommended that the RD greatly reduce the number of "on-
farm" replicated field trials and increase the quality and
precision of on-station replicated experiments to maximize
production of reliable data, allowing on-farm demonstrations to
provide farmers with first-hand information.
Dunn, James F.
1983 Project Evaluation Summary of Special Evaluation of
Farming Systems Research Project (No. 632-0065),
conducted by Cal Martin, Dan Galt, and Carolyn Barnes.
Frolik, Elvin F. and William N. Thompson
1986 Final Evaluation of Farming Systems Research Project
(No. 632-0065). (XD--AAV-915-A)
Martin, Cal, Ken McDermott, Ned Greeley, and Tom Bebout
1981 Interim Evaluation of Farming Systems Research Project
(No. 632-0065) .(PD-AAI-396).