Transition to sustainable agriculture: an agenda for AID

Material Information

Transition to sustainable agriculture: an agenda for AID
Ad Hoc Committee on Sustainability in Agriculture, Committee on Agricultural Sustainability for Developing Countries
Place of Publication:
New York, New York
Ad Hoc Committee on Sustainability in Agriculture, Committee on Agricultural Sustainability for Developing Countries
Publication Date:


Subjects / Keywords:
Farming ( LCSH )
Agriculture ( LCSH )
Farm life ( LCSH )


Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact Digital Services ( with any additional information they can provide.

Full Text
Ad Hoc Committee on Sustainability in Agriculture June 18, 1987

Objectives 1
Our Internal Goals 1
Sustainable Agriculture Defined 2
Guiding Principles for Agricultural Sustainabilit: Project 6 Other Important AID Programs Affecting Rural Development 9
What AID Should Do in Rural Development 10
Specific Elements that should be Included in AID Projects 10
Soil 11
Crop Systems 12
Water Management 13
Energy: 13
Wetlands, Coastal Systems, Forests & Grasslands 14
Rural Support- 14
Research 15
Adaptation and Spreading of Sustainable Farm Technologies 17 AID's Non-Agricultural Programs Closing Related to Agriculture 20
Institutionalization of Sustainable Agriculture 20
Personnel 21
AID Policy: and Planning: In Washington and in
the Field 22
AID Leadership on Agricultural Sustainability-with Developing Countries 23

This paper represents an initial effort by a committee from within the environmental, development and private voluntary organization communities to define in practical, non-technical terms what we mean by sustainable agriculture. In that context we have set out what we believe and hope AID should do to help the farmers and, secondarily, the scientists, private organizations, and governments of developing countries make the
very necessary transition to agricultural sustainability in a timely fashion.
Our objectives are to urge AID to consider the following:
1. Institutionalize sustainability as the norm for all its .
agricultural programs and policies.
2. Focus more of its agricultural and rural development program
on helping farmers of developing countries make the
transition to sustainable agricultural systems.
3. Discourage the use of non-sustainable agricultural methods
and inputs.
4. Progressively assume leadership in encouraging governments
of developing countries, development agencies, and global,
regional, and national research organizations to commit
significant resources for helping farmers, particularly poor
farmers, make the transition to sustainable agriculture.
Our Tnternal-Goa1s1. Carry out with AID a continuing, mutually beneficial and
increasingly well-informed exchange of views about
agricultural sustainability 'in general and AID's agricultural program and projects in particular.
2. Develop a better capacity to evaluate "in-place" projects
focussing on sustainable agriculture, and in particular
AID's projects.
3. Utilize our ability to mobilize effective support with
Congress for AID's efforts to assist farmers in developing
countries to make the transition to sustainable agriculture.

Sustainable Agricultural Systems Defined
Sustainable agriculture is but one, if perhaps the most
important, building block for overall sustainable development, a subject which we don't fully address.
We think of sustainability in agriculture as the ability of an agricultural system to maintain production, over time, in the face of ecological difficulties and social and economic pressures. More importantly, from a working viewpoint, we regard sustainable agricultural systems as follows:
1. Systems that maintain and improve soil productivity,
quality, and tilth.
2. Systems that augment the potential for achieving the highest
possible efficiency in the use and conservation of basic
farm resources (soil, water, sunlight, energy, and farmer's
3. Systems that incorporate as much biological interaction as
possible: for example, mulching, the use of nitrogen fixing
plants, the use of agroforestry techniques, and the use of
intercroppingand- crop rotations to control pests and weeds.
4. Systems that minimize the use of health endangering and
environmentally damaging external inputs (some chemical
fertilizers; non-selective pesticides and herbicides; and
some forms of energy) and, instead, maximize the use of
available, affordable, renewable, and environmentally benign
5. Systems that avoid the contamination of groundwater by using
only those fertilizers, pesticides and herbicides that do
not penetrate below the plants' growing zone and then only
in controlled doses.
6. Systems that meet the needs of farm families for energy to
work their land, cook, and heat from readily available and
affordable energy sources.
7. Systems that strengthen communal cooperation, that protect
rural survival systems, that through community support and
sharing allow farm families to keep going in difficult times
(famine, drought, and natural or political disasters), and
that make possible effective local management of communitycontrolled common property resources (ponds, woodlots, grazing lands, irrigation systems) in ways that permit
equitable sharing in benefits.

The achievement of sustainability and therefore the
protection of the natural resource base requires improvement in the lives of the poor majority. Unless there is a significantly greater return for a farmer's investment of time and whatever limited capital he may have, he will have no incentive to adopt ecologically sounder practices. It is because of this that we emphasize the close connection between improving all aspects of the well being of the poor majority (economic, social and nutritional) and agricultural sustainability.
Two additional important elements integral to sustainability which must be developed at other levels are the
1. willingness of research organizations, economic development
agencies and banks, and developing country governments and
institutions to accept the need for continuing evolution of the concepts and practices of sustainability and to provide
long term commitment to their achievement, and the
2. growing need for changes in the attitudes and practices of
governments in some developing countries so as not to
discourage but, rather, to provide encouragement and
incentives for the adoption of sustainable agricultural
We believe strongly that in a world of rapidly growing population pressure and rapidly increasing deterioration of natural resources, sustainability in every aspect of development
is an imperative. For us, sustainability in agriculture-nationally and internationally--is vital. In this paper, however, we only address the narrower questions of how AID should carry out its mandate to assist the poor in developing countries and,' more specifically, how it does this through its agricultural and rural development programs.
Central to our thinking is the proposition that AID should do much more to make agricultural sustainability a central focus of its agricultural and rural development program; and that given the vital importance of agricultural sustainability, projects to carry it out should enjoy the highest priority within AID, and even precedence over other aspects of agricultural development. We are convinced that in the light of budget cuts, present and possibly prospective, agricultural programs based Bn sustainability can better meet AID's mandates--in particular poverty alleviation--than can its current mix of agricultural programs. We urge that every aspect of AID's agricultural portfolio be reconsidered to ensure that everything aims at and contributes to sustainability and that nothing is done which detracts from that objective.

To achieve sustainability, we urge AID to concentrate of\
what it does better than other things (human development) and on
what most needs to be done (offering techniques which make sustainability possible under present conditions and limitations available to the poor majority of Third World farmers). Poor farmers need not only new techniques but also a limited number of additional resources of the kind that PVO's and community organizations can provide. With few exceptions, however, other important and more expensive agricultural development programs should be left to the World Bank, the regional banks, and other bilateral and multilateral givers of economic assistance.
Fortunately, a transition towards sustainability as a
central focus for AID should be relatively easy, for in our view AID is already embarked on the iight ship. To its credit, AID has already thought more and done more about agricultural sustainability than most development institutions. In a few good projects it has dealt creatively with sustainability. Many of AID's agricultural and natural resource management specialists and policy makers accept and.even advocate giving sustainability a central place in its agricultural programs and research. A growing number have practical experience in its application. And AID's present portfolio of agricultural, rural development, and natural resource projects provides a good base for expansion in
the right direction.
AID's most recent high-level formulation of agricultural policy, "Focus for Agency Agricultural Programs" dated May 1, 1987, reflects, in our view, important policy elements that favor sustainability: "maintaining and enhancing the natural resource base", a mandate that surely encompasses all or most aspects of sustainable agricultural systems as we see them, and a concentration on the "poor majority" of Third World farmers.
However, we believe that this policy statement should be slightly expanded. First, there is no recognition of the centrality of agricultural sustainability 2e/_.. Also the goal "to increase the incomes of the poor majority and to expand the availability and consumption of food" (presumably for all Third World people, not just farm families), while important to include, is not comprehensive enough. We recognize that poor farmers without income are too often just on the fringe of some market economy and, therefore, at a disadvantage in acquiring the things they need for a better life, let alone new technology that could help them achieve sustainability. However, in our view even more important than cash income for many poor farmers is food security. This concept is possibly included in AID's definition under "food availability" but the statement should be expanded in our view specifically? recognize the importance of food security. For poor farmers, the imperative must be to feed their families in good times and bad; however, a second imperative, both for them and for us, must be to do so in a way that does not degrade the lands they farm.

In our view, the focus statement has another shortcoming. It does not seem to reflect the importance of increasing farm productivity (not production), productivity which can and must be increaseiw-ithin sound environmental bounds if poor farmers are ever to achieve a better life. We believe that sustainability must encompass this concept of both increased productivity and increased economic return for poor farmers. Even the poorest have a desire to improve their situation and not remain just marginal farmers.
Despite favorable attitudes and some good projects, we fear that AID's commitment to agricultural sustainability is still fragile and still lacks the institutionalization which will ensure its survival and bureaucratic prosperity. AID's work on sustainability is still spotty. It is not yet conceptualized as it relates to every aspect of agriculture. It is not yet well enough integrated with AID's broader-based efforts to improve natural resource management and not well enough linked to programs in forestry and watershed protection. Furthermore, it is not yet well enough networked as a separate, complicated multi-sectoral discipline in a way that the lessons AID and others have learned about agricultural sustainability become part of the Agency's institutional memory and are easily available through its computer network.
Another weakness that AID shares with other development
agencies and the global academic community is the difficulty of
designing and then implementing true cross-sectoral projects in sustainable agriculture. The reductionist trend in the academic world has left its mark on most professions represented in AID.
It's still difficult, for example, to get crop-oriented agronomists to work closely with foresters. Agriculture and natural resources management are too often perceived as being separate sectors. While AID had made progress in overcoming inter-professional distrust and lack of understanding, it still has a way to go as do too many of its frequently used consulting firms. Strong interdisciplinary leadership is still badly needed.
Most important of all, we believe that AID's first need now is to gather and analyze its rather disparate experience and then determine how best it can adapt this experience to the many agricultural environments in which it is working. The United States knows enough about agricultural sustainability to make major contributions now.
Unfortunately, as far as we can determine, AID has yet in any country to carry out enough agricultural projects where sustainability is a major focus to have more than very local impact, let alone regional, national, or international impact, or even real impact beyond a narrow circle of experts. In our view there is an urgency about mounting efforts to promote one aspect

of agricultural sustainability largely overlooked by AID--by indeed other development agencies too--and by Congress: the fate of the poorer farmers on the more fragile lands who too often unwittingly contribute to the destruction of the very natural resource base on which their own families (and, ultimately, people in their own country and beyond) must depend for food. Heretofore, ecologically sustainable methods of slash-and-burn with long rotations were available to farmers cultivating poorer lands on every continent, although it must be admitted that their standard of living was both economically and nutritionally poor. Today, however, the growing number of farm families and the lack of land for long rotation makes even this relatively unsatisfactory method difficult and even impossible.
We suggest six general principles to guide AID's work in
agricultural sustainability.. None of them is new to AID, but all need reinforcement.
1. Concentrate major effort on developing the human and
material resources necessary to put sustainable agriculture systems within the grasp and understanding of poor farmers.
This means, in part, more AID attention to participatory and
"bottom up" work with farmers.
2. Defend the existing natural biological diversity of
agricultural areas in all AID's agricultural and rural
development policies. For us this means, inter alia, moving
from promoting large-scale monoculture to greater
utilization of polycultural methods. While
institutionalization of polyculture on a farm-by-farm basis
is not realistic, nevertheless for both ecological and
nutritional reasons, greater introduction of a broader crop
base on at least a community-by-community basis holds real
3. Enlist American private voluntary organizations (PVOs) in
the effort to help poor farmers adopt technologies of
agricultural sustainability AID should be more imaginative
and determined in its effort to recruit or develop
indigenous PVOs for this task.
4. Maintain a high level of support for research that
emphasizes sustainable agricultural systems, particularly as
it effects farmers on fragile lands.

5. Assume a greater leadership role in promoting sustainable
agricultural systems: with Congress, with those in the
American agricultural, scientific, and academic communities
concerned with Third World .agricultural development; with other development agencies, specially the World'Bank with
whom AID should develop joint or parallel programs; and with Third World governments and appropriate private institutions
in the developing countries.
6. Long-term public commitment to programs and projects in
support of agricultural sustainability, a commitment, if
possible, backed up by Congress.
This is specially important in this day of changing AID
priorities and funding levels. One of AID's principal
weaknesses in rural development has been its failure to
spend the necessary time and resources to bring projects to
fruition and then see to the wider adaptation of the lessons
learned. It often takes years and some false starts before
cultural, political, economic, and agricultural structures
are understood and impediments to success are mastered,. An_.-.
agricultural project--in fact, any development project-cannot be said to be "sustainable" in our view until there
is a high probability of an ongoing flow of benefits to
farmers once AID stops its assistance. This means in
practice that trained and competent local people are in
place who can carry out the functions (including support and
encouragement) that outsiders formerly filled and that an
institutionalized provision of necessary resources, such as money, technical assistance, and encouragement continues to
be available.
We will have more to say about these principles later in the context of suggested AID programs. However, human development is so central in our view to AID's urission and capacities that a few basic thoughts on this subject seem warranted at this point. First, there can be no doubt that the human problems of getting sustainable agricultural systems adopted by farmers and accepted by governments is an even harder task to accomplish than finding
technical solutions. Typically, AID experts and others wellgrounded in sustainable agriculture tend to be relatively clear about the technical aspects of this problem, but less clear about how to go about solving the human problems. They quite correctly point out how "site-specific" each project must be in human as
well as in ecological terms.

Second, at the level of the farmer, we know of very few
occasions when farmers have accepted the new disciplines and additional complexity of sustainable systems until they were up against a wall--from drastic and consistent reduction of productivity, from getting sick (or their families getting sick) from farm chemicals, or, in the case of a very small number of richer and better educated farmers, from being convinced that only thus could they help stave off financial ruin.
Third, and closely related, the economics of sustainability are as important to the subsistence farmer as to the more favored farmer. Farmers' willingness to adopt sustainable practices depends more on whether it is profitable than whether it is environmentally sound. The interrelationship between profitability and sustainability is not always apparent to the poorer farmer, but this must be at least the initial focus of how sustainability is presented.
Fourth, perhaps the hardest of all in human terms is finding ways to present or help present the methods of sustainability to farmers ("extension") Methods acceptable both to the farmersthemselves and to central government authorities have proved veiy difficult to devise. Central government controlled and financed extension systems on the US/European models have very often proved too expensive in terms of recurrent costs, and quite often
unacceptable to farmers who resist close contact with agents of the central government. Nor have these systems been particularly efficient. At the root of these problems is the global .rural/urban split, sometimes complicated by ethnic or religious factors. The answer to this problem, as with many others, seems to lie in Third World governments encouraging a greater degree of decentralization and rural empowerment. Unfortunately, we can find very little formal reflection of AID's understanding of this latter highly political aspect of the urban/rural problem.
One more aspect of human development policy to which we attach particular importance needs to be underlined: AID's agricultural program must. in our view, have an even Atroncer grounding in the collaborative and participatory approach. AID's basic task should be to orient and then to train developing country people at all levels in the potentials, problems, and procedures of sustainable agriculture. AID seems well launched
IThe term extension" is used for shorthand purposes though it too often has for Third World farmers a strong connotation of American and European cultural bias.

into collaborative work at the governmental and technical levels in developing countries. However, despite good rhetoric, not enough has been done in our view to ensure "dirt farmer' input, into program adaptation. Not only the awareness and sensitivity of the cultural anthropologist is needed. Also, the techniques of soliciting the farmers' own perceptions of their capacities and enlisting their "ownership" of whatever is new needs much more attention. The farm family and then the village and its chosen organizations must be the most important objects of AID's attention. Respect for farmers' understanding of their own environment and of rural values generally, including very specifically the crucial role of women, is essential.
Projects aimed at sustainability, while nec essarily complex and multi-faceted in design and while ideally bringing to bear inputs from several scientific disciplines, must fit the farmers' available time, must build from existing farm practices, should support existing survival strategies, and must be simple enough to be credible and understandable. Building simplicity out of complexity is admittedly a very difficult task.
All projects should be low cost, from the farmers'
perspective, and be based on adapting proven technical packages to local environments. They should build from existing farm practices and should use only easily available, affordable, and environmentally safe inputs.
In designing and carrying out agricultural projects, AID
must ensure that benefits are equitably distributed. This is a vital element in sustainability. It must also ensure, as some
-donors have not, that existing common property rights (communal
-woods, ponds, irrigation systems, grazing areas) are respected
and that the costs of adopting new techniques do not unfairly fall on any particular group.
AID has other rural development programs which are important to achieving sustainability, which we endorse, and which we will discuss later: population programs; rural reforms by central governments, sometimes supported by AID subsidy; an~d, very specially, land tenure and land entitlement programs which give landless farmers the incentive to protect and build up the lands they farm.
Other very valid aspects of AID's rural development policy are covered only in passing in this paper: health, general education, women in development, nutrition, farm housing, etc. We do not downgrade the importance of these programs but choose not to discuss them in detail in order to concentrate-attention on the ecologically oriented aspects of rural development.

We recognize that there is a complex set of
interrelationships between and among all facets of rural development. However, we urge, on the basis of AID's unhappy experience with the too broadly conceived "integrated rural development projects" of the '70s, that at least at first projects focussed on achieving sustainability should not include too many components. AID in any given village where it is involved in agricultural projects should normally introduce additional components such as assistance to health and education only incrementally and as farmers and their families grow into the system. There will be exceptions, particularly with PVO-run projects, but the general rule still holds.
AID's work in helping farmers in developing countries make
the transition to sustainability in agriculture is so important that it should largely leave to other development agencies most
other approaches to agricultural development, particularly thoseinvolving the construction of "public works": engineering rehabilitation of major irrigation systems; rural highway construction; agricultural storage, if this involves major construction; and the construction of buildings for government research and government services. The World Bank is usually willing to finance such "bricks and mortar" projects and, in fact, usually does them better.
In sum, we repeat our view that AID should focus its
agricultural programs and projects on introducing sustainable systems; and in doing this, by concentrating on what it does best; on what most urgently needs to be done; on what other development agencies cannot do as well; and on whit it can find the resources to do. On all counts, helping teach people at many levels to adapt to the many ecological and economic situations in which it works the new techniques of sustainable agriculture fills the bill.
The following pages include a list project elements that we believe will advance agricultural sustainability. It is meant to be suggestive rather than all-inclusive. Some project elements could and do appear under more than one heading.
We use the term "project elements" to emphasize the need, for example, not to have soil projects as such but rather to include elements dealing with various aspects of soil conservation or productivity in every agricultural project.

SOIL! Systems and Project Elements Which Conserve and
Improve Soils
Soil conservation is an area of special American expertise. AID has been relatively strong in soil research and mapping but weak in soils management conservation.
In our view, not enough has been done to *introduce simple, cheap soil conservation techniques into AID agricultural projects, with the exception of a few goodPVO-sponsored projects. This should include:
o Protection against water erosion: planting ground cover;
tree planting along terraces; bench terracing; bunds on steep areas; check dams; infiltration ditches; terrace construction, when cost-efficient or when built as the
result of village decision; gully plugs; low dams; stone and
trash lines; minimum till; alley cropping.
o Protection against wind erosion: dune stabilization by
palisade construction and tree planting; trees for wind
breaks; planting of ground cover.
o Restoration and maintenance of soil fertility: use of
mulching; crop residues; animal and "green manure"; dispersed farm tree planting and other agroforestry
practices; methods to increase organic content of soil;
particularly in arid and semi-arid zones.
o Restoration of abandoned lands.
Almost all soil preservation and restoration projects involve substantial investment of money and time. Some are relatively cheap (trash lines) and some expensive (terracing). Outside financing is very often necessary. However, such projects should not be undertaken unless farmers are genuinely convinced that such projects are in their vital interest, make a strong case for them, and are willing to invest their own time and even some funds in their construction and maintenance. We believe that major attention should be given in all AID's agricultural programs to basic, well-understood soil and water management as an integral part of all that AID does.

CROP SYSTEMS: Project elements that adapt to local use crops and
cropping systems (a) to increase productivity or crom
security, particularly in difficult environments (low
rainfall, poor soils, etc.); (b) to augment soil
fertility. (c) to reduce the need for outside inputs
for pest and weed control? and (d) to improve the
quality of rural life (better nutrition, sources of
cash income to purchase inputs, basic goods not
produced on the farm. and amenities).
- Crops that reduce the need for fertilizers whose use
leads to nitrate pollution or the destruction of soil
structure or quality.
- Nitrogen fixing systems: alley cropping; use of
nitrogenous trees; leguminous cover crops, etc.
- Systems that provide for better crop spacing, better
timing of planting, intercropping, crop rotation, and use of biological interacting plants for weed control.
- Low till systems.
- Improved crops which have been adapted to acid,
saline, or other problem soils, which increase drought
and stress resistance, and which increase pest
resistance, thus reducing the need for those
pesticides which destroy pest predators or pose health
- Relay cropping, polycultures, and crop rotation.
- "Minor" crops, both annuals and perennials, that
provide better farm diet or that can be processed to provide cash income, including honey, gums, tanning,
edible seeds, fruits, leaves, bark, and fibers.
- Agroforestry combinations, including forest and home
gardens, particularly for moist tropical forest
situations, where on-farm production can also provide
wood, forage, fruits, medicines, and other products;
project elements which will make it unnecessary to
resort to primary forest destruction.

Incorporation of animals including small ruminants and
fish (aquaculture) into farming systems to provide power, organic fertilizer, and new sources of food.
Incorporation of tree or other plants or animals which
can be processed on the farm for cash income at
periods of cash shortage.
We support the introduction of more productive and more sturdy species which are developed through agricultural research when and if they meet actual farmer requirements. However, we urge that AID not wait for "the perfect species" and put less emphasis on crops as commodities and more as part of a total farm system. We support the move away from monoculture and towards the preservation of crop progenitors.
WATER MANAGEMENT: Low cost systems or techniques which provide
for conservation of water and improved and locally controlled water management at local village level.
Small scale irrigation systems, including systems
which prevent salinization, waterlogging, and water
borne diseases.
- Watershed management and runoff control on or above
village lands.
- Deepening of wells and low-cost pumping.
- Dry land techniques for water conservation: field grid
systems, water spreading, water channelling, and water
harvesting (infiltration) systems.
- Water storage conservation methods.
ENERGY! Project components and systems which meet needs for
energy at the farm or village level (preferably from
energy). renewable energy sources for reducing the burden
of heavy farm work. for heating. and for cooking.
Introduction of draft animals (absence of good or
cheap animal fodder is often the chief constraint).

Home and village tree planting for fuelwood (often a
by-product to the production of construction
materials, animal fodder, food, and the provision of
Biogas digesters (also supply nutrient rich slurry),
bicycle power, etc.
- Solar energy for crop drying and preservation.
- More efficient use of biomass fuels through better
stoves, improved techniques for the production of
charcoal, etc.
techniques that provide for sustainable use and
protection of village-exploited forests, aquatic systems-and grasslands.
Improved management of natural forests and equitable
distribution of benefits.
Improved management of wetlands and coastal systems.
- Development of local fisheries, particularly as a
.. .... supplemental source of farm income and better
Rangeland rehabilitation (AID should also continue
some practical experimentation to find more efficient
traditional range management systems despite
widespread failures to date).
RURAL SUPPORT? systems and techniques which make possible more
efficient and equitable village-level support of
agricultural production. storage, and commercialization.
Farm or village level crop storage.
- Locally controlled marketing systems.
- Local solutions to meet the need for more cash at the
time of purchase of seeds and other inputs through the

planting of supplementary cash crops and the
establishment of local micro-enterprises to process
agricultural products (too many'top-down farm credit
systems have proven to be unsustainable because of
high cost and difficulty of financial control and
Locally controlled means of purchasing and reselling
needed farm inputs.
- AID, as a matter of high priority, should restore its
previous level of financial support to the CGIAR
research institutions and a limited number of other research institutions like ICRAF, the International
Fertilizer Development Center, and CATIE. While some of these organizations have made a start in addressing
sustainability, AID should continue to use its
intellectual influence and financial leverage to
persuade them to concentrate more of their time and attention specifically on sustainable agricultural
systems and on helping the poorer farmers on fragile
lands to increase their productivity and food
security. These research institutions should be encouraged to find better ways to get meaningful
inputs from dirt farmers themselves; from the outset
of research planning to the adaptation of this
research to various ecological situations, farmers
must be involved. These research institutions should
continue to concentrate their genetic research on
plants which can resist climatic and soil stresses as
well as combat pbsts and weeds. They should be
encouraged to be more imaginative and creative in
finding ways to deliver the fruits of their research
to farmers when Third World government research and
extension services are weak or unpopular with farmers.
We urge continuing attention to collaborative social
science research related to the establishment of sustainable agricultural systems. Two important
subjects that come to mind are (1) how to measure
success in achieving sustainability (both human and
agronomic indicators), and (2) the attitudes of
various governments to decentralization and
empowerment of rural citizens, information which is
vital to planning successful programs in sustainable
We agree with AID's policy of helping to build up
national or, in some cases, regional research
capacity. AID has already decided it cannot help

build research centers in every country, but rather
that it should concentrate on helping countries with the most promise and potential for efficient regional outreach. AID should encourage research institutions to move away from classic export-crop and commodityoriented research towards giving more attention to the problems of the poorer farmer and to helping solve the
problems of transition to sustainable agriculture.
So far, US academic commitment to agricultural
sustainability is thin, and AID should encourage the
U.S. academic community to move faster towards a multi-disciplinary, sustainable approach to Third
World agriculture. In this regard, we consider BIFAD
to be an important vehicle for encouraging
collaborative research between U.S. academic
institutions and Third World scientists working on
agricultural sustainability.
AID should continue and even accelerate its efforts to encourage, train and utilize American academic experts on both the biological and social scientific aspects -/
of agricultural sustainability in the developing world. While individual experts in the various
disciplines involved in designing and implementing
sustainable agriculture systems are available, there
are fewer American academic experts on countryspecific agricultural problems. AID should help
create and then use a few centers of excellence where
all country-specific aspects of development, including
in particular agricultural sustainability, will be
Closely related are efforts, in which AID has played a pioneering role, to support networking, data gathering,
and retraining of scientists, all related to
sustainability. AID should also continue to support
the growing and laudable trend towards scientific
exchange and cooperation among developing countries,
which cooperation is already quite effective among some
countries in Asia and Latin America. AID should
continue to find ways to support individual scientists
in developing countries who have shown real
understanding of the problems of sustainability.
We would not presume to be much more specific about the
agenda which the various research institutions we have mentioned here should follow. This agenda must vary widely from institution to institution and we don't pretend to know enough to
make firm recommendations. However, there are common characteristics that we believe research projects should share: solid grounding in farms systems analysis; a broad interdisciplinary approach; and continuing participation in refocusing and adaptation by farmers themselves.

A few research topics which we believe should continue to receive special attention are:
- Designing 'no-till" systems which don't depend on
chemical herbicides.
- Further research and dissemination of successful
tropical agroforestry models, and finding and
developing fast growing woody perennials adapted to the
various ecological situations in which AID works.
- Finding low-cost and widely available ways to provide
basic soil nutrients, particularly phosphorous, to poor
farmers in order to avoid the need to import costly
- Testing of systems of resource-conserving range
management which will be acceptable to migratory
Confecting systems of low-cost integrated pest
management whichdo not rely on harmful or costly
chemical pesticides.
- Discovering low-cost ways of measuring and monitoring
water output from shallow aquifers.
Developing additional and lower cost sources of farm
Comparisons of total performance of a large number of
low-cost, projects (in AID and elsewhere) with that of
conventional high-input agricultural systems.
We applaud AID for beginning to move away from
financing and then relying on classic Third World
extension services, patterned after US models, models which have progressively fallen into disuse here. We
encourage AID to move creatively farther and faster in
the direction of perfecting imaginative "bottom up"
modes of helping farmers to--adapt sustainable-----agricultural systems through utilizing local "NGO's"
as well as new uses of mass communications and-even
the formal educational systems at local levels.
Better utilization of local people who understand and
are dedicated to sustainability, people who have a
superior knowledge of local culture, appears to us to

be the best way to mobilize these absolutely vital
local human resources.
- We believe that almost invariably the best base for
success in reaching large numbers of poor people for
the adaptation and teaching of new techniques is
participation by farmers at all stages of research and planning and, of course, implementation by farmers and their chosen organizations. This must be accomplished at the village level and through choosing credible (to
farmers) organizing groups of manageable size. AID
should have more barefoot" seminars on agricultural
AID should expend a lot more effort in helping to
build local institutions. We recognize that the
selection of local cooperating organizations is messy,
highly pragmatic and almost site-specific. Usually, but not always, the best group to choose is the one
that is most widely supported by the farmers
themselves: farmers' cooperatives or unions; women's groups, particularly where women do most of the farm work; village councils; and religious groups. If no appropriate group exists, we believe that AID should
help create one.
But AID's working at the level of individual villages
won't solve the problem of spreading new techniques more broadly. At this stage, AID does not have the
money to tackle the problems of sustainability in all villages of a country or even a region. Nevertheless, even while working at the village level, AID must aim
to construct solutions which are of broad enough
applicability as to be widely acceptable in a large
number of villages and which need only small low-cost
local adaptation. For the moment, AID must
realistically look to cooperating organizations and
cooperating donors -- to shoulder the broader
dissemination of information on new and proven systems and techniques more broadly. This further means that
project methods and results must be recorded a way
which can be widely understood and then made widely
Typically, the most difficult problem in bringing
about proper acceptance of proven systems and
"technical packages" is finding more widely based
regional or national groups with the technical skills,
governmental receptivity, and financial strength to act as conveyor belts to reach new farmers' groups.
We believe AID is on the right track in turning for
this purpose toz

- US private voluntary agencies;
- "Indigenous" PVOs with national or regional reach;
- The US Peace Corps;
- The African Development Foundation- The World Food Program;
- Local university outreach programs;
- And, increasingly, local private sector channels, particularly in Asia.
AID should give greater emphasis to in-country training of people at every level involved in
agricultural sustainability: government officials, indigenous NGO and PVO leaders. They need to be helped to get information about the need for the ready availability of many sound soil, cropping, water management, energy techniques that protect fields, forests, rangelands, and aquatic systems.
Reliance on any single group or combination to do the whole job in any given country is impossible today. There are limits in the case of each of these categories of cooperants, usually self-imposed, on their capacity for expansion. We recommend that AID urgently make more systematic efforts to identify organizations, particularly local PVO's, which can effectively play this role. AID should help in training them technically and financially. We also urge that AID do even more to encourage reluctant developing country governments to look with a kindlier eye on the use of such groups, particularly local groups. Finally, we urge that AID identify and seek to remedy constraints that make it difficult for PVOs to use their funds efficiently; that it make a careful survey of success stories in the use of PVOs in achieving sustainability; and that it examine how it can help give the better indigenous PVOs the financial stability which will attract and keep good people.
As a very central point, given the fact that AID will be depending increasingly on PVOs, American and indigenous, we believe that AID should reassess and rationalize its methods of working with them. Every
effort should be made to be more flexible and encouraging. For example, we believe that matching fund requirements should be relaxed for AID-PVO cooperative activities involving agriculture and forestry just as it has done with child survival programs.

We strongly support the following programs:
O Land reform or, where communal property is firmly rooted,
more security for farmers in their tenure: the purpose of such measures is to give farmers a sense of ownership and
responsibility for the land they till so that they will,
among other things, not disincentives against protecting the
natural resource base of their lands.
o AID programs proposing and supporting, through structural
adjustment and reforms or otherwise, a whole series of
changes in governmental institutions and policies which
could help or hurt the cause of sustainability (agricultural
pricing, agriculture taxes, abolition or limiting of cheap
food imports, changes in government agricultural
institutions, and abolition of subsidies which will
discourage farm populations).
o Population Programs: including demographic studies to
determine the impact of population trends on agricultural J
sustainability. Reducing overly rapid population growth is
one of the keys of the transition to sustainable agriculture in many parts of the world. Even in areas where the lack of population could be a constraint on agriculture, the ability
of farm families to decide on family size will be a factor
for stability.
o Health: particularly infant health.
o Education
o Women in development programs.
o Non-farm rural employment and other programs which imprQve
the quality of rural life.
Experienced AID observers all agree on one thing: the
surest way to ensure that AID moves more rapidly toward effective
incorporation of the principle and practices of sustainable
agriculture into all its agricultural and related programs is to
institutionalize the principle by bringing it into every aspect
of AID activities: personnel recruitment and assignment;
training; promotion; policy formulation; planning at individual
missions, as well as central planning and budgeting; planning and
coordination with other donors; and, very important, policy and
technical discussions with developing countries.

As with any key program, AID's program for agricultural
sustainability must receive increased resources, money as well as personnel, if it is to succeed. A major effort should be made by AID, particularly in these times of budgetary stringency, to convince Congress of the priority and importance of U.S. support of agricultural sustainability, interA1JA as a key element in the protection of our own national interests. Private American organizations, like those which participated in the formulation of these views, should be asked to lend important support with Congress. AID should particularly examine one instrument for giving greater support to this effort, Public Law 480. We believe that greater use can be made of funds so generated for support of agricultural sustainability without creating dependence on imported food stocks.
AID must also take the lead in convincing other development agencies to support sustainable agriculture. AID has a comparative advantage over other donors in helping developing countries make the shift toward agricultural sustainability. In addition to sharing in the U.S. national advantages in adapting agricultural advances to many situations, it has other strengthsthe presence of AID field staff which can monitor and encourage promising initiatives (although field staffs are being cut too rapidly under present budgetary pressures); ability to mobilize grant funds and PL-480 counterpart funds to help meet local costs; a dynamic relation with U.S. PVOs which allows creative spread of sustainable agricultural concepts; the willingness to help build cooperating nongovernmental organizations in developing countries; and the ability to work closely with experienced people in American academic and scientific institutions in networking with developing countries.
AID also has some distinct disadvantages: growing budgetary limitations; personnel ceilings; and, perhaps as serious as any, Congressional restrictions on AID's activities.
Most of AID's younger generation of agricultural experts -now in the majority -- are firm believers, at least in theory, in the centrality of sustainability to agricultural development. The trouble is that AID does not yet have enough field-based agricultural and especially natural resource experts: to plan the kinds of projects we have described; to inform and motivate developing country counterparts; to keep close touch with farmers and their organizations; to monitor U.S. and other agricultural and natural resource projects and programs; and to seek, evaluate and disseminate the successes and failures of sustainable agriculture. Unfortunately, the number of experienced in-house
agriculturalists is shrinking, and AID never had a sufficient number of natural resource specialists. Without trying to quantify exactly how many such experts should be stationed in

Washington or in a given country, we believe that every mission where there is a serious agricultural-natural resources program (and with limited exceptions that should include all AID missions) should have one or more resident experts. AID should make special efforts to recruit more of the kind of people who understand the participatory approach such as ex-Peace Corps people. AID should also encourage American PVOs to help in training local experts skilled in the agronomic and human skills of sustainability.
We strongly urge that Mission Directors and, in fact, all AID personnel be judged for promotion in part on their ability to support sustainable agricultural projects. This means, unlike what is the case too often today, seeing such programs through to a successful conclusion, not just designing such projects and getting them underway.
We recognize that AID should and will have to rely
heavily and increasingly on outside experts from the U.S., other developed countries and from the Third World, from a few excellent agricultural development institutions, from freestanding groups of consultants, and from the academic community to carry out its agricultural programs. We urge discriminating emphasis in selection of these experts on experience and performance and not on their institutional affiliation as such. We also urge a careful review of the criteria for selecting MIQCs" to ensure that agriculture consultants so chosen are truly qualified in the principles of sustainability.
We urge AID to sponsor a growing number of. regional workshops on sustainability in agriculture, some quite specifically on sustainability for farmers on fragile lands. AID should also hold or sponsor more interregional workshops in Washington. These workshops should include people from all regions, all bureaus, Congressional aides, PVO and NGO representatives, and outside experts.
AID Policy and Planning: In Washington and in the Field
We believe that AID has made'a good start in formulating
policy on agricultural sustainability. Nevertheless, we believe that AID's working consensus .has far outrun written down policy in both detail and creativity, particularly in the good agricultural and natural resource policy statements of some regions. As already indicated, we believe that AID's new agricultural focus statement should be expanded. We urge AID within the coming months to make new efforts to formulate global and regional agriculture policies which will, to a greater extent, reflect the centrality of sustainability. This theme should also be reflected much more specifically in Congressional presentations.

We urge that more of AID's centrally funded agriculture and natural resource projects be focused specifically on sustainability and be mutually better integrated.
With growing decentralization, a major emphasis on
agricultural sustainability in the planning and allocation of funds by AID missions is essential to institutionalizing sustainability. Some missions--reflecting the experiences and predilections of top staff and often the views of host-country government--put more emphasis on sustainability than others. We urge AID in its screening of mission Action Plans to insist that missions include proposals for projects and project elements that address agricultural sustainability. Furthermore, we believe every mission should have at least one ongoing program which addresses at the national level one or more of the more serious natural resource conservation problems related to agricultural
and rural development. Individual Country Development Strategy Statements as well as the guidance for preparing those statements should also be reviewed to ensure they have taken agricultural sustainability and particularly its natural resource components fully into account in their longer range planning.
We urge AID to push on toward completion of Phase II of the Congressional mandated program for financing the writing of environmental profiles for developing countries. These profiles, appropriately done largely by experts from the country being
studied, should be required to put new emphasis on natural resource problems associated with agricultural sustainability. The same should be the case with AID-financed National Conservation Strategies.
AID Leadership on Agricultural Sustainability with Developing
Countries and Other Donors
Outside but more often within the context of their projects, AID, Embassy, and other U.S. officials have a continuing exchange of ideas and proposals on agriculture and related subjects with officials of the country to which they are tied. These discussions range from macro-economic aspects of economic policy and reforms--in or out of fundamental discussions on structural adjustment or reforms--to quite technical discussions among experts. Sustainability in all its aspects should be central to those discussions. We also urge that AID generate good discussions of the demographics of agriculture and its relation to population programs.
We recognize the difficulty of persuading other givers of economic assistance fundamentally to change the course of their development assistance programs and to undertake programs and projects to which they are not politically, institutionally, or
intellectually committed. However, agricultural sustainability

and natural resource protection are themes whose time has come, and other donors--and some of the more advanced developing countries like India, China, and Thailand--have very important contributions to make. We urge that agricultural sustainability--and not just agriculture--should be the theme of regular exchanges of information and views with and among donors and of attempts to avoid unproductive donor duplication and rivalries, both in-country and on a more global or regional basis. AID should also, in this period of budget austerity, interest other donors in financing projects for the spreading of sustainability solutions, once such systems have been proven in various environments. This should be particularly so when AID doesn't have the money to do the job itself. AID should urge the OECD's Development Assistance Committee (DAC) to hold a special session dedicated to sustainability. AID, in reviewing World Bank and other regional bank projects, should give special emphasis to those institutions' performance on sustainability.
Finally, we urge that AID produce an annual state-of-the-art report on its and other's experiences in finding successful techniques and ways to get them adapted and adopted. PVO experience in this regard will be particularly important.

Consultative Group on International.Agricultural Research
Mailing Address: 1818 H Street, N.W., Washington, D.C. 20433, U.S.A.
Office Location: 801 19th Street, N.W. Telephone (Area Code 202) 334-8021 Cable Address-INTBAFRAD
From: The Secretariat 20 October, 1987
International Centers Week October 26-30, 1987
Washington, D.C.
The attached draft paper "Sustainable Agricultural
Production: Implications for International Agricultural
Research" (AGR/TAC:IAR/87/22) is transmitted by TAC for the information of the Group. It will be the basis for
discussion at the Workshop to be held in Rome, 12-14
January 1988. Group members are invited to send written comments to the TAC Secretariat for use in preparing the
final document.

Draft Report of TAC Continuing Sub-Committee II
October 1987

AVRDC Asian Vegetable Research and Development Center
CASAFA Commission on the Application of Science to Agriculture,
Forestry and Aquaculture
CGIAR Consultative Group on International Agricultural Research CIAT Centro Internacional de Agricultura Tropical
CIMMYT Centro Internacional de Majoramiento de Maiz y Trigo CIP Centro Internacional de la Papa
FAO Food and Agriculture Organization"of thE United Nations
GDP Gross domestic product
IARC International Agricultural Research Center
IBN International Biosciences Network
IBPGR International Board for Plant Genetic Resources IBSRAM International Board for Soil Research and Management ICARDA International Center for Agricultural Research in the Dry Areas ICRAF International Council for Research in Agroforestry
ICIPE International Center for Insect Physiology and Ecology ICRISAT International Crops Research Institute for the Semi-Arid Tropics ICSU International Council of Scientific Unions
IFDC International Fertilizer Development Center
IFPRI International Food Policy Research Institute IIMI International Irrigation Management Institute
IITA Internagonal Institute of Tropical Agriculture
ILCA International Livestock Center for Africa
ILRAD International Laboratory for Research on Animal Diseases IPM Integrated pest management
IRRI International Rice Research Institute
ISNAR International Service for National Agricultural Research IUBS International Biosciences Netwdrk
LDC Less developed country
OFR On-farm research
TAC Technical Advisory Committee to the CGIAR
UNDP United Nations Development Program
UNEP United Nations Environmental Program
UNESCO United Nations Educational, Scientific and Cultural Organization WARDA West Africa Rice Development Association

1.1. Introduction I
1.2. The CGIAR's Concern with Sustainability 1
1.3. Concepts and Goals of Sustainability 2
i.4. Environmental Degradation and Sustainability 4
1.5. Who Wins? Who Loses? Who Pays? 5
1.6. Short-term versus Long-term Considerations 6
1.7. Favourable versus Less Favourable Environments 7
1.8. "Closed" versus "Open" Agricultural Systems 7
1.9. Forest Resources and Rangeland 8
1.10. Desertification 9
1.11. Trends in Agricultural Production 10
2.1. Limits to Production 13
2.2. Rising Demand 13
2.3. Problems of Meeting the Demand 14
2.3.1. Limited Availability of Land 14
2.3.2.. Difficulties in Maintaining the Rate of
Increase in Yields 15
2.4. Low Priority Accorded to Agriculture 16
2.5. Economic Policies Unfavourable to Agriculture 18
2.6. Inadequate Infrastructure and Markets 19
2.7. Inadequate Inputs and Credit 20
2.8. Weak Institutions for Research, Extension and Education 21
2.9. Problems Related to Land Tenure 24
2.10. Inadequate or Inappropriate Laws that Affect Sustainability 25

3.1. Physical Factors 27
3.1.1. Soil Resources 27
(i) Soil Erosion 27
(ii) Soil Fertility 30
3.1.2. Water Resources 31
(i) Irrigation 32
S(ii) Rainfall 32
3.1.3. Toxic Chemicals 33
3.1.4. Changes in Climate 35
3.2. Biological Factors 37
3.2.1. Crop Pests 37
3.2.2. Diseases and Parasites of Animals 40
3.2.3. Genetic Resources 40
4.1. CGIAR Centers 41
4.1.1. Centers Working on Crop and Livestock Productivity 41
(i) Germplasm 41
(ii) Resource Management 41
(iii) Agro-ecological Characterization 42
4.1.2. Centers with Other Mandates 43
4.2. Non-CGIAR Centers 43
5.1. Suggestions from the IARCs 46
5.2. Approaches of Other Organizations to Sustainability Concerns 48
5.2.1. FAO 48
5.2.2. UNEP 48
5.2.3. World Commission on Environment and Development (WCED) 49 5.2.4. Office of Technology.Assessment, U.S. Congress 50
6.1. Responses to the Challenge 51
6.2. RePsearch with a Sustainability Perspective 52

6.3. Short-term versus Long-term Objectives 54
6.4. Low-input Agriculture 55
6.5. High-input Agriculture 57
6.6.. Sustainability and Equity 58
6.7. Improved Production Systems, including Agroforestry 59
6.8. Balance in Research: Productivity versus Resource Management 61 6.9. Techniques in Biotechnology 63
6.10. Policy Research 64
6.11. Relations with National Agricultural Research Systems 65
6.12. Training Programs 65
6.13. Collaboration with Institutions outside the CGIAR System 66
6.14. Reseach Needs and Resource Implications 66
6.15. Conclusions 68
ANNEX 1 Analysis of the Centers Responses to the CSC II
Request for Information on Sustainabitity

, ix
From its inception, a major goal of the CGIAR 1/ has been to increase food production in developing countries. Much of the work it supports has also been concerned with-sustaining production for the needs of future generations.
In:its study of CGIAR prorities-and future-strategies, TAC
recommended that the word "sustainable" be included in the System's goal statement and that greater emphasis be placed on sustainable production
systems in future work of the Centers. In this paper, TAC reviews the
circumstances threatening sustainability, analyzes areas where international research could contribute more effectively to the
development of sustainable agricultural production, and makes
recommendations for the future work of the Centers.
TAC's Concept of Sustainability
A dictionary definition of sustainability refers to "keeping an
effort going continuously, the ability to last out and keep from
falling". Such a definition would suggest that agricultural systems
would besustainable if production could be maintained at current
levels. This would be a static concept of sustainability. But
sustainability should be treated as a dynamic concept, allowing for the
changing needs of a steadily increasing global population. In the static sense, many traditional agricultural production systems were sustainable for centuries in terms of their ability to maintain a
continuing, stable level of production. However, the needs and
increasing aspirations of expanding numbers of people have forced
changes in production practices that have imposed excessive demands on
the natural resource base.
1/ A List of Acronyms is given at the front of this document.

Within this context, sustainable agriculture should involve the successful management of resources for agriculture to satisfy changing human needs while maintaining or enhancing the natural resource base and avoiding environmental degradation.
Trends in Agricultural Production
Characterized in this way, sustainability.should be considered in the light of past and current trends in agricultural production.
Aggregate statistics look very positive in terms of food
production in developing countries in the 30-year period from 1950Qto 1980. During this period, food production in the Third World experienced a compound rate of growth of 3% annually. Per capita food production in the developing world also improved, with a compound rate of growth of 0.6%, even though populations grew rapidly during the same period.
Despite this remarkable progress in expanding food production, the needs for further improvement continue to mount. Food deficits remain critical in Africa, where per capita food production has dropped by almost 20% in the last quarter-century. Furthermore, despite the overall -increase in per capita food production in the developing world, it is estimated that half the population cannot afford a diet that meets the minimum energy needs for a healthy, active life.
Circumstances that Limit the Achievement of Sustainability
Difficulties in Maintaining Progress in Food Production
There is much evidence to suggest that in the developing
countries it will be difficult, but by no means impossible, to maintain into the forseeable future the rate of progress in food production

,, ,xi
realized over the past 20-30 years. For example, it is doubtful if
those regions that have benefitted from the green revolution in rice and
wheat production can continue to enjoy the same rates of gain in
productivity that have occurred in recent decades. Furthermore, there
will be great difficulties in extending the green revolution in rice and
wheat to the other half of rice and wheat producers in the developing
world whom it has not yet reached.
But-aside from the difficulty in gustaianng the rates of gain
in food production through the green revolution, there are many
circumstances that will make the realization of sustainable production
systems extremely difficult, unless remedies can be found and
Population Growth
On a global basis, agriculture must produce enough to feed some
80-100 million additio6al people each year. The enormity of this
problem is highlighted by the fact that about 90% of this increase in
population is occurring in the developing world.
Such growth in population poses one of the greatest threats to
the achievement of sustainable production systems. Expansion in the
numbers of people increases the demand for more cropland while,
simultaneously, expanding the need to take land out of production to
accommodate other requirements. Furthermore, expanded food production
in response to rising demand increases the pressures on those natural
resources that are vital to sustain production often with serious
consequences for the environment.
Limited Opportunities for Expanding Cultivated Areas
There will be limited opportunities to expand the global base
of productive agricultural land. New areas brought into cultivation will do little more than compensate for the loss of agricultural land

diverted to other uses or otherwise lost through various processes of degradation. It is apparent, therefore, that growing demands for agricultural products must be met primarily by intensification of.
production on existing arable land, rather than by bringing new areas
into cultivation.
Unfavourable Political, Economic and Policy Environments
Political instability in some developing countries has been a major deterrant to sustained agricultural development, which has been further hampered by the low priority accorded to agriculture by many
national and local governments. Such low priority is reflected in many
ways, including a low level of investment in the development of the
agricultural sector as well as in policies for trade, taxation and
pricing. Often, the terms of trade are skewed against the agricultural
sector, leading to artificially low internal prices for agricultural
commodities, which favour the urban consumer at the expense of the ..farmer. In many countries, agriculture is not provided with the
financial-and institutional support its central role in the economy
warrants. Moreover, such policies act as deterrants to the achievement
of sustainable agricultural production because they do not provide the necessary incentives to producers to invest in sound farming practices
that minimize degradation of the environment.
In many developing countries, weak infrastructure is a major
constraint to the delivery of inputs and transport of farm commodities to market. Extending infrastructure helps to remove these constraints
and allows further-intensification of production in favourable areas,
which helps to reduce the pressure for increased production in the more
fragile environments.
In many circumstances, the achievement of sustainability will
require the use of purchased inputs such as seed, fertilizers,
pesticides, implements and machinery.. In addition to the limitations to their availability imposed by poor infrastructure, their availability is
also limited by high prices and the lack of credit to purchase them.

The development of strong, viable and effective agricultural research, extension and education programs is vital to the achievement of sustainability. Despite the high financial returns to investment in such programs revealed by many studies, there is widespread evidence that they are underfunded in most developing countries.
Systems of land tenure iiay also impose limitations on
agricultural development and the achievement of sustainability, by acting as disincentives to producers, for example,: to conserve natural resources and invest in the future productivity of the land.
Physical and Biological Factors Affecting Sustainability
Physical Factors
Soils. No single resource is more important to the achievement of a sustainable agriculture than the soil which contains nutrients and stores the water for -plant growth. Deficiencies or excesses of either can seriously limit productivity. Their availability in appropriate amounts is heavily dependent on the manner in which the soil is managed.
Moreover, in many parts of the world, soil erosion has
increased to the point-where losses exceed the formation of new soils through weathering. When this occurs, the soil is, in effect' being mined, converting renewable resources to non-renewable ones. When topsoil is lost through erosion, there is a loss of fertility and a deterioration of physical properties, resulting in a decline in productivity.
Population pressures are contributing to other difficulties in maintaining soil productivity. As fuelwood supplies are diminished, expanding populations in many areas have become increasingly dependent upon crop residues and animal manures for fuel, thereby reducing their use in replenishing nutrients and organic matter.

Water. Agriculture is the principal user of global water
supplies. From historic times, irrigation has been used to help farmers to secure a reliable and timely supply of water for their crops. During the 1950s and 1960s, irrigated areas expanded at the rate of about 4% annually. By the early 1980s, the rate of growth had declined to less than 1%.
Non-sustainable use of water is occurring in a number of agriculturar-areas throughout the world, involvin both the use of fossil water as well as the overdrafting of rechargeable aquifers.
Irrigation water is often used inefficiently, with much more water transported and applied than crops require. Furthermore, poor irrigation practices result in severe problems of land degradation through water-logging, salinization or both.
For those vast areas in developing countries that depend on rainfall for their agriculture, efficiency in its use is just as important as for irrigation water. Inappropriate soil and water management under rainfed agriculture is one of the primary causes of land degradation.
Toxic Chemicals. Human activities are responsible for releasing chemicals into the environment that may have serious deleterious effects on plants and animals. For example, certain industrial processes, along with the combustion of fossil fuels result in the release of large quantities of sulphur and nitrogen oxides into the atmosphere. These gases combine with moisture and come down as acid-contaminated rainfall, which has the potential to damage both terrestrial and aquatic organisms. Furthermore, acid rain may contribute to the acidification of soils which, in turn, may adversely affect productivity.
Industrial activities involving high-temperature processes have also resulted in pollution of the atmosphere through the release of a

number of metals, which are potentially toxic to plant and animal life, if allowed to accumulate in soils or water. In addition, chemicals with beneficial uses, such as fertilizers and pesticides, can become harmful if used inappropriately. The ability to benefit fully from such chemicals in the future may well depend on refinements in their use.
Climatic Change. There is considerable evidence pointing to a significant warming trend in world climate. Such a trend is a result of a build-up of carbon dioxide and other gases in the:atmosphere, which could have significant, longer-term effects on agricultural production systems and their sustainability.
For example, coastal lowlands could suffer increased risk of flooding caused by a greater melt of polar land ice. Possible effects of the warming trend on precipitation are more speculative, although various models suggest that some regions could become more arid and others more humid.
Biological Factors
If the food needs of rapidly increasing populations are to be met, both yields per unit area and per unit time must be substantially increased. Such intensified production favours the build-up of weeds, diseases, arthropods, rodents and birds (collectively referred to as pests") which, unless adequately controlled, seriously limit productivity. Although there is wide variation, it is estimated that pests contribute to field losses of some 35% of the potential production of major food crops, with the greatest losses occurring in the developing countries.
The long-term control of pests is also threatened by
break-downs in the effectiveness of pesticides and host-plant resistance through mutation of the pests. The research required to maintain the levels of control already achieved has an important part to play in the achievement of sustainability.

The sustainability of animal production is partly dependent on finding improved methods of controlling diseases and parasites, which cause high mortality among livestock in developing countries and seriously reduce productivity.
For the continued genetic improvement of both plants and animals, the conservation of genetic resources is of paramount importance and should continue to receive urgent attention from national .governments and relevant international organizations.
Contributions by International Institutes and Organizations to the Goal of Sustainability
TAC requested all the IARCs, as well as a wide range of other international organizations, to provide information on their current activities related to sustainability. They were also asked for their views on the needs for further research to assist in achieving sustainable agricultural production.
The information provided in response to these requests is
summarized in Annex 1 and briefly reviewed in Chapters 4 and 5 of this paper.
Recommended Strategies for Progress Towards Sustainability
If sustainability in agricultural production is to become a
reality, not only must the circumstances that threaten it be alleviated, but there must also be major efforts to increase productivity to meet growing needs. TAC views the challenge of finding timely and workable solutions to these problems as one which should receive the highest priority from all organizations that can make a contribution.
Many of the circumstances that limit the achievement of sustainability, however, cannot be solved by the CGIAR or through

agricultural research alone. It is national governments and their development services that must bear the brunt of the problems and on whose commitment progress in achieving sustainability depends. Nonetheless, continuing research is crucial for success and international research institutions, as well as national agricultural research systems, must continually examine their programs to give greater emphasis and visibility to those aspects that relate to sustainability.
Although the resouces of the CGIAR are small relative to total global expenditure on agricultural research by the public sector, the Centers can have a disproportionate impact through their ability to influence the nature of research at other institutions.* Furthermore, donors and other components of the CGIAR System can be helpful in focusing attention on sustainability, and encouraging governments and relevant institutions to accord it high priority.
A great deal of the work being undertaken by the Centers
already relates, to a greater or lesser extent, to problems affecting sustainability; The question that arises, therefore, is not so much whether the Centers are working .to make agriculture more sustainable, but whether they should be doing more, and whether there should be a different emphasis in the work.
Research with a Sustainability Perspective
TAC does not view research related to sustainability as a separate or discrete activity. Rather, concern for sustainability should be reflected in the way in which the research is approached. TAC therefore recommends that research at the Centers designed to generate new agricultural technology should be planned and conducted with asustainability perspective. TAC further suggests that in formulating or revising their strategic plans, Centers should include proposals for maintaining a sustainability perspective throughout their total program.

Short-term versus Long-term-Objectives
If the goal of sustainable agriculture is to meet the changing needs of people, research must clearly cater for both short-term and long-term needs. Nevertheless, a guiding principle for Centers must be that stability of the environment should never be consciously sacrificed for short-term gains. The aim should be to devise technologies that can meet short-term requirements while, at the same time, maintaining or enhancing the ability to meet long-term needs. .;."
Low-input Agriculture
TAC considers that research On low-input agriculture should
feature more strongly in Center programs. The aim should be to optimize productivity from the use of low levels of purchased inputs, consistent with the requirements of sustainability. The ultimate aim would be to promote a gradual evolution towards the use of higher levels of inputs, where needed, and the development of technologies that reduce the risk of uneconomic returns.
Technologies geared towards the more effective use of scarce resources can aggravate the problems of soil mining, unless nutrients are recycled as manure or plant residues, or replenished through the use of fertilizer. There are large differences, however, in the demands for nutrients made by different crops and differentproduction systems. Cassava production systems, for example, are sustainable with-very low inputs.
Centers should review the emphasis given to low-input farming in their research programs, and increase it where appropriate. They should also review their approaches to research on low-input farming to ensure that the sustainability perspective is adequately taken into account.

High-input Agriculture
Without the existence of high-input technologies, it would be impossible to meet the food demands of the increasing world population unless more, but less suitable, land were brought into cultivation, further raping the surface of the earth and'destroying-natural ecosystems in the process.
TAC-considers that, under appropriate conditions, the use of high levels of industrial inputs can make important contributions to sustainability and recommends that high-input technologies be included in research programs of the CGIAR Centers. TAC suggests, however, a selective approach to research related to sustainability in these production systems that restricts in-depth investigation to problems that are especially relevant to tropical and sub-tropical environments.
Sustainability and Equity
TAC reaffirms its earlier recommendation that the Centers give greater emphasis to the development.of technologies that are especially applicable in less-endowed regions. In addition, TAC stresses that assessment of these technologies with respect to sustainability requires a thorough analysis of evolving agricultural policies in the domains of their application.
Improved Production Systems, Including Agroforestry
There are dangers both in disregarding the principles of
traditional production systems and in assuming that, because they.are appropriate in some circumstances, they will remain appropriate in others.
TAC encourages Centers to continue to investigate aspects of more intensive production systems based on sound ecological principles and the conservation of resources. Whenever appropriate, this work should include aspects of agroforestry.

Balance in Research: Productivity versus Resource Management
Although productivity research includes many aspects of
resource management, the strengths of the various components of the
multidisciplinary approach must be kept under review to ensure an appropriate balance. Plant breeding, fo-- example, can continue to
contribute much to sustainability, but must not dominate Center programs.
to the extent that other approaches are neglected.
TAC recommends that Centers with commodity mandates review
carefully the relative emphasis being given to genetic improvement in
comparison with other aspects of productivity research. Sustainability
concerns may make it desirable, if not essential, for some Centers to
give increased attention to research on problems of resource management.
Techniques in Biotechnology
Centers must constantly assess, in relation to other needs and
opportunities, the potential contribution to their work of new
techniques emerging from advances in the biological sciences.
TAC considers that Centers involved in productivity research
should have the capability to monitor advances in biotechnology and,
when appropriate, develop the in-house capacity to use techniques that
would assist their programs in a cost-effective manner.
Policy Research
* .. Policy research has a particularly valuable role- to play in the
CGIAR System through its intimate interaction with technology research
and the changing comparative advantage that this research provides.
In its study of priorities and future strategies, TAC
recommended a significant increase in policy research. TAC reaffirms
this recommendation and urges that a concerted effort be made to provide
additional funding-for this purpose.

Relations with National Agricultural Research Systems
Centers could be very effective in encouraging national
agricultural research systems to give greater attention and priority to considerations of sustainability, as well as in helping to strengthen their capacity to do so.
TAC recommends that Centers give high priority to strengthening the capacity-of national agricultural research systems to incorporate a sustainability perspective into their research approach.
TAC further recommends that Centers give high priority to the incorporation of a sustainability perspective in training programs,
making adjustments, where necessary to meet, more effectively, the needs of national agricultural research systems in this respect.
Collaboration with Institutions Outside the.CGIAR System
TAC recommends that Centers continue to explore the potential for collaboration with other research institutions, including those in the private sector, particularly with a view to strengthening their research related to sustainability.
Research Needs and Resource Implications
While the Centers and the-national agricultural research
systems already make important contributions to the solution of problems related to sustainability, the total current effort is unlikely to be adequate.
In view of the serious problems limiting the achievement of sutainability, and the urgency for additional research to assist in their solution, TAC recommends a substantial increase in Center funding as a significant international contribution to meeting this need.

TAC considers that, because much of the required additional work relates to protection of the environment and the conservation of natural resources, it might well be possible to widen the avenuesfor donor support for this vital new thrust, compared with support for productivity research per se.
While many of the circumstances that limit sustainability
cannot be alleviated through work supported by the CGIAR, members of the Group can bring their influence to bear in.creating'a greater sense of urgency amongst all concerned. TAC further suggests that the issue of agricultural sustainability has major implications for the further development of the Third World and, indeed, for future global security.
TAC considers that the international donor community, as well as the governments of developing countries, have crucial roles to play in emphasizing the need to consider sustainability in allocating future resources and orientating future thrusts.
TAC has characterized sustainability in terms of the dynamics of population growth and resource conservation. The common challenge facing all concerned is to find ways of removing the impediments to sustainable agricultural production, whether the causes are technical, social, institutional, political, or some combination of all four.
A significant part of this challenge rests with the
International Agricultural Research Centers. Accepting it offers them opportunities for unprecedented contributions to the global community, as they help to find solutions to serious problems that significantly affect the future of mankind.

1.1. Introduction
Globally, much of the increase in agricultural production prior to World War II resulted from horizontal expansion by Vtinging more land into cultivation. The nineteenth century, for example, saw tremendous geographical expansion of agiiculture.7..Vast new areas were brought into cultivation in North and South America, Australia, Asia and Africa. Improved irrigation systems opened for cultivation large areas of the Indian sub-continent and other regions in Asia. Such geographical expansion made possible steady increases in production to meet the food needs of growing populations.
During the last half-century, however, there has been a sharp decline in the rate at which new land has been cultivated. In fact, cropland expanded at the rate of 1% per year during the 1950s but only at 0.3% during the 1970s (I). There is general agreement that most of the future gains in the production of food and other agricultural commodities must come from vertical expansion from increasing the productivity of existing agricultural areas.
While there have been remarkable gains in agricultural
productivity in some parts of the world during the last two to three decades, there is growing concern in the global community that such gains cannot be easily sustained. Consequently, it may become increasingly difficult for agriculture to meet the food needs of an expanding world population.
1.2. The CGIAR's Concern with Sustainability
From its inception, a major goal of the CGIAR has been to
increase food production in the developing countries. Much of the work it Supports has also been concerned with sustaining production for the needs of future generations. In its recent study of CGIAR Priorities

and Future Strategies (2), TAC recommended that the.word "sustainable" be included in the System's goal statement and that greater emphasis be placed on sustainable production in the future work of the Centers. A comprehensive study of the achievements and potential of the Centers (the "Impact Study") also recommended that sustainability, as well as stability of .production should be criteria of performance, especially in the hunid tropics.
At its meeting in May, 1986, the qGIAR.rook note of both sets of recommendations and expressed its own concerns about how the Centers could best respond to the challenges posed. The Group noted that in addressing problems related to sustainability, consideration should be given to the rehabilitation of areas that had lost their productive base. The Group particularly stressed the need for TAC to consider relationships between the work of the CGIAR Centers and those non-CGIAR Centers that deal with key interacting environmental factors, such as soil, water, trees and shrubs.
Accordingly, TAC was explicitly requested to examine how its recommendations relating to sustainability could be implemented and monitored. The first step in responding to the CGIAR's request is the preparation of this paper.
1.3. Concepts and Goals of Sustainability
The Oxford Advanced Learners' English Dictionary indicates that sustainability refers to "keeping an effort going continuously, the ability to last out and keep from falling". Such a definition would suggest that agricultural systems would be sustainable if production or output could be maintained at current levels. This would be a static concept. But sustainability should be treated as a dynamic concept, reflecting changing needs, especially those of a steadily increasing population.
In the static sense, many traditional agricultural production systems were sustainable for centuries i .srms of.,-their ability to

"* 3
maintain a continuing, stable level of production. However, the needs and increasing aspirations of expanding numbers of people have forced changes in land usage and Imposed excessive demands on natural resources, making the modified farming systems unsustainable. The goal of a sustainable agriculture should be to maintain a level of production necessary to meet the increasing needs of an expanding world population without degrading the natural resources on which agriculture depends.
Against this background, sustainable agriculture should involve the successful management of resources for agriculture to satisfy changing huan needs while maintaining or enhancing the natural resource base and avoiding environmental degradation.
In the above characterization
"successful" implies, that the production system is
economically viable and socially acceptable;
"management" includes policy decisions that might affect agriculture, made at all levels from national governments
down to individual producers;
"resources" includes inputs and manufactured items that come
from outside the agricultural sector (i.e. agricultural
chemicals, machinery, etc.);
"satisfy ... human needs" suggests that production and
prices should be at levels to allow low income groups
sufficient access to food;
"changing needs" suggests a continuing evolution without
specifying any particular time horizon;
"maintaining" suggests that changes in the natural resource
base should never threaten the capacity to meet changing.
With this characterization of sustainability, the term
"resource management" takes on a wider meaning. Consequently, research on resource management would include research on purchased inputs such as fertilizers and pesticides. Use of "resource management" in this sense would obviate the need for terms such as "factor research", which is ambiguous and even misleading in relation to the use of the term "factor" in other senses, such as in economics. TAC therefore suggests that the term "factor-research" be dropped from the-CGIA. -dabulary and

that the expression "research on resource management" be substituted. Where greater specificity is required, "resources" could be subdivided into "industrial", "cultural" and "natural". TAC proposes this adjustment in terminology to facilitate future discussions.
It is important that sustainability should be clearly
distinguished from stability. Stability implies limited change in the general sense, as in environmental stability. It is also used to mean limited fluctuation, as in yield stability. Unstable yields caused by environmental fluctuations can, however, be entirely sustainable.
Sustainability should also be distinguished from productivity. While an adequate level of productivity is a necessary condition for sustainability, it is not by itself sufficient to achieve sustainability.
The term "sustainability", as used in this paper, applies equally to all production systems that are dependent on the natural environment, including forestry, fisheries and aquaculture. While recognizing this broader context of sustiinability, however, the present paper is concerned mainly with agriculture, i.e. crop husbandry, animal husbandry and horticulture.
1.4. Environmental Degradation and Sustainability
The increasing global interest in issues related to
agricultural sustainability grows out of concerns that many factors are contributing to the degradation of the natural resources on which future agricultural production depends. Many have suggested that modification of traditional agricultural systems to meet growing needs often has undesirable environmental or ecological consequences. More intensive agricultural systems that are ill-conceived or poorly managed may damage the environment in various ways such as through increased soil erosion, greater salinity, waterlogging and the contamination of aquifers.
V. A

Indeed, much current writing on technology seems to imply that chemical and mechanical technology is bad, that biological technology is good, and small is always to be preferred. But agricultural technologies are seldom inherently good, bad or appropriate. Their appropriateness is determined by the manner in which they are applied which, in-turn, depends on the agro-ecological environment and changing socio-economic circumstances.
A ma-jor goal of efforts to achieve.sustainability should,
therefore, be to devise production systems that can meet the growing needs of society while avoiding. further degradation of the environment. Indeed, if long-term sustainability goals are to be achieved, production systems must not only avoid further degradation of the natural environment on which they depend, they must also contribute to restoring
and enhancing its productive capacity.
1.5. Who Wins? Who Loses? Who Pays?
While sustainability is a global problem, the environmental
degradation affecting sustainability is often the result of action taken locally. An individual farmer may follow practices that are beneficial to himself but harmful to others. Furthermore, disregard for the environment in one locality, through deforestation or overgrazing, may contribute to problems elsewhere through siltation of reservoirs and downstream flooding. Acid rain, resulting from environmental pollution by industry, may have sustainability consequences to others completely removed from the source of pollution. Indeed, action taken in one. country may adversely affect the natural resources of another.
A related question is who should pay for the effort to avoid or correct problems affecting sustainability? Many studies suggest that the immediate returns from various soil conservation practices, for example, may not be sufficient to pay for their cost. Should farmers expect to pay the total cost of such measures? Or should the costs be borne by society as a whole as an investment in the future? In many countries, governments are sharing in the costs of-conservation measures

undertaken by farmers, a practice that may have to be intensified and more widely applied in future.
1.6. Short-term versus Long-term Considerations
The foregoing discussion touches on another issue the choice of conserving a resource for future use versus its exploitation to meet current needs. Farmers and communities respond ta this dilemma according
to their perceived needs and circumstances. Some condone the
exploitation of fossil reserves of fuels and water, for example, on the
grounds that science will keep ahead of needs and provide alternative
technologies. Others.justify the exploitation of natural resources, on
the grounds of dire.necessity. The problems are complex'and the
attitudes of farmers and governments are crucial.Subsistence farmers with very limited resources usually have a
short-term planning horizon. They often exploit soils, but degradation
has traditionally been prevented by practices, such as shifting
cultivation. As the cycle is shortened through population pressure,
total production in the region may increase for some time. However, the
farmer ultimately faces deteriorating soil fertility, lower production,
poorer nutrition and a potentially shorter life-span. Increased
exploitation, although undesirable, can be justified in the short-term
if it increases income to the extent that the farner obtains a large enough cash surplus to enable resource management, with a long-term
perspective, to become part of the regular farm practices. In general,
as incomes increase, longer-term benefits are accorded lower discount
rates and sustainability can feature more prominently in the
decision-making process.
Governments of developing countries may well be faced with
similar problems. Their priorities are often dominated by short-term
constraints relating to the budget or balance of payments. Adequate
provision for the future may be precluded by the urgency of short-term
needs, thus leaving farmers to fend for themselves in matters of .:. ustainability until greater national prosperity is-oachieved.

* 7
1.7. Favourable versus Less Favourable Environments
Because of the relative scarcity of additional arable land, the
growing need for food must be met largely by greater intensification of
production, which could be more easily achieved in some agro-systems
than in others. Indeed, given the necessary resources, intensification
of production in the more favourable environments could probably produce
ample food for global populations into the foreseeable future. The
problem of equitable distribution could not: solved, however,
and resettlement of people on the scale required is neither socially nor
politically acceptable.
Consequently, a major proportion of the increased demand for
food will have to be met where the demand exists, much of this being in
the less favourable environments. Unfortunately, many of these areas
are already suffering from the consequences of population pressure,
making the achievement of sustainability greater than for the favourable J environments, And posing greater challenges to all concerned.
1.8. "Closed" versus "Open" Agricultural Systems
As a means of characterizing the changes occuring in the
* transition from traditional to more intensive production systems,
comparisons might be made between "closed" agricultural systems that
require little external inputs and produce little more than subsistence
needs and "open" systems that require industrial inputs and meet the
-needs of growing urban populations. Closed systems have been
characterized by an average yield increase of less than 1% per year (3).
With an average population increase of 2.5%, the food deficit must be
met by expanding the cultivated area at a rate of up to 2% annually at
* the expense of fallow periods, pastures and virgin land. In contrast,
yield levels in an open production system are substantially higher and
the annual increase can be more than enough to meet growing needs of an
expanding population.
t: Theconsequences of closed systems are ecologically alarming in
that they can satisfy the increasing demand for food only by expanding

into previously uncultivated, areas and destroying the natural ecosystems in the process. Consequently, preservation of the environment is dependent on accelerating the transition to more open systems in which the increases in yield can accommodate the growing demands for food without continuously encroaching on to virgin land.
It follows that a major development goal should be to meet food demands by better and more intensive use of land that is already cultivated and to avoid the indiscriminate .and o.ten temporary clearing of natural ecosystems. Because of the alarming rate at which the landscape is being denuded, the fear is that the natural resource base may be destroyed before the economic base can be strengthened to the extent that it can support the research and development needed for sustainable production systems to become a reality.
1.9. Forest Resources and Rangeland
These fears are further reflected in the exploitation of
forests and rangelands. It has been estimated that the amount of wood harvested, per capita, worldwide reached a peak in 1964 and has fallen since that time by 11% (4). However, a report by FAO suggests that the prospects for the long-term supply of wood are quite different for the two types of consumption, industrial wood and firewood (5). The report suggests that world forests could meet the demands projected to 2000 for industrial wood on a sustainable basis, but the prospects for firewood are quite alarming.
An estimated 250 million people in the developing countries live in areas of fuelwood shortage. In 1979, the estimated fuelwood consumption was 1,300 million cubic metres, but fell short of the amount required by an estimated 100 million cubic metres. At present consumption levels, the total requirement will rise to 2,600 million cubic metres by 2000 but, because of shrinkage in the resource base, total production could only be in the region of 1,500 cubic metres, representing a shortfall of about 1,000 million cubic metres and an increased demand for other soume&of fuel (5).

* 9
While the shortage of fuelwood is, in itself,-. an extremely serious problem, its effects on the environment through the loss of trees and-shrubs is likely to prove devastating unless remedial measures can be instituted. The role of forest cover is especially important to sustainability in protecting the soil and preventing runoff and erosion, wh!,le conserving precious water for recharging aquifers.
Removal of ground cover by overgrazing can have some of the same effects-as deforestation, in terms of leaving:land vulnerable to erosion as well as causing deterioration in soil fertility. Especially in Africa, there is widespread evidence of deterioration of grasslands as livestock numbers expand nearly as fast as the human population. In 1950, the estimated livestock population was 295 million, compared with
219 millionpeople. By 1983, the figures had risen to 521 and 515 million, respectively (6).
An FAO report in 1984 indicated the extent of overgrazing in the eight nations of.southern Africa. It was suggested that in some
-countries, and in major areas of others, existing herds considerably exceeded carrying capacities. The report concluded that this had led to deterioration of the soil, thereby lowering the carrying capacity even more, and eventually causing erosion, in an accelerating cycle of
degradation (7).
1.10. Desertification
Desertification has been defined as "the diminution or
destruction. of the biological potential of the land, a process that can lead, ultimately to desert-like conditions" (8). Such a process has contributed to the widespread deterioration of ecosystems and has destroyed the biological potential for plant and animal production at a time when increasing productivity is needed.
The United Nations Conference on Desertification in 1977
concluded that desertification was threatening the productivity of as muchas one third of the world's land ttaiU!g"15Z of the global

population. The World Comaission on Environment and Developaent indicated that some 29% of the earth's land area suffers slight, moderate, or severe desertification, with an additional 6% classified as "extemely severely desertified" (9).
Some have described desertification as a product of climate and human activity (10), a process that in the Sahel, for example, has been accelerated alarmingly by changes in both. Others have distinguished between desert spread in the more arid areas and~desertification in more humid areas caused by soil degradation and erosion (11). Although in these more humid regions, the land is rendered barren, the term desertification is somewhat confusing in that it has an arid connotation. Nevertheless, in estimates of the extent of desertification,. both types are usually included.
Estimates of the area affected by desertification in Africa vary from 2% to 22%. FAO suggests that this variation is so large mainly because an oversimplified view is often taken of a process that is both complex and dynamic. Attempts to control desertification usually fail unlesss this complexity is recognized and both the social causes of desertification as well as its physical symptoms are addressed
(11). But until solutions are found and implemented, the rapidly growing areas subjected to desertification throughout the world will pose serious problems for sustaining agricultural production in the years ahead.
1.11. Trends in Agricultural Production
If the concepts defined in this paper are accepted, production trends and sustainability have to be considered together, for each has implications for the other.
Prior to the mid-twentieth century, each of the major
developing regions of the world Asia, Africa and Latin America was a net exporter of cereals. Even in the early 1950s, developing countries" were still net exporters of food. By 1980, however, A:.a= .Ar6iq and

Latin America were major importers of cereals, with only Argentina, Thailand and South Africa in these regions remaining as net exporters. Domestic production had not declined; it actually increased, and even per capita production grew in Asia and Latin America. Increased imports of cereals reflected demand not met by domestic production.
Such imports were made possible by growth in real incomes
resulting, in part, from increased agricultural exports. The recent worldwide recession and .the weak demand for. agricultural exports from developing countries, however, resulted in declines in both agricultural exports and imports, with the result that, in the mid-1980s, developing countries as a whole became net importers of agricultural commodities (1).
Aggregate statistics look very positive in terms of food
production in developing countries in the 30-year period from 1950 to 1980. In fact, during this period, total food production in the Third World experienced a compound rate of growth of 3% per annum. Despite rapid population growth, per capita food production also improved, with a compound rate of growth of 0.6% annually. Africa was the only major region that experienced a decline in per capita food production, with A compound rate of change of -0.6% per annum (I).
There are many indications that had it not been for the green revolution in Asia and Latin America during the past quarter of a century, these two regions would have experienced declines in per capita food production similar to those that occurred in Africa. The rapid spread of new technologies based on improved varieties of wheat and rice, initially bred at CIMMYT and IRRI, but often developed further by national programs, had enormous impact on food production in these regions and transformed the lives of hundreds of millions of people. India, for example, doubled its wheat production in a six-year period, an accomplishment unequalled by any other major country in history. Many other countries, including Mexico, Pakistan, Turkey, Indonesia and the Philippines also increased cereal production dramatically (12). By 1983, it is estimated that developing countries were planting half of their wheat areas and almost 60% of their rice areas to improved varieties of the two crops. Some 80% of the wheat areas in both Lati'zt.- .

America and India and 95% of the rice areas in China are now planted to
the improved varieties (13).
Nevertheless, there are strong indications that in Asia and
Latin America, a great effort will be required to maintain the rate of
increase in food "production that has been achieved over the last quarter of a century. With further growth in per capita income, it is likely to
be even more difficult to meet the increased demand projected for the
year 2000. In Africa, progress has been inhibita.-by a number 6f
factors, including the manner in which priorities for research and
development'have conventionally been set. For example, the three major
staples, roots, tubers and starchy bananas, have seldom been accorded
priorities commensurate with their importance in the diet. For the
future, much will depend on government policies and the extent to which farmers adopt the new technologies increasingly becoming available from
the Centers and national agricultural research systems.
Food deficits remain critical in Afra where per capita food
production has almost 20% in the last quarter of a century (11).
Furthermore, by the year 2,000, sub-Saharan Africa is projected to face
a shortfall of 20 million metric tons of basic food staples just to meet
the requirements of growth in population. If the 1966-77 trend in
growth of per capita income were to continue,.the net deficit in basic
food staples by the year 2,000 would be 36 million tons seven times
the 1977 food gap of the region (14).
Despite the overall increase in per capita food production in
-- the developing world,.it is estimated that about half the population
cannot afford a diet that meets the minimum energy needs for a healthy,
active life. Furthermore, there is little prospect of reducing this
proportion of the population by the year 2000 (15), by which time food output in the developing world is projected to fall short of demand by
about 80 million tons, if the 1966-77 trends in growth of per capita
income continue. This would represent a fourfold increase in the
estimated net deficit of staple foods in 1977 (14).

2.1. Limits to Production
A major concern threatening the achievement of sustainability is the capacity of individual communities to increase their agricultural productivity to meet increasing demands, without degrading the enviroument.'- Although large areas of the earth's: surface are unsuitable for agriculture, it is suggested that the amount of land potentially useable is about twice the area presently in use. Furthermore, there is evidence that agricultural production could.,, potentially, reach many times current levels if appropriate technologies were developed and applied (16). For the foreseeable future, therefore, the problem is not so much the ultimate global capacity for production, but whether the local rate of progress in production can keep up with increasing demand in a sustainable manner.
2.2. Rising Demand
Population growth, income growth and urbanization will be the key factors shaping food demand in the future. On a global basis, agriculture must produce enough to feed about 80 to 100 million additional people each year, representing an approximate 2% annual increase. The enormity of this problem is highlighted by the fact that about 90% of the increase in population is occurring in the developing part of the world, where food deficits are already critical in a number of regions.
Income growth accelerates the increase in demand caused by population growth. The World Bank projects annual rates of growth of per capita income of about 3%. With an average income elasticity for
food consumption of 0.5, the per capita demand for food is predicted to increase each year by another 1.5% Income growth is also a major force
shaping the composition of food demand. As incomes rise, consumers
-...shift from staple grains towards livestock product and more fruits and vegetables.

Urbanization is a third factor shaping food demand and is also one of the main reasons for substantial growth in consumption of wheat and rice in developing countries. Projections show that by the year 2000 about 40% of the total population in the developing regions as a whole will be living in towns and cities (17), and the agricultural population will be about 50% of the total.-'By the year 2025, the urban population will be about 60% of the total, and the agricultural population will further decrease to about 35%. Consequently, the remaining farming community will have to producemuch more than their own family needs. The example of the developed world shows that this can be done, but it requires considerable external industrial inputs.
Many of the circumstances that threaten sustainability of agriculture are caused and accentuated by the rapid increase in the total demand for food and by changes in its composition. In the long run, measures to enhance sustainability have little chance of succeeding unless this demand can be checked, but for the coming decades the die is cast and agriculture has to meet a formidable challenge.
2.3. Problems of Meeting the Demand
2.3.1. Limited Availability of Land
It is possible to clear more land for agriculture, but only at the expense of sacrificing natural ecosystems at the alarming rate that has characterized the recent past. In reality, it is likely that new areas brought under cultivation will do little more than compensate for the losses caused by various forms of degradation and by the spread of cities, villages and highways.
Moreover, population and land resources are unevenly
distributed. Within the African continent, North Africa has little unused land, while West and East Africa have limited reserves relative to population growth. Only Central Africa has large unused areas, but the population density is low,'the infrastructure is weak, the soils are fragie- and.the-snvironment such that both human beings and animals are prone to diaeas&erc

Much of Asia already has high ratios of population to arable land, with limited opportunities for expanding the area under cultivation. Temperate and tropical South America still have large areas that could be brought into more productive use, but many of them have major problems that must be solved before their potential can be realized in a sustainable way.
In view of all these circumstances, it is apparant that the relentlessly growing demand for agricultural products has to be met primarily by the more productive and sustainable use of existing arable land, rather than by bringing new land into cultivation.
2.3.2. Difficulties in Maintaining the Rate of Increase in Yields
Yields of crops and livestock products have increased rapidly during the last 40 years in the industrial part of the world and during. the last 20 years in parts of the developing world. Indeed, per capita production of many, but not all,' basic commodities increased during the third quarter of the century. Moreover, some of those that increased reached a peak. For example, the amount of wood harvested per capita peaked in 1964 and has since fallen by 11%, while the fish.catch-per capita has declined by 13% since 1970 (4). Similar trends can be seen in various other commodities and, in Africa, 40% of the population live in countries where grain yields are lower today than they were a generation ago (6).
Even with high inputs, crop yields are not consistently being maintained. In the Ching Mal Valley of northern Thailand, for example, paddy rice yields were increased to about 7 metric tons per ha, from about 4 metric tons, by changing from traditional to intensive practices. After ten years, however, yields had reverted to traditional levels in spite of continued high inputs of lime and fertilizer (18). Other examples include the rice/wheat rotations in Pakistan, where problems have been encountered in maintaining yield levels despite the rising use of inputs and growing managerial skills of the farmers (19). The problems are ndt zecessantly insuperable, but -they point to the need

for more penetrating research and add to the problems of achieving sustainability.
There will undoubtedly be difficulties in extending the green revolution in rice and wheat, for example, to less endowed regions. Farmers who have benefitted most have been those who have had the most ready access to irrigation, fertilizers, pesticides and markets. The remainder are primarily farmers on marginal, rainfed land, often with limited access to purchased.inputs.. Although there--is considerable potential for increasing yields in these areas, the benefits of new technologies will likely be realized gradually, rather than in the burst of adoption that characterized the green revolution (20).
Furthermore, large numbers of farmers in the developing world rely on crops other than rice and wheat, which have, up to now, attracted less research. Nevertheless,.prospects for increasing yields in a sustainable way are encouraging, provided improved technologies can be made readily available. Their development requires renewed research effort, in which the International Centers- could make a substantial contribution. Failure would force farmers 'to use agricultural production systems that might meet the increasing demand for the immediate future, but would threaten sustainability in the longer term.
2.4. Low Priority Accorded to Agriculture
In many developing countries, sustained agricultural
development is hampered by the low priority accorded to agriculture by national and local governments. Such low priority is reflected in many ways, including a low level of investment in the development of the agricultural sector as well as in policies for trade, taxation and pricing. which often tend to favour the urban consumer at the expense of the farmer. The result is that in many countries, agriculture is not provided with the financial and institutional support that its central role in the economy warrants.
This is especially true in A;-&tca,,.,-he~e many governments have allocated a relatively small proportion o! hi;'a- vailable resources to

* 17
agriculture. For example, in Botswana, 80% of the labour force depends primarily on agriculture, but the government allocates only 1-3% of its gross fixed investment to this sector. Kenya has 75% of its labour force in agriculture; yet the government invests only about 8% in the agricultural sector (21). Overall, expenditures by national governments for agricultural research in Africa decreased from $460 million to $285 million in the period from 1980 to 1982-83 (22). More recently, however, there have been indications of a greater determination to redress the balance, as exemplified by the.special-.program-for the rehabilitation of agriculture in Africa, adopted by the Organization of African Unity in 1985 (23).
The FAO review of Third World public expenditure on agriculture in 1978-82 found that the most troublesome budgetary problems for agricultural expenditure appeared to exist in the LDCs in Africa and in some Latin American countries (24). Over half of the 21 African states (but only one-third of the 36 non-African states) that were reviewed in this study had negative annual rates of change in planned agricultural expenditure during-that period.
It appears that an urban-biased development strategy is
responsible for the failure of many developing countries to supply agriculture with the financial and administrative resources, as well as the political support, that its central role in the economy would justify. Such a bias seems to represent a formidable obstacle to agricultural development (24).
The bias of development strategies can be illustrated by
contrasting agriculture's roles both as a source of government revenue and as a recipient of public expenditure. Taxes on exports of agricultural commodities have been the most Important component of agricultural taxation in many developing countries; in some they have been a major source of total government revenue. But taxation becomes a major constraint if it penalizes agriculture by reducing the resources available for investment in the sector, or acts as a major disincentive to farmers to increase their productivity or invest In the longer-term needs of sustainability. -

2.5. Economic Policies Unfavourable to Agriculture
Discontinuity in government administration, often associated with political instability, has been one of the causes of limited agricultural development in some developing countries (25). Not only does discontinuity cause inconsistencies in policies and their implementation, but there is a tendency for each new administration to dissociate itself from previous administrations and go through a time-consumihg learning process before it can famnulate and implement new policies (26).
Moreover, the low priority that many developing countries give to agriculture is often reflected in policies that-act as deterrents to the achievement of sustainable agricultural production. Many countries have pursued strategies in which the domestic terms of trade are sharply skewed against the rural sector. Overvalued currencies have contributed to this effect by making imports of food artificially cheap and reducing the'internal prices of agricultural export commodities, particularly raw materials. Such distorted terms of trade not only favour urban consumers but also have the effect of keeping wages down and increasing profits in the industrial sector, which, if reinvested, become the major source of capital formation (26). Attempts to initiate capital transfers from agriculture to industry ("syphoning off the agricultural surplus" or "mining agriculture") is a policy attractive for countries in the early stages of their'economic development, whose economies are still dominated by agriculture. All these circumstances act as serious disincentives to producers to invest in'sound farming practices that minimize degradation of the environment.
In contrast, the establishment of more reasonable price levels for food commodities, as a consequence of government policies, leads to increasing production and greater incentives for investment in sound agriculture. For example, the Ugandan government doubled food prices in 1983, resulting in a 400% increase in locally produced food (27). By contrast, Zambian and Ethiopian farmers in 1980 received approximately half the price for their cereals that they had received a decade earlier. Production of cereals in both countries declined significaftrly, = during this period. c

FAO studies indicate that price policies were significantly less favourable to farmers in Africa than in any other developing regions in the period from 1969/71 to 1981/83. Prices paid to African producers for export crops fell by 17% in real terms over the period, compared with rises of 3% in the Near East, 11% in Asia and 17% in Latin A-erica. It was concluded that the decline in prices was caused by the policies of African governments rather than by declining world market* prices (24).
Prices to producers have generally been more favourable for
cereals than for export crops, but the FAO study indicated that African farmers have still fared less well than their counterparts in other developing regions. During the period of this study, cereal prices to producers in Africa rose by 3% in real terms, but they rose in Asia by 18%, in the Near East by 23% and in Latin America by 30% (24).
2.6. Inadequate Infrastructure and Markets
The importance of infrastructure to agricultural development and hence to sustainability has been well documented (26). In many developing countries weak infrastructure including roads and rail systems is a major constraint to the delivery of inputs and the transport of farm commodities to market. Extending infrastructure helps to remove these constraints and allows further intensification of production in favourable areas, which helps to reduce pressure for increased production in the more fragile environments.
The problems of inadequate infrastructure are particularly
severe in Africa. For example, in 1982, there were only 206,177 km of roads in the 14 landlocked countries of Africa. (Less than 20,000 km were paved. One country, Zimbabwe, had some 40% of the paved roads.) Furthermore, the rail system in Africa was developed during the colonial period primarily as a means of transporting exports and imports between the coastal ports and inland areas. But Central Africa because of its vast distances from ports has no major links by rail in spite of its agricultural potential (28). Furthermore, because-of low population

densities in Central Africa, as well as in many other areas, the cost per capita of providing roads and other services is much greater than in most other regions of the world (29).
Improved access to channels of marketing and processing can act as a major production incentive, but marketing networks are inadequate in a large number of developing countries. Studies in Bangladesh have compared areas with different levels of infrastructure. Compared with the poorly developed areas, those with well-develDped infrastructure
showed significant increases in the level of adoption of new technology, in the proportion of land devoted to high-value crops, in employment,
and in prices of marketed produce as well as in wage rates. The studies also showed strikingly lower marketing costs and thus greater
competitive power on international markets in Asian countries compared with those in Africa. On average, the Asian farmers received 75-85% of the prices paid by the final users for agricultural commodities, whereas
African farmers received only 30-50%. Half of this difference was
attributed to better infrastructure in Asia (30). Lower marketing costs
as a result of better infrastructure could serve as a significant
incentive to farmers in their efforts to achieve sustainable
agricultural production.
2.7. Inadequate Inputs and Credit
In many circumstances, the achievement of sustainable
production will require increased use of purchased inputs such as seed, fertilizers, pesticides, implements and.equipnent. The elimination of
some of the major constraints to their availability at reasonable
prices, as well as the provision of credit for their purchase,must be
accorded high priority in development plans.
In some instances, declining export revenue and the high
priority that governments attach to other imports, including food,
result in inadequate importation of needed agricultural inputs.
Furthermore, because of poor systems of marketing and distribution in ..many developing countries, farmers have difficulty in obtaining inputs
-an a timely basis, if at all.

The ratio of prices of inputs to those received for farm
products often discourages a more intensive use of purchased inputs.
The reason for unfavourable price ratios may be high costs of production of inputs coming from outside agriculture, high costs of transportation,
or a deliberate policy to keep food prices down. In order to minimize
the effect of the resulting disincentive, some over ments have resorted
to subsidies. Such subsidies may not be necessary if producer prices
can be set sufficiently high to provide the necessary incentives. Where
this would itply intolerable hardship for poor ceheiners, consumer
subsidies, carefully targeted to low-income groups, may be an
alternative to subsidizing inputs.
Many farmers in developing countries do not-use adequate inputs
because they do not have the resources to purchase them. Institutions
to provide credit are often weak and ineffective or inaccessible. They
commonly favour export crops over staple food crops. Furthermore, small
re/source-poor farmers and especially women often have great
difficulty in securing credit, either because of their lack of
" collateral, or because of the cost to banks of administering credit
schemes for small producers of food-crops.
.2.8. Weak Institutions for Research, Extension and Education
Because the adoption of improved technology lies at the heart
of sustainability, research, extension and education must be given high priority. In research, for example, although considerable progress has
been made in some Asian and Latin American countries, the performance of
national-agricultural research systems in Africa in recent years has
generally been weak.
The reasons for the lack of success of research systems are
numerous and complex, but they usually include one or more of the
following (25).
- Political support for research has fluctuated widely
-.... ,resulting in inadequate funding.
. -.,. 'U

- The demand from producers for improved research is widely
dispersed among millions of unorganized and politically
impotent small farmers, and has not consistently been
- Research gains have not been'sustainable, not o.uny because of
inadequate funding but also because the funding has been
highly unstable, resulting in uncompleted and abandoned
research projects.
- Many national research institutions have lacked a critical
mass of experienced research workers.
-Many national research systems have lacked continuity owing
to the high rate of turnover of staff.
- Continuity has also been adversely affected by
discontinuities in research direction and management.
- The allocation system for research resources has not been
Despite widespread evidence of high returns to investment in agricultural research, it remains surprisingly underfunded in many developing countries. An FAO/UNDP study suggests that an allocation to research of between 10 and 20% of the total funds committed to agricultural development would be appropriate, or about 1% of agricultural GDP. Different authorities have suggested targets for research funding ranging from 0.5% to 2.0% of agricultural GDP. Even though several Asian countries have large and effective research systems, the minimum suggested level of 0.5% has not yet been reached (26).
Limitations to the achievement of sustainability imposed by
inadequate national research systems are often aggravated by weaknesses in extension services. Many extension services tend to be orientated more towards export crops than to food crops and they often fail to
reach small farmer!..- The*-.generally bypass or neglect women who,
............ ... . . .. . . .
. . . .~ . . . . . .

particularly in Africa, play key roles in the production, processing and marketing of food crops. In most countries there are weak linkages between research and extension which restrict the flow of information to and from the farmer. Such problems often grow out of the manner in which research and extension services are structured within the g6vernment service.
The development of capabilities in research and extension, as well as in the other activities needed for agricultural development, requires adequate educational and training institutions to meet the need for well-trained technical and professional personnel. It is estimated, for example, that nearly a quarter of the African countries will have a major shortage of professional manpower for the next fifteen years or longer and that more than one-third will have a major shortage of technical (intermediate) level manpower (11). Most developing countries have needs for more well-trained agricultural personnel than their institutions are currently providing.
A major problem in many, if not most, developing countries is that salaries in educational institutions often take up a high proportion of recurrent expenditure, leaving the staff with limited funds for operation. Many governments have also found it difficult to provide adequate funding for education and training institutions once the donor support ends (11). Similar considerations apply equally to research institutions.
In addition to the importance of developing trained technical personnel, there is considerable evidence that greater emphasis on general education of farm and rural people contributes significantly to the development process. World Bank studies have shown, for example,
that primary education yields significant returns to investment in view of the fact that farmers having several years of basic schooling become more effective users of technology than those without education. If wider awareness of the need to conserve resources is to become a reality, then the principles of sustainability must. be built into the curricula of both schools and universities.

2.9. Problems Related to.Land Tenure
A recent FAO publication (31) suggests that agricultural
development maybe affected by land tenure in various ways, some of which are described below.
There is a common belief that a system of communal land ownership tends to discourage the development of long-term, high-investment, crop operations. This may: well.,be. true in many instances. However, in Africa, there have been a number of situations where significant production of perennial crops, such as oil palm, was developed on land under customary or communal tenure, and the indigenous farmers were able to compete successfully with plantations.
There are cases, however, where communal tenurial rights may
impede long-term agricultural development. For example, in many African and Middle Eastern countries, communal rights permit the communal winter-grazing of cattle on land that is cultivated by individuals:Such overriding communal rights seem to inhibit longer-term advances in land improvement that might be achieved, for. example, through fencing, planting perennial crops, or the development of agroforestry.
Land tenure may have other significant effects upon the
degradation of natural resources and the sustainability of production systems. In some Latin American countries, for example, where much of the better land is held by large owners, many of the peasants have to cultivate less productive, more fragile land on the hillsides and mountains, with strong negative environmental consequences.
In the case of communal land, a conflict of individual or household interests against those of the community may arise. The individual household may perceive no short-term costs in increasing its herd on the communal grazing land. But if all households were to do the same, there could be overgrazing and long-term ecological damage as is already being experienced in many countries. In these and other ways, tenurial systems may pit individuals against each other for the use of scarce communal resources, thus accelerating the process of euitrormei.tal degradation.

Special recognition should be given to the-rights of women under various land-tenure systems. In many areas, particularly in
Africa, women are the primary producers of agricultural products.' Yet they often do not own the land they farm. This may serve as a disincentive to long-term improvements in the land and have unfavourabe effects on agricultural production. Furthermore, when women have no legal right to land, they cannot use it as collateral, with the adverse effects already discussed.
Contrary to the view often expressed, some forms of customary or communal tenure allow for inheritance and provide sufficient security to encourage investment by recognizing individual rights to improvement and to continued occupancy during usage. For example, about 45% of sub-Saharan land under irrigation is dependent on rights from customary
or communal tenure (31).
The traditional forms of customary-tenure, however, may no
longer be sufficient under many circumstances. Rapid growth of human and livestock populations has led to unsustainable pressures on the land. In some areas, holdings have become excessively fragmented and are now too small to provide adequate subsistence using traditional technologies. Furthermore, arable farmers have spread out on the land traditionally occupied by pastoralists, leading to conflicts between the two groups and to overgrazing on the remaining area, not under cultivation (11).
2.10. Inadequate or Inappropriate Laws that Affect Sustainability
A number of countries in the developing world have not
promulgated adequate laws and regulations to control the use of land or to protect forests and rangelands from indiscriminate exploitation. Neither is there adequate control of the use of water resources in catchments, rivers, lakes and aquifers nor legislation to facilitate schemes for land improvement or irrigation. There is a need for stronger international support of the efforts of national governments to address these matters,

Some countries have introduced legislation, which is adequate in principle, but they have been unable to enforce it in a satisfactory manner. Without appropriae public control or regulations, the chances of maintaining or achieving environmental stability and sustainable agriculture are considerably diminished.
. ......................

3.1. Physical Factors
3.1.1. Soil Resources
No single resource is more important to the achievement of a sustainable agriculture than the soil whict contains the nutrients and stores the water essential for plant growth. Deficiencies or excesses of either may seriously limit productivity. Soils may be too acid or too alkaline, they may suppress germination or root growth, they may have a low water-holding capacity and subject the plant to drought, or they may have poor infiltration leading to inadequate aeration and water-logging. Consequently, the manner in which soils are managed has tremendous impact on productivity and sustainability.
(i) Soil Erosion
Reference has already been made to the manner in which
population pressures are forcing farmers to cultivate increasingly marginal lands. In most developing countries, scarcity of arable land is forcing farmers to move up the hillsides, clearing the natural vegetation that protects the soil and opening land for cultivation that is far too steep to avoid runoff unless terraced or protected in some other way. In Malawi, for example, escarpment land that has a slope of more than 12% is being cultivated without adequate conservation practices, causing serious erosion and. the .flooding of fertile croplands below. Such erosion is threatening the future of one of the few countries in Africa that is successfully feeding itself (11).
Research on land planted to cassava in Nigeria demonstrated
that with a 15% slope there was an annual loss of 221 tons of topsoil/ha
- a rate that would erode away the entire topsoil in 10 years (32). Similar examples are legion and, in many parts of the world, soil erosion has increased to the point where losses exceed the formation of
* new soil through weathering. When this occurs, the' coil ia;. in effect, being mined, converting renewable resources to nonrenewa. Ocnei4.

Such soil erosion is truly a global blight, affecting developed. as well as developing countries. In the USA, for example, the loss of soil through erosion is estimated to exceed tolerable levels on some 44% of the cropland. This represents a loss of 1.7 billion tons annually, with over 90% coming from less than a quarter of the cropland. In India, a study has indicated that 60% of the cropland is eroding excessively, constituting a loss of 4.7 billion tons of topsoil annually (33).
In many tropical soils, most of the primary plant nutrients
available to arable crops are contained in the topsoil. When topsoil is lost through erosion, there is a consequent loss in fertility and a deterioration in the physical structure of the soil as more of the subsoil is exposed or brought to the surface through tillage. As a consequence, infiltration rates decline, runoff increases, crop growth decreases leaving the soil more exposed, and the rate of erosion increases. Unless this process is arrested, erosion can become so severe that the land has to be abandoned. Its reclamation is extremely difficult and prohibitively'costly for many developing countries. Few problems pose greater threats to sustainability of agricultural production systems than those of soil erosion.
Technological solutions to the problems of soil erosion have been known, to a greater or lesser extent, for centuries. Their thorough elucidation In the present century owes much to research and development on a massive scale in the USA during the 1930s. It was shown that erosion control involves correct land use over the whole of the catchment area, because bad. practices at the top increase the hazards of erosion lower down. Indeed, the effects of erosion extend far beyond the farm, as soil is carried into streams and canals, with harmful consequences to aquatic life as well as to irrigation and hydroelectric systems and even to navigable waterways (33).
The methods of control that were developed in the USA were based on systems of land management involving suitable cropping sequences, fertilizer application, contour cultivation, strip cropping and suitably designed systems of terraces and protected waterways. .*. o, .

recommendations formulated as a result of this work have made a major
contribution to soil conservation measures throughout the world,
although they need to be adapted to local circumstances.
The widespread adoption of these practices in the developing
countries is fraught with difficulty, however, even though they have
been tested and shown to succeed almost universally. There are
difficulties in organizing whole communities to do the work, as well as
in convincing them that it is worthwhile in' termi of the demonstrable
benefits, before serious erosion has occurred. Historically, the
problems.have been aggravated, especially in former colonial countries,
through the attempted enforcement of control measures by local
legislation. Some of these were impracticable under the prevailing
circumstances and led to disenchantment of the small farmer.
Nevertheless, there are many traditional farming systems,
including those based on agroforestry, that have built-in methods of
erosion control demonstrating that, in certain circumstances,
communities will respond to the challenge and institute collective
* control measures. Even in sub-Saharan Africa, where widespread
agriculture has evolved more recently than in other regions, there are
examples of the effective adoption of erosion control measures, both
past and present.
Perhaps the most striking example is that of .Tkara island
(Tanzania) in Lake Victoria. Prior to colonial times, the inhabitants
were prevented from migrating because of the hostility of the
inhabitants of the nearby mainland. Faced with a continually growing
* population, the islanders responded by intensifying their production
* systems. Among the many improved technologies they developed, were
several designed to prevent soil erosion. They constructed rock-faced
terraces on the hillsides and further controlled runoff by preparing the
land as a mosaic of small basins, comparable with tie-ridging.
Waterways were protected with bunds and grazing carefully controlled.
Later, however, as migration to the mainland became possible, the
communal effort was not maintained and deterioration set in, leading to
* soil degradation and the familiar declining spiral.

A more recent example of a community responding to the
challenge of preventing soil erosion is provided by the efforts of
village cooperatives in Burkina Faso. They have constructed
micro-catchment areas, consisting of mosaics of rectangular earth bunds
to concentrate rainfall around trees, planted to protect the upper slopes and eventually to supply firewood. In addition, areas of
crusted soil are broken up with hoes, rock barriers are built as well as
woven barriers of stalks to reduce runoff and trap silt. The effort
shows promise of restoring to production considaeahle areas of land that
had previously been abandoned (11).
(ii) Soil Fertility
Another major constraint to sustainability is a lack of certain
nutrients essential for plant growth. Whiie all essential elements are
normally found in soil, the amounts of certain nutrients may be
insufficient to achieve desired levels of production. Furthermore, harvested crops have the effect of mining the soil of iti nutrients,
unless they are continually supplemented through plant residues, manures
or fertilizers. In Africa, for example, FAO estimates that some ten
times as many nutrients are being removed in crops as are being put back
into the soil. Soil mining on such a scale represents a serious threat
to sustainability and if unchecked will eventually lead to significant
soil degradation.
Traditional systems of shifting cultivation in the tropics
depend on a natural nutrient cycle in which nutrients removed in crops
or leached into the deep subsoil during the arable phase are replenished"
in the topsoil during the resting phase. For effective restoration of
fertility after cropping, the resting phase must be long enough for
deep-rooted plants, such as trees and shrubs, to become well established
and for their roots to penetrate into the deeper zones of the subsoil.
During the resting phase, there is a net increase of available nutrients
by weathering, nitrogen fixation and upward movement of nutrients from the subsoil to the topsoil. The organic matter is replenished through
the shedding of leaves and the accumulation of other plant debris.

In former times, the resting phase might last for 25 years or
more before the land was again cleared for cropping. Crops were then
grown for 2-4 years before allowing the land to revert to natural
vegetation for another period of regeneration. As population pressure has increased, however, the increasing demand for arable land has been.
met by shortening the resting phase so that the restoration of fertility
between cropping periods has gradually been diminished..' Ultimately a low-level equilibrium is reached in which, even in favourable seasons, yields are barely adequate to justify the I-abourj: anid the soils become
* dangerously degraded and vulnerable to erosion.
Population pressure is also aggravating the problem of
* maintaining soil fertility through the demand for fuel for cooking. As
firewood becomes more scarce, greater use is made of crop residues and
animal manure for fuel, making them less available for the replenishment of soil fertility, both in terms of nutrients and in the organic matter
that is so Important for soil structure.*
3.1.2. Water Resources
Water resources present a global paradox. it is estimated that
the volume of fresh water, anually renewed through a cyclic flow
involving sea, air and land, is sufficient to meet the material needs of
* many times the existing world population.- Despite such apparent
abundance, periodic shortages of water threaten millions with famine,
* and water tables in parts of India, China, Mexico and the USA are
falling precipitously (34).
The commonly-held view that water is an inexhaustible gift from
nature is dangerous in that it leads to complacency. The relentless
growth of population with the'associated increase in demand for water
for domestic consumption, industry and agriculture impose increased
pressure on the the world's water resources. If this trend continues,
there will be increasing shortages of water of suitable quality, unless
serious attempts are made to improve the efficiency of its use (35).

(i) Irrigation
Agriculture is the principal user of global water suppiies.
From historic times, irrigation has been used to help farmers to secure a reliable and timely supply of water for their crops. By 1900, irrigation was supplying some 40 million hectares of land. The area had more than doubled to 94 million hectares by 1950 and tripled by i985. This dramatic increase in the area under irrigation was a major contributingfactor to the increases in global fobd production (36).
Today, the 17% of the world's cropland that is irrigated
supplies about one-third of the agricultural production. As arable land becomes more scarce, irrigation enables farmers to increase production on existing land. During the 1950s and 1960s, irrigated areas expanded at the rate of about 4% per annum, but by the early 1980s, the rate of growth had declined to less than 1% (34).
Unfortunately, some of this expansion in the irrigated area has been at the expense of using fossil water or of the overdrafting of rechargeable aquifers. Both practices are unsustainable. In addition, in some localities water is being rendered unsuitable for irrigation through chemical and biological pollution.
These problems are aggravated by the fact that a great deal of irrigation water is used inefficiently, with much more water often being transported and applied to crops than is required (37). Poor irrigation practices result in severe problems of land degradation through waterlogging and salinization, which is estimated to render unproductive some 1.0-1.5 million hectares of good cropland annually. These problems are particularly acute in India and Pakistan where it is estimated that 12 million hectares have been seriously degraded. Indeed, in many areas of the world, waterlogging and salinization threaten to diminish the very gains that costly new irrigation projects are intended to provide (34).
(ii) Rainfall
In the absence if major problems of soil-toxicity,* agricultural productivity in vast &. s cr! --he tropics is greatly affected by the

amount, distribution and reliability of rainfall. Although it is unlikely that science will find ways of controlling the supply of rainfall, at least in the foreseeable future, science has much to
contribute to its efficient use. Attitudes towards rainfall, especially
in semi-arid regions, must be similar to those for irrigation water: it
must be regarded as a scarce resource that should not be wasted. Its
efficient use is crucial to raising and su staining yields in areas where
irrigation is impossible.
There are two main aspects to the problem. one involves the
* management of soil; the other, the management of the crop. Both are
designed to conserve rain water and improve the efficiency of its use.
The soil assumes the role of a bank in which rainfall can be
stored; soil management that prevents runoff and allows rapid
* infiltration also permits maximum 'storage of rain water. Moreover,
appropriate use of fertilizers improves ;water-use efficiency. Crop
management that optimizes the relationship between plant water
requirement and rainfall expectation,.by timely sowing for example, is
likely to optimize yield.
other ways of conserving rain water involve various methods of
* water harvesting", for- example by holding the water on the land in
* small basins, or tied ridges, or by collecting runoff in various ways..
The application of all of these principles is often limited in
* developing countries, however, by bottlenecks in the availability of
-. labour or by inadequate backstopping from the research and extension
services. Nevertheless, the more efficient-use of rainfall lies at the
* heart of meeting the rising demand for food in all rainfed areas of
* production.
3.1.3. Toxic Chemicals
Human activities are responsible for releasing chemicals,
including radio-active elements, iqo thc.-,"4iro=Pent that may have

deleterious, effects upon plants and animals and also threaten sustainability. The publicity that has been given to "acid rain", for example, is well-known. It is a problem that arises from combustion of fossil fuels and smelting of metallic ores, resulting in the release of large quantities of sulphur dioxide into the atmosphere, where it may be carried for great distances, ultimately to.come down as acid-contaminated rainfall Nitrogen oxides, also the products of industrial processes, similarly find their way into rainfall as strong acids.
This acid-contaminated rain has the potential to damage both terrestrial and aquatic organisms, and many areas in Europe and North America have already incurred serious damage, especially areas of forest in Europe (38). Another potentially harmful effect of acid rain is its contribution to the acidification of soils, especially in humid regions, where soils may already be too acid for optimum crop performance. Furthermore, acidification of soils through acid rain may increase the solubility of some'elements in the soil to the point that they, too, become toxic to plants.
Industrial activity has also resulted in pollution of the environment by a number of metals, several of which have increased rapidly. For example, lead, large amounts of which is used in petroleum, is being released into the atmosphere at an.estimated rate of 2 million-metric tons annually (39). If thesemetals reach sufficiently high concentrations in soils or in water, they have the potential for becoming toxic to living organisms.
Some chemicals with beneficial uses, such as fertilizers and
pesticides, can have harmful effects, especially'if used inappropriately. There is growing global concern about the pollution of both surface and subterranean water, for example, with compounds of nitrogen and phosphorus as well as with pesticides. The ability to continue to benefit fully from the use of many of these chemicals in future may well depend upon refinement in their use.
Moreover, some chemicals that occur natuvll lu.-'Me soil may be found in such high concentrations as to be toxic t. Wfs j

especially in acid soils where high concentrations of aluminiun, manganese, iron and possibly other elements may adversely affect plants. Under more alkaline conditions, high concentrations of sodium chloride and other salts may also be harmful to plants. In addition, high acidity or alkalinity may contribute to low.solubility and deficiencies of still other plant nutrients.. In either case, crop production may be lowered and sustainability threatened.
3.1.4. Changes in Climate
Climate is one of the dominant factors that determines the
nature of ecosystems and hence the distribution of crops and ruminant livestock as well as their productivity. Consequently, climatic factors and long-term changes in climate have major implications for sustainability.
The severe drought in the Sahel for over two decades has
prompted great speculation concerning its nature whether .it was caused by changing atmospheric circulation patterns that have historically caused droughts, or whether it was the result of major changes in land use and soil degradation that may have altered rainfall patterns. Most of the changes in land use have resulted in increased run-off, a reduction in both evaporation and transpiration and an increase in the reflectivity of the land.
Various studies have shown that this African region has been subjected to similar drought conditions before. Accordingly, there is an emerging preponderance of opinion that what has happened over the past two decades in the Sahel does not represent a long-term or permanent change to more arid conditions. Nevertheless, the combination of drought and the pressures of increasing numbers of people and. livestock have contributed to severe ecological degradation, to the point of desertification in much of the region..
While the effects on long-term changes in climate on the
degradation of natural resources, especially severe deforestatio'.,;ar e--. ) .

not entirely understood, there is growing evidence that environmental'
degradation may have some undesirable effects on the 'hydrological cycle.
Research on the watershed in the central Amazon, for example, indicates
that when rain falls in a tropical rain-forest, approximately one-fourth
of the precipitation runs off, returning to the ocean while three-.
- fourths returns to the atmosphere through evaporation or transpiration.
When the rain-forest is cleared for cropping, this ratio is approximately reversed, with three-fourths of the precipitation running off into rivers
and ultimately into the ocean, and one-fourth evaporating (13).
In addition to changes that may affect rainfall distribution,
there is considerable scientific evidence pointing to a significant
warming trend in the climate. Various mathematical models predict a
global warming in the range of 20 to 50C during the next century (16).
* Others have suggested an increase of 2-30 by the year 2050.
This "greenhouse effect" is the result of a build-up of carbon
dioxide and other gases, such as methane, in the atmosphere. Some
studies indicate that the carbon dioxide level in the atmosphere will
double in the next 60-80 years. This build-up of carbon dioxide, which
is a consequence of the combustion of fossil fuels, wood and other
*materials, is not, in itself, harmful. Indeed, higher than normal
concentrations of carbon dioxide in the atmosphere enhance plant growth,
water-use efficiency and crop productivity.
According to some, the warming trend is likely to be most
dramatic in high latitudes and least significant in equatorial regions..
Coastal lowlands, however, may suffer increased risk of flooding due to
a greater melt of polar land-ice. Possible effects of the warming trend
upon precipitation are more speculative, although there appears to be
considerable agreement that the cornbelt in the USA could become drier
as well as warmer, pushing the effective climatic zone for corn
production further North (26).
Although current understanding of these phenomena is inadequate
to make reliable predictions of the consequences, the long-term. effects
*of such changes could have significant effects on the sustainability of

agricultural production in various parts of the world. 'Consequently the trends need to be carefully monitored in order to assess whether or not technology can be adapted rapidly enough to meet the changing circumstances.
3.2. Biological Factors
3.2".. Crop Pests 1/
Intensification of production to meet growing food needs brings with it greater risks of the build-up of pests which, if not controlled, affect stability of production in the short-term and sustainability in the longer term. Although losses vary widely, it has been estimated that pests contribute to the loss in the field of some 35 of the potential production of major food crops, with the greatest losses occurring in developing countries (40).
Although-high priority is often given to diseases and arthropod pests perhaps the most persistent cause of depressed yields in tropical environments is competition from weeds. It follows that there could probably be no greater contribution to crop productivity in developing countries than by effectively controlling weeds. Moreover, effective weed control demands well-grown crops which, in turn, contribute to the prevention of soil erosion and, through vigorous root growth, to the maintenance of soil structure.
Various parasitic weeds, belonging to the genera Striga and Orabanche, are also increasing in importance as crops spread into new areas, rotations are shortened and both organic matter and nitrogen in the soil are depleted. Their control presents new challenges to science, but ieft unchecked they pose an additional threat to sustainability.
I/ The term "pests" is used herein to refer to weeds, diseases,
insects, mites; nematodes, birds, rodents and other mammals.

Insect pests are generally most abundant in the warm and humid environments of the tropics and sub-tropics. Crops grown under improved agronomic practices are sometimes more prone to insect attack than those more sparsely grown. Moreover, insect pests are more abundant in those areas that grow the same crop, such as rice, all the year round. In the Philppines, for example, insect problems have increased significantly in areas where a single rain-grown crop of rice has been replaced by several successive crops grown under irrigation in the same year, illustrating. the potential loss of sustainabilityras intensity of production is increased. Under these conditions, appropriate control measures for insect pests have given increases in yield of 20-25% or about 1.0 t/ha (41).
Crop diseases also build up more rapidly and become more
serious in tropical environments than in temperate zones. Crops are particularly vulnerable to disease attack when large areas are planted to the same variety, or when successive crops of the same variety are grown on the same field. Diseases can build up to the extent that they threaten sustainability, a threat that can be partly avoided by suitable cropping patterns or sequences. But a greater threat to the long-term control of diseases arises from a breakdown in the effectiveness of pesticides and host-plant resistance through mutation in the pest or parasite.
The occurrence of resistance in insect. pests to insecticides has been known for over 70 years. The greatest increase in resistance has occurred during the last 40 years following the extensive use of organic insecticides. Resistance in plant pathogens to chemical treatment has also built up during the last 40 years, with reports being most frequent during the last 15 years, coinciding with the introduction of systemic fungicides. Resistance of noxious weeds to herbicides is more recent, but is now being reported more frequently in species that have been treated intensively (42).
The first report of host-plant resistance to a disease being overcome by changes in the parasite also appeared nearly 70 years ago after the release in the USA of the first wheat variety resistant to

stem rust. Since that time, the apparent breakdown of-resistance to insects and diseases has become commonplace and numerous strategies have been evolved to maintain resistance in the face of continuing changes in the pest. This type of research, which has become known as "maintenance research", now constitutes a major part of plant breeding programs.
For example, CIMMYT estimates that two-thirds of its total
effort in wheat improvement relates to maintenance research, primarily, but not exclusively, on disease resistance. The-aignificance of such research is made apparent in an analysis of the potential losses in production from the major wheat pathogens, resistance to which is incorporated into improved wheat cultivars. Estimates of potential yield losses from these diseases were based on a range of loss measurements made during actual epidemics in the past. The maximum potential losses are based on the assumption of a serious breakdown of resistance in the most popular CIMXYT-based variety, Sonalika, which is now planted on some 8 million hectares in the Third World. The analysis shows potential yield losses from the various pathogens ranging from 5-60% with the total potential reduction of 760,000 tons/year (43).
Such potential losses emphasize the important contribution which maintenance research must make to sustainability. But other approaches to the achievement of sustained control of pests involve integrating crop management with control measures. For example, mixtures of varieties of the same crop or of different crops show reduced build-up of pests under some circumstances. With appropriate mixtures of spring barley in the UK, for example, reductions of infection by powdery mildew of up to 80 per cent have been recorded, compared with the mean disease level of the component varieties grown in pure stand. Similar effects have been observed in tropical crops. For example, a reduction of the damaged caused by the brown plant hopper has been reported in mixed stands of rice-(44).
These are relatively simple and well-researched aspects of the wider concept of integrated pest management which must increasingly be featured in efforts to control pests at the higher levels of productivity thct.-wll1.e required in future.
............... ... "

3.2.2. Diseases and Parasites of Animals
The control of pests is also important in sustaining livestock production. It is estimated that, globally, diseases and parasites are responsible for the death of 50 million cattle and water-buffalo and 100 million sheep and goats each year (40). Such mortality figures do not indicate the full extent of the problem, however, since diseases and parasites may contribute to a serious reduction in the productivity of animals without being reflected in mortality statistics. More effective control of animal diseases would, consequently, make a major contribution to the sustainability of animal production.
3.2.3. Genetic Resources
The continued genetic improvement of crops and livestock, which is so important for maintaining and increasing productivity, is dependent on the availability of suitable sources of genetic diversity. The preservation of obsolete varieties of crop plants and breeds of animals is vital for future breeding programs. Primitive land races of crops and their wild relatives also provide valuable sources bf genes, particularly those conferring resistance to pests and various environmental stresses. These genes have become even more valuable with the prospect of their transfer across reproductive barriers, which is becoming increasingly possible through the development of new techniques in protoplast fusion and molecular biology.
The CGIAR has given leadership to the conservation of plant genetic resources and the creation of gene banks. it is essential for future sustainability of production that this work should continue and that national governments give greater consideration to the conservation of genetic resources, particularly to the preservation in situ of wild species of both plants and animals.

4.1. CGIAR Centers
Historically, the CGIAR Centers have effectively promoted the
development of technologies that contribute to increased food production
in developing countries, and much of their work has long-term
*Implications for sustainability. Their current activities are analysed
from this perspective in Annex 1. This section summarizes the main
4.1.1. Centers Working on Crop and Livestock Productivity
Although the mandates of the ten Centers working on crop and
livestock productivity vary in scope and.complexity, their research can be grouped into three broad areas of investigation: germplasm, resource
management and agro-ecology.
(i) Germplasm
In that the dynamic concept of sustainability implies
increasing productivity to meet the needs of a growing population, all
of the work on germplasm has implications for sustainability. The
conservation of genetic resources, for example, through exploration,
collection, storage and evalutation, is designed to preserve sources of
genetic diversity, without erosion or deterioration. Work on crop
improvement is designed to enhance yielding capacity while introducing
and maintaining resistance to environmental stresses, such as pests,
diseases, weeds, drought, salinity and soil toxicity. A basic aim is to breed for broad adaptation and achieve greater yield stability, in order
to reduce those wide fluctuations in seasonal production that are so
harmful to the small producer.
(ii) Resource Management
Within the context of resource management, major emphasis has
been placed throughout the CGIAR System on the devcL'pment of integrated

pest management (IPM). Such efforts often involve monitoring the build-up of pests under different systems of crop management and pest control, in order to discover more cost-effective and sustainablemethods of crop production. A major objective is to minimize the use of pesticides by developing other methods of pest control and incorporating them into production systems.Research on the control of diseases is not confined to crops, but is extended to encompass cattle through. the .vork of ILRAD, which works mainly on the control of theileriosis and trypanosomiasis. Success could have enormous implications for livestock production and increasing food supplies in Africa and elsewhere. ILRAD is already studying the potential social, economic and environmental consequences of the introduction of control measures for these two diseases, while ILCA is advancing knowledge of livestock production systems with the aim of optimizing the use of resources.
The Centers working on productivity also devote considerable effort to research on soil and water management. Most are working on soil fertility, determining nutrient requirements, especially of nitrogen and phosphorus, under different soil conditions and cropping systems. In addition, some of the research involves studies of soil acidity and alkalinity, along with related chemical toxicities and nutrient deficiencies. There is considerable work on the biological fixation of atmospheric nitrogen, involving both symbiotic and free-living organisms. Likewise, current work on mycorrizal associations is important in that it offers opportunities for increasing the uptake of phosphorus in certain tropical soils.
(iii) Agro-ecological Characterization
The development of sustainable agricultural production systems requires a thorough knowledge of the circumstances in which they are intended to be used. Most of the Centers are therefore concerned with the agro-ecological characterization of the environments in which their mandates take them. The main emphasis is on land form, soils and climate, but socio-economic factors are increasingly taken inttrA account.

Stimulated by an inter-center workshop in 1986 at FAO, cooperation with the World Meteorological Organization and the International Soil Science Society has been strengthened, and attempts are being made to coordinate the working contacts with national systems.
4.1.2. Centers with Other Mandates
There are three additional Centers in the CGIAR System, whose
mandates are not directly concerned with research on aspects of crop and livestock productivity, namely IBPGR, ISNAR and IFPRI. IBPGR is dedicated to the conservation of genetic resources, the importance of which has already been discussed.
ISNAR's role is to assist in the strengthening of national agricultural research systems. Its principle contribution in the context of this paper is to assist developing countries to manage their national agricultural research systems so that appropriate priority is given to sustainability.
IFPRI is playing a vital role in addressing'some of the
critical policy issues that affect sustainability, including the future dynamics of food demand and supply; food policy, choice of commodities and agricultural technology; trade and food security; and development strategy.
4.2. Non-CGIAR Centers
The IARCs.that are not part of the CGIAR System are also
heavily involved in research related to sustainability. AVRDC, for example, has a research agenda very similar to those of the CGIAR Centers with commodity mandates. Other major non-CGIAR centers deal with factors closely related to sustainability and some have indicated that sustainability is at the core of their concerns.

For example, ICRAF states that sustainability of production systems is a central and integral theme of its work, and that agroforestry is often justified on the basis of its potential to sustain crop and animal productivity. ICRAF suggests that the most obvious need for research in agroforestry is to elucidate the potential role of trees in sustaining soil productivity. ICRAF further suggests that agroforestry has a broader potential role in terms of sustaining the carrying capacity of land-use systems in the light of increased populations. In the process of intensified land-use, trees can play both an ecological role (e.g., fertility maintenance and erosion control) and an economic role (e.g., increased income through sale of tree products) (45). Moreover, by providing fuelwood, systems of agroforestry can make it unnecessary for rural households to use animal manures and crop residues for fuel, allowing these nutrient-enhancing organic materials to be returned to the soil.
IBSRAM indicates that its goal is to "promote sustainable improved soil management technologies in order to remove soil constraints to food and other agricultural production- in developing countries". IBSRAM suggests that "the state of soil would appear to be the key factor to sustainability. Soil fertility depletion, acidification, salinization, deterioration of the soil's structure, soil erosion, laterization, and depletion of the soil's bacteriological activity are the most recognized consequences of non-sustainable agricultural systems in the tropics" (46).
IFDC emphasizes the key role which fertilizers must play in
achieving sustainability goals through the restoration and amelioration of soil fertility. IFDC further stresses the role it can play, both independently and in collaboration with other Centers, in research relating to the use of fertilizers (47).
IIMI is concerned in a number of ways with the relationship between irrigation and sustainability, including research directed towards more efficient operation and management of irrigation systems; the rehabilitation and improvement of existing irrigation systems; and the use of irrigation to improve the productivity of cropping systems (48).

, 45
ICIPE is engaged in developing strategies for integrated management of insect. pests tailored to the needs of resource-poor, small-scale farmers. Such strategies emphasize the development and use of non-chemical means of pest control, including the incorporation of resistance to pests in crop cultivars, intercropping, and biological control measures. The livestock pests studied by ICIPE include the ticks that transmit the protozoan agents causing East Coast Fever in
cattle, and tsetse, which is a vector of trypanomsomiasis in cattle and sleeping sickness in human beings (49) .. .-"
4 ,*

* 46
TAC requested the IARCs both within and outside the CGIAR. System, as well as a number of other international organizations, to outline their opinions on the need for research to advance sustainability. Details of the responses oQbtained are given in Annex 1. This chapter summarizes the main views expressed.
5.1. Suggestions from the IARCs
. Many of the comments from the IARCs indicated a need to
continue much of the work currently underway. At the same time, most of them suggested that sustainability should receive higher priority in their programs. Furthermore, emphasis was placed on the need to help. sensitize "national research system colleagues" to the urgency of the problem. Centers also stressed the need to give priority to consideration of issues related to sustainability in their training
All of the Centers with commodity mandates indicated that there should be continued effort on breeding to maintain the gains already achieved. But it was recognized that production stability was not enough and that there must be continued increases in production on a sustained basis to meet the needs of an expanding global population.
Center responses suggested that plant breeding would continue to play an extremely important role in achieving higher levels of productivity. Most Centers indicated that-major emphasis would be given to the development of cultivars adapted to stress environments. Furthermore, emphasis was placed on the continued collection and characterization of germplasm from threatened areas, as well as from remote-.and isolated habitats.

Most IARCs (both CGIAR and non-CGIAR) indicated that major emphasis must be given to research relating to soil and water management. CIMXYT, for example, which is currently giving primary emphasis to plant breeding, indicated that it would be giving serious consideration to shifting resources towards research related to soils, as well as to strengthening its relationships-with other institutions specializing in such work. Major emphasis was also placed by many
Centers on the need for a better characterization of the environments of large agricultural regions'and the interaction Qf-such environments with cropping and management systems. Centers also stressed the importance of research relating to the effect of various government policies upon sustainability.
IFPRI stressed its continuing contributions to both the demand and supply aspects of'sustainability. Research on choices among crops and technologies, factors influencing production variability, infrastructural influences on production, institutional viability, prices and subsidies, and potential benefits of structural adjustments all have important sustainability implications.
Some of the other problems identified by the Centers clearly lie outside the scope of the IARCs, but fall within the ambit of other organizations. The Global Environmental Monitoring System (GEMS), for example, is monitoring changes in atmospheric, soil, and water parameters. Furthermore, there are many other organizations involved in research having sustainability implications, including the FAO, the Man and the Biosphere Program of UNESCO, the World Wildlife Fund as well as
various organizations that work under the auspices of the International Council of Scientific Unions (ICSU), such as the Commission for the Application of Science to Agriculture, Forestry and Aquaculture (CASAFA), the International Union of Biological Sciences (IUBS) and the International Biological Network (IBN). There is obvious need for close, continuing interaction between the IARCs and these other organizations.

5.2. Approaches of Other Organizations to Sustainability Concerns
5.2-1. FAO
The FAO response to TAC's enquiry listed a series of problems relating to soil and water management and fertilizer use, while stressing the importance of genetic conservation of both plants and animals. ftphasis was also placed on the lnportgnce of'integrated pest management to curtail over-reliance on pesticides (50).
In its recent study of African Agriculture (11), FAO points out that during the past 10 to 15 years, the primary emphasis in the development of agricultural technology has been on the use of high inputs including industrial fertilizers, machinery, pesticides and irrigation. Because of the risks involved, however, subsistence farmers tend to use a minimum of inputs to ensure some return for their outlay, even under the most unfavourable seasonal conditions. It is recognized that, in marginal areas, farmers are particularly cautious about using inputs and they need technologies that increase yields and income, or that reduce risks without incurring the costs of major external inputs.
The study emphasizes a strategy of development based on
conservation that would enable land to be managed for the benefit of the community in ways that conserve and expand its potential for use by future generations. It suggests that the greatest potential for increasing food production at low cost is through water harvesting, the control of soil erosion, alley cropping, using crop residues to recycle nutrients, community afforestation and small-scale irrigation.
5.2.2. UNEP
The UNEP response to TAC's enquiry emphasized many of the points made by FAO. Among those aspects stressed, were the need for assessing the impact on the environment of new agricultural technologies, the full exploitation of the potential for biological

- 499
nitrogen fixation and biofertilizers, the need for continuous surveying and recording of ecosystems, the promotion of awareness and training programs in developing countries, and the need to evaluate the effect of introducing novel organisms into the environment (51).
The UNEP "State of the Environment" report (52) deals with environmental aspects of emerging agricultural technologies. After sumarizing the effects on the environmentof modern and emerging agricultural practices, the report notes that ex-sting agricultural technologies are continually being improved, or applied in new situations, in attempts to increase food production and to reduce both .the costs and the adverse effects on the environment.
The report acknowledges that much research is underway, but stresses the need for accelerated effort on integrated research and development at both the national and international levels, to develop appropriate and sound agricultural practices. The report further stresses, however, that research alone is insufficient. It suggests that national development plans should emphasize the formulation and implementation of environmentally sound agricultural policies and. practices, with priority being given to conservation and the non-wasteful use of resources. Among the highest priority items listed
in the report were the formulation and implementation of national soil conservation policies, national water management plans, measures (regulations, incentives, pricing, control mechanisms, etc.) to conserve different agricultural resources'and the further development and modernization of indigenous agricultural technologies.
5.2.3. World Commission on Environment and Development (WCED)
The report of the advisory panel on Food Security, Agriculture, Forestry and Environment of the World Commission on Environment and Development sets forth a number of steps needed to achieve "sustainable livelihood security" (53). While not focusing on research, per se, the report reinforces the propositions that conservation-based production systems should be the aim of agricultural development and

technologies that are carefully tailored to agro-ecological and socio-economic circunstances should be developed and applied. The Panel further proposed that location-specific conditions should be the main determinants of types of technology and encouraged the strengthening of national and sub-national research capabilities.
5.2.4. Office of Technology Assessment, US Congress
The Office of Technology Assessment, an advisory body to the US Congress, is completing a major study of the potential of low-resource agriculture in Africa. A draft copy of the report indicates that low-resource agriculturalists produce the-majority of the continent's major food crops (21).
The report recognizes, however, that low-resource agriculture in Africa is increasingly unable to keep pace with the pressures and demands being placed upon it in certain areas because of unprecedented population growth and increasing poverty. It suggests, therefore, that technological innovation in low-resource agricultural systems will be a major factor in determining Africa's ability to meet the challenges ahead.
The report further suggests that rather than relying on the introduction of well-defined packages of inputs, the most viable strategy for promoting food security calls for the evolution of existing farming systems. Consequently, technical assistance should be geared primarily to the needs of low-resource agriculture.
o-t i p

6.1. Responses to the Challenge
If sustainability is considered in the context TAC proposes, agriculture must be managed to meet the needs of steadily increasing numbers of people. This means that the circumstances that threaten to reduce current production levels must not only b. alleviated, but that there must also be major efforts to increase productivity to meet
growing needs.
Throughout the world, the enormity of these tasks is causing
serious and rapidly growing concern. TAC views the challenge of finding timely and workable solutions as one of major dimensions, which should receive highest priority on a continuing basis, not only by the CGIAR but by all organizations and institutions that can make a contribution. Further emphasis of the need for high priority stems from the knowledge that all regions of tbe world, including both developing and industrialized countries, face problems of agricultural sustainability to a greater or lesser extent. The sense of urgency that this engenders should reinforce the concerted effort that will be necessary for
It must be emphasized from the outset that many of the
circumstances that limit the achievement of sustainability (chapter 2) cannot be solved by the CGIAR or through agricultural research alone. National governments and their development services will bear the brunt of the problems, and success in achieving sustainability will largely depend upon their commitments and efforts. Nonetheless, continuing research is crucial for success and the CGIAR institutions as well as the national agricultural research systems must constantly examine their programs to give greater emphasis and visibility to those aspects that relate to sustainability.
Although TAC believes that the CGIAR System can make a
significant contribution to sustaining agricultural- production at the

increasing levels required in. the future, its total effort must be kept
in perspective. In 1980, total expenditure in the CGIAR System
represented only 1.6% of the global public sector expenditure and.
approximately 5% of developing country expenditure on agricultural
research (2). But the CGIAR Centers can have an Impact far in excess of
their relative level of expenditure through their ability to influence
research activities in other institutions throughout the world.
Furthermore, donors and other components of the CGIAR System can be
* helpful in focusing attention on sustainability %nd encouraging
governments and relevant institutions to give it appropriate priority.
If, as TAC has suggested, sustainability relates to the
successful management of resources for agriculture to satisfy future
human needs, it might well be argued that most of the work of the IARCs
is already contributing to this goal. Indeed, if research were not
orientated towards sustainability, it should have little place in the programs of the Centers. The key question, therefore, is not so much
* whether the Centers are working to make agriculture more sustainable,
but whether they should be doing more and whether there should be a
different emphasis in the work.
A main task of the IARCs is to develop new agricultural
technologies that can be adapted with relatively little modification to diverse socio-economic and agro-ecological conditions. The question of
whether or not such new technologies contribute to sustainable
agriculture cannot be considered in isolation but depends on where and
how they are applied. Consequently, consideration of sustainability must be an integral part of the procedures for planning and reviewing
all research projects that are directly concerned with the development A of new agricultural technology.
6.2. Research with a Sustainability Perspective
TAC does not view research related to sustainability as a
separate or discrete area of activity. Rather, concern for
sustainability should be reflected in the way in which the research is

approached. TAC therefore recommends that research at the Centers
designed to generate new agricultural technology should be planned and
conducted with a sustainability perspective.
The alms and progress of a research program should be reviewed
with a sustainability perspective from its conception to the
implementation of its results. The various technological innovations
suggested by the research should be assessed in relation to
sustainability and their potential domains.of application specified.
Within each domain, the resource demands of the proposed technology
should be quantified, as far as possible, and the implications for
sustainability further considered. The process of assessment should be continuous and iterative and should be an integral part of the planning
and review of research programs.
At the level of the single domain of application, the process
of assessment would naturally be linked'to the farming systems approach.
* Indeed considerations of sustainability should feature strongly in
on-farm research.(OFR). Sustainability should always be an evaluation
criterion in the process for identifying, screening and adapting
technological solutions to farmer problems. During this process, the
disciplinary scientists interact with the OFR team who have a detailed
knowledge of the local circumstances, including the existing farming
system, in which the proposed. technological solution has to be applied.
The specialists prescribe the content, timing and method of each
potential solution. They should also spell out the requirements for,
and effects on, sustainability.
The OFR team can then use their knowledge of local
circumstances to analyse whether the sustainability requirement can be
met. Sustainability should be given weight alongside other criteria,
such as farmer priorities and resources, research costs, input
availability, profitability and risk, in deciding, ex ante, if the
proposed technology is feasible and suitable for inclusion in the OFR
The Implications for applied research are clear. Just as it
has become accepted that new technologies have labour, cash and

management implications that. will be important in the eyes of the farmer, so they will also have implications for sustainability that will be important in the eyes of the community as a whole. Both the soil and water characteristics required for successful use of a technology and the medium-term and long-term effects of its adoption on soil, water and other resources, need to be specified as a routine part of applied research projects.
The-se principles imply a continuing nee- for the Centers to
remain in close touch with on-farm research through their involvement in networking arrangements with national research systems. They also have implications for the further development of training in the methodologies of on-farm research to ensure that evaluation of sustainability'is adequately emphasized (see also section 6.12).
... ...... .....
TAC considers that, in formulating or revising their strategic plans, Centers should include proposals for maintaining a sustainability perspective throughout their total programs.
6.3. Short-term versus Long-term Objectives
Just as the farmer is faced with the choice between short-term and long-term considerations (see section 1.6), so the research.worker has to consider both horizons. If the goal of sustainable agricultural production is to meet the changing needs of people, research must clearly cater for both short-term and long-term needs. Centers already direct their activities largely towards meeting the needs of people, but the extent to which they specifically include long-term considerations of sustainability in their research planning differs with the nature of their mandates.
In spite of the dilemma of the resource-poor farmer in the
choice between short-term and long-term gains, a guiding principle for all Centers must be that stability of the environment should never be consciously sacrificed for short-term gains. The aim should obviously be to devise technologies 'tat.can-meet short-term-requirements while,