Reducing food poverty by increasing agricultural sustainability

Full text

1
Reducing food poverty by increasing agricultural 1
sustainability in developing countries 2
3
4
J.N. Pretty, J.I.L. Morison and R.E. Hine 5
Centre for Environment and Society and Department of Biological Sciences, 6
University of Essex, Colchester UK 7
8
9
Agriculture, Ecosystems and Environment 95(1), 217-234 10
11
12
13
Corresponding author: 14
Prof. Jules Pretty 15
Centre for Environment and Society and Department of Biological Sciences, 16
University of Essex, 17
Wivenhoe Park 18
Colchester CO4 3SQ, UK 19
20
Tel: +44-1206-873323 21
Fax: +44-1206-873416 22
Email: jpretty@essex.ac.uk 23

2
Abstract 24
We examined the extent to which farmers have improved food production in recent years 25
with low-cost, locally-available and environmentally-sensitive practices and technologies. 26
We analysed by survey during 1999-2000 208 projects in 52 developing countries, in 27
which 8.98 million farmers have adopted these practices and technologies on 28.92 28
million hectares, representing 3.0% of the 960 million hectares of arable and permanent 29
crops in Africa, Asia and Latin America. We found improvements in food production 30
occurring through one or more of four mechanisms: i) intensification of a single 31
component of farm system; ii) addition of a new productive element to a farm system; iii) 32
better use of water and land, so increasing cropping intensity); iv) improvements in per 33
hectare yields of staples through introduction of new regenerative elements into farm 34
systems and new locally-appropriate crop varieties and animal breeds. The 89 projects 35
with reliable yield data show an average per project increase in per hectare food 36
production of 93%. The weighted average increases across these projects were 37% per 37
farm and 48% per hectare. In the 80 projects with small (< 5 ha) farms where cereals were 38
the main staples, the 4.42 million farms on 3.58 million hectares increased household food 39
production by 1.71 t yr-1. We report on the practices and technologies that have led to 40
these increases: increased water use efficiency, improvements to soil health and fertility, 41
and pest control with minimal or zero-pesticide use. This research reveals promising 42
advances in the adoption of practices and technologies that are likely to be more 43
sustainable, with substantial benefits for the rural poor. With further explicit support, 44
particularly through national policy reforms and better markets, these improvements in 45
food security could spread to much larger numbers of farmers and rural people in the 46
coming decades. 47
48
Key words: agricultural sustainability, food security, rural livelihoods, policies 49

3
Introduction 50
Over the past 40 years, per capita world food production has grown by 25%, with 51
average cereal yields rising from 1.2 t ha-1 to 2.52 t ha-1 in developing countries (1.71 t ha-1 52
on rainfed lands and 3.82 t ha-1 on irrigated lands), and annual cereal production up from 53
420 to 1176 million tonnes (FAO, 2000). These global increases have helped to raise 54
average per capita consumption of food by 17% over 30 years to 2760 kcal day-1, a period 55
during which world population grew from 3.69 to 6.0 billion. Despite such advances in 56
productivity, the world still faces a persistent food security challenge. There are an 57
estimated 790 million people lacking adequate access to food, of whom 31% are in East 58
and South-East Asia, 31% in South Asia, 25% in Sub-Saharan Africa, 8% in Latin America 59
and the Caribbean, and 5% in North Africa and Near East (Pinstrup-Andersen et al., 60
1999). A total of 33 countries still have an average per capita food consumption of less 61
than 2200 kcal day-1 (FAO, 2000). 62
63
An adequate and appropriate food supply is a necessary condition for eliminating 64
hunger. But increased food supply does not automatically mean increased food security 65
for all. A growing world population for at least another half century, combined with 66
changing diets arising from increasing urbanisation and consumption of meat products, 67
will bring greater pressures on the existing food system (Popkin, 1998; Delgado et al., 68
1999; Pinstrup-Andersen et al., 1999; UN, 1999; ACC/SCN, 2000; Smil, 2000). If food 69
poverty is to be reduced, then it is important to ask who produces the food, who has 70
access to the technology and knowledge to produce it, and who has the purchasing 71
power to acquire it? Modern agricultural methods have been shown to be able to increase 72
food production, yet food poverty persists. Poor and hungry people need low-cost and 73
readily-available technologies and practices to increase food production. A further 74
challenge is that this needs to happen without further damage to an environment 75
increasingly harmed by existing agricultural practices (Pretty et al., 2000; Wood et al., 76
2000; McNeely and Scherr, 2001). 77
78
79
Key Questions for Research on Agricultural Sustainability 80
81
There are three strategic options for agricultural development if food supply is to be 82
increased: 83
i. expand the area of agriculture, by converting new lands to agriculture, but 84
resulting in losses of ecosystem services from forests, grasslands and other areas of 85
important biodiversity; 86
ii. increase per hectare production in agricultural exporting countries (mostly 87
industrialised), but meaning that food still has to be transferred or sold to those who need 88
it, whose very poverty excludes these possibilities; 89
iii. increase total farm productivity in developing countries which most need the 90
food, but which have not seen substantial increases in agricultural productivity in the 91
past. 92
93
In this research, we explore the capacity to which more sustainable technologies and 94
practices can address the third option. We draw tentative conclusions about the value of 95
such approaches to agricultural development. This is not to say that industrialised
96
agriculture cannot successfully increase food production. Manifestly, any farmer or 97
agricultural system with unlimited access to sufficient inputs, knowledge and skills can 98
produce large amounts of food. But most farmers in developing countries are not in such 99

4
a position, and the poorest generally lack the financial assets to purchase costly inputs 100
and technologies. The central questions, therefore, focus on: 101
i) to what extent can farmers increase food production by using low-cost and 102
locally-available technologies and inputs? 103
ii) what impacts do such methods have on environmental goods and services and 104
the livelihoods of people who rely on them? 105
106
The success of industrialised agriculture in recent decades has often masked significant 107
environmental and health externalities (actions that affect the welfare of or opportunities 108
available to an individual or group without direct payment or compensation). 109
Environmental and health problems associated with industrialised agriculture have been 110
well documented (cf Balfour, 1943; Carson, 1963; Conway and Pretty, 1991; EEA, 1998; 111
Wood et al., 2000), but it is only recently that the scale of the costs has come to be 112
appreciated through studies in China, Germany, UK, the Philippines and the USA 113
(Steiner et al., 1995; Pimentel et al., 1995; Pingali and Roger, 1995; Waibel and Fleischer, 114
1998; Norse et al., 2000; Pretty et al., 2000, 2001). 115
116
What do we understand by agricultural sustainability? Systems high in sustainability are 117
making the best use of nature’s goods and services whilst not damaging these assets 118
(Altieri, 1995; Pretty, 1995, 1998; Thrupp, 1996; Conway, 1997; Hinchliffe et al., 1999; NRC, 119
2000; Li Wenhua, 2001; McNeely and Scherr, 2001; Uphoff, 2002). The aims are to: i) 120
integrate natural processes such as nutrient cycling, nitrogen fixation, soil regeneration and 121
natural enemies of pests into food production processes; ii) minimise the use of non-122
renewable inputs that damage the environment or harm the health of farmers and 123
consumers; iii) make productive use of the knowledge and skills of farmers, so improving 124
their self-reliance and substituting human capital for costly inputs; and iv) make productive 125
use of people’s capacities to work together to solve common agricultural and natural 126
resource problems, such as pest, watershed, irrigation, forest and credit management. 127
128
Agricultural systems emphasising these principles are also multi-functional within 129
landscapes and economies. They jointly produce food and other goods for farm families 130
and markets, but also contribute to a range of valued public goods, such as clean water,
131
wildlife, carbon sequestration in soils, flood protection, groundwater recharge, and 132
landscape amenity value. As a more sustainable agriculture seeks to make the best use of 133
nature’s goods and services, so technologies and practices must be locally-adapted. They 134
are most likely to emerge from new configurations of social capital, comprising relations 135
of trust embodied in new social organisations, and new horizontal and vertical 136
partnerships between institutions, and human capital comprising leadership, ingenuity, 137
management skills, and capacity to innovate. Agricultural systems with high levels of 138
social and human assets are more able to innovate in the face of uncertainty (Uphoff, 139
1999; Pretty and Ward, 2001). 140
141
Research Methods 142
143
The aim of the research was to audit recent progress in agricultural sustainability in 144
developing countries. We accessed an international network of key professionals in the 145
field of agricultural sustainability and food security, and asked them both to suggest 146
projects and initiatives, and to pass on details of this research project to other relevant 147
people or institutions. We also accessed other datasets (eg Hinchcliffe et al., 1996; FAO, 148
1999). We asked for nominations for three types of initiatives: i) research projects with 149
active farmer involvement, but which may not yet have spread; ii) community-based 150

5
projects with proven impacts; and iii) regional projects/initiatives that have spread to 151
many communities. We use the term `project/initiative’ here, as these have emerged 152
from many types of institutional context – some are international development projects, 153
some are activities within government programmes, some are non-government 154
organisation or private company led, and some are promoted entirely by farmers’ 155
organisations themselves. 156
157
We developed a four-page questionnaire as the survey instrument, with a short 158
descriptive rationale on agricultural sustainability and the aims of this research project. 159
The questionnaire survey instrument was based on an assets-based model of agricultural 160
systems, and was developed to understand both the role of these assets as inputs to 161
agriculture and the consequences of agriculture upon them (Conway, 1997; Pretty, 2000; 162
Pretty and Hine, 2000). The questionnaire addressed key impacts on total food 163
production, and on natural, social and human capital, the project/initiative structure and 164
institutions, details of the context and reasons for success, and spread and scaling-up 165
(institutional, technical and policy constraints). The questionnaire was sent out in 166
English, French and Spanish to all potential projects by email and conventional post in 167
early 1999. Field operatives from nodal organisations were contacted with specific 168
requests and questionnaires. Each project was contacted with a personalised covering 169
letter and questionnaire, and the resultant high response rate (some 60% of those 170
contacted replied with some information) appears to be a consequence of this personal 171
contact. Some 200 reminders and questionnaires were sent out by email in autumn 1999 172
to attempt to access those who had not yet answered. Follow-up contacts were made to 173
many of these during the course of the year 2000. In a number of instances, we received 174
secondary data on the project rather than a completed questionnaire. We collated 175
returned questionnaires and secondary material, and added these to country files. All 176
datasets were re-examined to identify gaps, and correspondents contacted again. 177
178
Not all proposed cases were accepted for the dataset, and rejections were made: i) where 179
there was no obvious link to agricultural sustainability; ii) where payments were used to 180
encourage farmer participation, as there have long been doubts that ensuing 181
improvements persist after such incentives end; iii) where there was heavy reliance on 182
fossil-fuel derived inputs, or only on their targeted use (this is not to negate these 183
technologies, but to simply indicate that they were not the focus of this research); and iv) 184
where the data provided was too weak or the findings unsubstantiated. We also 185
acknowledge that just because projects/initiatives have been accepted for this dataset, 186
this does not necessarily mean they will be sustained indefinitely. The problem of
187
agricultural development activities not persisting beyond the end of projects has been 188
widely-analysed through post-project reviews (Bunch, 1983; Chambers, 1983; Cernea, 189
1991; Carter, 1995). However, we do have confidence about these projects/initiatives, as 190
farmers are adopting novel technologies and practices on their own terms and because 191
they pay – not because they are being offered distorting incentives to do what an external 192
agency wishes. 193
194
The questionnaires were self-completed, so were subject to potential bias. We therefore 195
established trustworthiness checks through checks with secondary data, by critical 196
review by external reviewers and experts, and by engaging in regular personal dialogue 197
with respondents. We verified projects by sending full details entered on the database to 198
the named verifier on the questionnaire. We also sent batches of projects to key 199
authorities to obtain a second or third view on the project. This research, therefore, 200
comprises a purposive sample of existing `best practice’ projects/initiatives explicitly 201

6
addressing agricultural sustainability. It was not a random sample of all agricultural 202
projects, and thus the findings are not representative of all developing country farms. 203
Our aim was to discover the impacts of existing initiatives, to understand the processes 204
and policies that encouraged or restricted them, and to indicate the potential for 205
addressing food poverty through a focus on agricultural sustainability. 206
207
208
Survey Results 209
210
This was the largest known survey of sustainable agricultural practices and technologies 211
in developing countries, with 45 projects in Latin America, 63 in Asia and 100 in Africa, in 212
which 8.98 million farmers have adopted more sustainable practices and technologies on 213
28.92 million hectares. As there are 960 million hectares of land under cultivation (arable 214
and permanent crops) in Africa, Asia and Latin America, more sustainable practices are 215
now present on at least 3.0% of this land (total arable land comprises some 1600 million 216
hectares in 1995/97, of which 388 million hectares are in industrialised countries, 267 217
million hectares in transition countries, and 960 million hectares in developing countries: 218
FAO, 2000). 219
220
The most common country representations in the dataset are India (23 221
projects/initiatives); Uganda (20); Kenya (17); Tanzania (10); China (8); the Philippines 222
(7); Malawi (6); Honduras, Peru, Brazil, Mexico, Burkina Faso and Ethiopia (5); and 223
Bangladesh (4). The projects range very widely in scale - from 10 households on 5 224
hectares in one project in Chile to 200,000 farmers on more than ten million hectares in 225
southern Brazil. Most of the farmers in the projects surveyed are small farmers. Of farms 226
in the total dataset, 50% are in projects with a mean area per farmer of less than one 227
hectare, and 90% with less than or equal to 2 hectares (Figures 1a and 1b). Thus of the 228
total, there are some 8.64 million small farmers practising forms of more sustainable 229
farming on 8.33 million hectares. Most of these initiatives increasing agricultural 230
sustainability have emerged in the past decade. Using project records, we estimate that 231
the area a decade ago in these 208 initiatives was no more than 500,000 hectares. 232
233
[FIGURES 1a and 1b] 234
235
236
Changes in Farm and Household Food Productivity 237
238
We found improvements in food production were occurring through one or more of four 239
different mechanisms: 240
i. the intensification of a single component of a farm system, with little change to 241
the rest of the farm, such as home garden intensification with vegetables or tree crops, 242
vegetables on rice bunds, and introduction of fish ponds or a dairy cow; 243
ii. the addition of a new productive element to a farm system, such as fish or 244
shrimps in paddy rice fields, or trees, which provide a boost to total farm food 245
production and/or income, but which do not necessarily affect cereal productivity; 246
iii. the better use of natural resources to increase total farm production, especially 247
water (by water harvesting and irrigation scheduling), and land (by reclamation of 248
degraded land), so leading to additional new dryland crops and/or increased supply of 249
additional water for irrigated crops (both increasing cropping intensity); 250
iv. improvements in per hectare yields of staple cereals through introduction of new 251
regenerative elements into farm systems, such as legumes and integrated pest 252

7
management, and new and locally-appropriate crop varieties and animal breeds. 253
254
Thus a successful project increasing agricultural sustainability may be substantially 255
improving domestic food consumption or increasing local food barters or sales through 256
home gardens or fish in rice fields, or better water management, without necessarily 257
affecting the per hectare yields of cereals. Figure 2 illustrates the frequency of occurrence 258
of each of these mechanisms in the dataset. The most common mechanisms were yield 259
improvements with regenerative technologies and new seeds/breeds, occurring in 60% of 260
the projects, by 56% of the farmers and over 89% of the area. Home garden intensification 261
occurred in 20% of projects, but given its small scale only accounted for 0.7% of area. 262
Better use of land and water, giving rise to increased cropping intensity, occurred in 14% 263
of projects, with 31% of farmers and 8% of the area. The incorporation of new productive 264
elements into farm systems, mainly fish/shrimps in paddy rice, occurred in 4% of 265
projects, and accounted for the smallest proportion of farmers and area. 266
267
[FIGURE 2] 268
269
As mechanism 4 was the most common, we analysed these projects in greater detail. The 270
dataset contains 89 projects (139 entries of crop x projects combinations) with reliable 271
data on per hectare yield changes with mechanism 4, and these are shown as relative 272
yield changes over a baseline of 1.0 according to yields before or without the 273
interventions (Figure 3). Agricultural sustainability has led to a 93% increase in per 274
hectare food production through mechanism 4 averaged across all projects. The weighted 275
averages are a 37% increase per farm household, and a 48% increase per hectare in these 276
projects. The relative yield increases are higher at lower yields, indicating greater benefits 277
for poor farmers, most of whom have been missed by recent decades of agricultural 278
development. We also analysed the yield data according to crop types (Figure 4). The 279
largest relative increases in yield occur for vegetables and roots, and the smallest for rice 280
and beans/soya/peas. 281
282
[FIGURES 3 AND 4] 283
284
We also calculated the marginal increase in food production per household for these 89 285
projects with reliable data on yields, area and numbers of farmers. Using the data for 286
average farm size in each project, we calculated the average increase in annual food 287
production per household after adoption of more sustainable practices and technologies 288
(Figure 5). In the 80 projects with small (< 5 ha) farms where cereals were the main 289
staples, the 4.42 million farms on 3.58 million hectares increased household food
290
production by 1.71 t yr-1 (an increase of 73%). In the 14 projects with roots as main staples 291
(potato, sweet potato and cassava), the 146,000 farms on 542,000 hectares increased 292
household food production by 17 t yr-1 (increase of 150%). In the four projects in southern 293
Latin America with larger farm size (average size of 90 ha farm-1), household production 294
increased by 150 t yr-1 (increase of 46%). 295
296
[FIGURE 5] 297
298
These aggregate figures understate the benefits of increased diversity in the diet as well 299
as increased quantity. Most of these agricultural sustainability initiatives have seen 300
increases in farm diversity. In many cases, this translates into increased diversity of food 301
consumed by the household, such as availability of fish protein from rice fields or fish 302
ponds, milk and animal products from dairy cows, poultry and pigs kept in the home 303

8
garden, and vegetables and fruit from home gardens and other farm micro-304
environments. Although these initiatives are reporting significant increases in food 305
production, some as yield improvements, and some as increases in cropping intensity or 306
diversity of produce, few are reporting surpluses of food being sold to local markets. We 307
suggest this is because of a significant elasticity of consumption amongst rural 308
households in this dataset experiencing any degree of food insecurity. As production 309
increases, so also does domestic consumption, with direct benefit particularly for the 310
health of women and children. 311
312
As indicated earlier, for an average farm size of 1.4 ha (for the 4.4 million households for 313
which good data exists), the annual increase in gross food production (not including root 314
crops) was 1.71 tonnes. The net amount of food available to each household will, of 315
course, be lower than this – owing to post-harvest losses to pests, conversion of harvested 316
crops to consumable food, and feeding of some as feed to animals. Assuming a worst case 317
of 30% loss to pests, and a further 30% reduction in available food, this still leaves 800 kg 318
of available food per household. This is sufficient to feed two adults or one adult with 319
two children for a whole year. 320
321
We acknowledge that these findings on agricultural sustainability may sound too good to 322
be true for those who would disbelieve these advances. Many still believe that food 323
production and nature must be separated, that practices increasing agricultural 324
sustainability offer only marginal opportunities to increase food production, and that 325
industrialised approaches represent the best, and perhaps only, way forward (cf Avery, 326
1995). However, prevailing views have gradually changed in the last decade, and some 327
sceptics are beginning to recognise the value of innovative capacity emerging from 328
poorer communities in developing countries. 329
330
331
Technical Options for Improving Food Production and Agricultural Sustainability 332
333
We discern in the dataset three types of technical improvement that have played 334
substantial roles in these food production increases: 335
i) more efficient water use in both dryland and irrigated farming; 336
337
ii) improvements to soil health and fertility; 338
339
iii) pest and weed control with minimum or zero-pesticide use. 340
341
342
i) More Efficient Use of Water 343
344
Improvements in the efficiency of water use can benefit both irrigated and rainfed 345
farmers by allowing new or formerly-degraded lands to be brought under farming, and 346
to increased cropping intensity on existing lands. In the projects analysed, water 347
harvesting has been widely applied in dryland areas. The Indo-British Rainfed Farming 348
project, for example, works with 230 local groups in 70 villages on water-harvesting, tree 349
planting, and grazing land improvements (Sodhi, 2001). Basic grain yields of rice, wheat, 350
pigeonpeas and sorghum have increased from 400 to 800-1000 kg ha-1, and the increased 351
fodder grass production from the terrace bunds is valued highly for the livestock. 352
Improved water retention has resulted in water tables rising by one metre over 3-4 years, 353
meaning that an extra crop is now possible for many farmers, thus turning an 354
unproductive season into a productive one. 355

9
356
Women are major beneficiaries. Sodhi (2001) puts it this way: “In these regions, women 357
never had seen themselves at the front edge of doing things, taking decisions, and dealing with 358
financial transactions. The learning by doing approach of the project has given them much needed 359
confidence, skills, importance and awareness.” The wider benefits of a transformed 360
agriculture are also evident, as “the project has indirectly affected migration as people are 361
gaining more income locally through the various enterprises carried out in the project. People are 362
now thinking that they must diversify more into new strategies. There has also been a decline in 363
drawing on resources from the forests”. Other projects in India have seen similar 364
environmental and social changes (Devavaram et al., 1999; Lobo and Palghadmal, 1999). 365
366
In Sub-Saharan Africa, water harvesting is also transforming barren lands. Again, the 367
technologies are not complex and costly. In central Burkina Faso, 130,000 hectares of 368
abandoned and degraded lands have been restored with the adoption of tassas and zaï. 369
These are 20-30 cm holes dug in soils that have been sealed by a surface layer hardened 370
by wind and water erosion. The holes are filled with manure to promote termite activity 371
and enhance infiltration. When it rains, water is channelled by simple stone bunds to the 372
holes, which fill with water, and into which are planted seeds of millet or sorghum. 373
Cereal yields in these regions rarely exceed 300 kg ha-1, yet these improved lands now 374
produce 700-1000 kg ha-1. Reij (1996) calculated that the average family in Burkina Faso 375
using these technologies had shifted from being in annual cereal deficit amounting to 650 376
kg, equivalent to six and a half months of food shortage, to producing a surplus of 150 kg 377
yr-1. Furthermore, tassas are best suited to landholdings where family labour is available, 378
or where farm labour can be hired, so that this soil and water conservation method has 379
led to a market for young day labourers who, rather than migrating, now earn money by 380
building these structures. 381
382
Good organisation also helps to improve irrigated agriculture. Despite great investment, 383
many irrigation systems have become inefficient and subject to persistent conflict. 384
Irrigation engineers assume that they know best how to distribute water, yet can never 385
know enough about the specific conditions and needs of large numbers of farmers. 386
Recent years, though, have seen the spread of programmes to organise farmers into water 387
users’ groups, and let them manage water distribution for themselves (Cernea, 1991; 388
Uphoff, 2002). One of the best examples comes from the Gal Oya region in Sri Lanka. 389
Before this approach, Gal Oya was the largest and most run-down scheme in the country. 390
Now, farmers’ groups manage water for 26,000 hectares of rice fields, and produce more 391
rice crops per year and per unit of water. Moreover, when farmers took control, the 392
number of complaints received by the Irrigation Department about water distribution fell to 393
nearly zero (Uphoff, 1999). The benefits were dramatically shown during the 1998 394
drought. According to government, there was only enough water for irrigation of 18% of 395
the rice area. But farmers persuaded the Irrigation Department to let this water through 396
on the grounds that they would carefully irrigate the whole area. Through cooperation 397
and careful management, they achieved a better than average harvest, earning the 398
country $20 million in foreign exchange. Throughout Sri Lanka, 33,000 water users’ 399
associations have now been formed – a dramatic increase in local social organisation that 400
has increased farmers’ own capacities for problem-solving and cooperation, and for using 401
nature more efficiently and effectively to produce more food. 402
403
404
405
ii) Improvements in Soil Health and Fertility 406
407

10
Soil health is fundamental for agricultural sustainability, yet is under widespread threat 408
from degradation processes (Cleaver and Schreiber, 1995; World Bank/FAO, 1996; 409
Smaling et al., 1997; Hinchcliffe et al., 1999; Petersen et al., 2000; Koohafkan and Stewart, 410
2001). Agricultural sustainability starts with the soil by seeking both to reduce soil 411
erosion and to make improvements to soil physical structure, organic matter content, 412
water-holding capacity and nutrient balances. Soil health is improved through the use of 413
legumes, green manures and cover crops, incorporation of plants with the capacity to 414
release phosphate from the soil into rotations, use of composts and animal manures, 415
adoption of zero-tillage, and use of inorganic fertilizers where needed (Reicosky et al., 416
1997; Sanchez and Jama, 2000). In projects in Central America, the incorporation of 417
nitrogen-fixing legumes into agroecosystems has substantially affected productivity, 418
particularly the velvetbean (Mucuna pruriens). This grows rapidly, fixes 150-200 kg N ha-1 419
yr-1, suppresses weeds, and can produce 35-50 tonnes of biomass ha-1 yr-1 (Bunch, 2000; 420
Anderson et al., 2001). Addition of this biomass to soils substantially improves soil 421
organic matter content, and has helped to increase cereal productivity for some 45,000 422
families in Guatemala, Honduras and Nicaragua. 423
424
In the past decade, Latin American farmers have found that eliminating tillage can be 425
highly beneficial for soils. After harvest, crop residues are left on the surface to protect 426
against erosion and, seed is directly planted into a groove cut into the soil. Weeds are 427
controlled with herbicides or cover crops. The fastest uptake of minimum till systems has 428
been in Brazil, where there are now 15 million hectares under plantio direto (also called 429
zero-tillage even though there is some disturbance of the soil) mostly in three southern 430
states of Santa Caterina, Rio Grande do Sul and Paraná, and in the central Cerrado. In 431
neighbouring Argentina, there are more than 11 million hectares under zero-tillage, up 432
from less than 100,000 hectares in 1990, and in Paraguay, another one million hectares of 433
zero-tillage (Sorrenson et al., 1998; Petersen et al., 1999; de Freitas, 1999; Peiretti, 2000; 434
Landers et al., 2001). 435
436
Elsewhere in Brazil, the transformations in the landscape and in farmers’ attitudes are 437
equally impressive. The Cerrado is a vast area of formerly unproductive lands colonised 438
for farming in the past two decades. These lands needed lime and phosphorus before 439
they could become productive, and now zero-tillage is being widely adopted (Landers et 440
al., 2001). In the early days, there was a widespread belief that zero-tillage was only for 441
large farmers. That has now changed, and small farmers are benefiting from technology 442
breakthroughs developed for mechanical farming. A core element of zero-tillage 443
adoption in South America has been adaptive research – working with farmers at 444
microcatchment level to ensure technologies are fitted well to local circumstances. There 445
are many types of farmers groups: from local (farmer micro-catchment and credit 446
groups), to municipal (soil commissions, Friends of Land clubs, commercial farmers and 447
farm workers’ unions), to multi-municipal (farmer foundations and cooperatives), to 448
river basin (basin committees for all water users), and to state and national level (state 449
zero-tillage associations and the national zero-tillage federation). 450
451
Farmers are now adapting technologies – organic matter levels have sufficiently 452
improved that fertilizer use has been reduced and rainfall infiltration improved, such that
453
some farmers are removing contour terraces. Other side-effects zero-tillage include 454
reduced siltation of reservoirs, less flooding, higher aquifer recharge, lowered costs of 455
water treatment, cleaner rivers, and more winter feed for wild biodiversity (Landers et al., 456
2001). However, there is still controversy over zero-tillage, as some feel the use of 457
herbicides to control weeds, or of genetically-modified crops, means we cannot call these 458

11
systems sustainable. However, the environmental benefits are substantial, and new 459
research is showing that farmers have some effective alternatives, particularly if they use 460
cover crops for green manures to raise organic matter levels. Using 20 species of cover 461
crops and green manures, Petersen and colleagues have shown how small farmers can 462
adopt zero-tillage systems without herbicides (Petersen et al., 2000; von der Weid, 2000). 463
464
A public good is also being created when soil health is improved with increased organic 465
matter. Soil organic matter contains carbon, and soils with above-ground biomass can act 466
as `carbon sinks’ or sites for carbon sequestration (Reicosky et al., 1995; Smith et al., 1998; 467
Sanchez et al., 1999; Watson et al., 2000; Pretty and Ball, 2001). Conservation tillage 468
systems and those using legumes and/or cover crops contribute to organic matter and 469
carbon accumulation in the soil. 470
471
In the Sahelian countries of Africa, the major constraints to food production are also 472
related to soils, most of which are sandy and low in organic matter. In Senegal, where soil 473
erosion and degradation threaten large areas of agricultural land, the Rodale Institute 474
Regenerative Agriculture Resource Center works closely with farmers’ associations and 475
government researchers to improve the quality of soils. The primary cropping system of 476
the region is a millet-groundnut rotation. Fields are cleared by burning, and then 477
cultivated with shallow tillage using animals. But fallow periods have decreased 478
dramatically, and inorganic fertilizers do not return high yields unless there are 479
concurrent improvements in organic matter, which helps to retain moisture. The Center 480
collaborates with 2000 farmers organised into 59 groups on improving soil quality by 481
integrating stall-fed livestock into crop systems, adding legumes and green manures, 482
increasing the use of manures, composts and rock phosphate, and developing water-483
harvesting systems. The result has been a 75%-190% improvement in millet and 484
groundnut yields – from about 300 to 600-900 kg ha-1. Yields are also less variable year on 485
year, with consequent improvements in household food security (Diop, 1999). 486
487
Thus if the soil is improved, the whole agricultural system’s health improves. Even if this 488
is done on a very small scale, people can benefit substantially. In Kenya, the Association 489
for Better Land Husbandry found that farmers who constructed double-dug beds in their 490
gardens could produce enough vegetables to see them through the hungry dry season. 491
These raised beds are improved with composts, and green and animal manures. A 492
considerable investment in labour is required, but the better water holding capacity and 493
higher organic matter means that these beds are both more productive and better able to 494
sustain vegetable growth through the dry season. Once this investment is made, little 495
more has to be done for the next two to three years. Women in particular are cultivating 496
many vegetable and fruit crops, including kale, onion, tomato, cabbage, passion fruit, 497
pigeon pea, spinach, pepper, green bean and soya. According to one review of 26 498
communities, 75% of participating households are now free from hunger during the year, 499
and the proportion having to buy vegetables had fallen from 85% to 11%. For too long, 500
agriculturalists have been sceptical about these organic and conservation methods. They 501
say they need too much labour, are too traditional, and have no impact on the rest of the 502
farm. Yet the spin-off benefits are substantial, as giving women the means to improve 503
their food production means that food gets into the mouths of children. They suffer fewer 504
months of hunger, and so are less likely to miss school (Hamilton, 1998). 505
506
507
iii) Pest Control with Minimal or Zero-Pesticide Use 508
509

12
Modern farmers have come to depend on a great variety of insecticides, herbicides and 510
fungicides to control pests, weeds and diseases, and each year, some 5 billion kg of 511
pesticide active ingredients are applied to farms (BAA, 2000). But farmers in these 512
projects have found many effective and more sustainable alternatives. In some crops, it 513
may mean the end of pesticides altogether, as cheaper and more environmentally-benign 514
practices are found to be effective. 515
516
Many projects in our survey reported large reductions in pesticide use in irrigated rice 517
systems. Following the discovery that pest attack on rice was proportional to pesticide 518
use (Kenmore et al., 1984), farmer field schools were later developed to teach farmers the 519
benefits of agro-biodiversity. In Indonesia, one million farmers have now attended 50,000 520
field schools, the largest number in any Asian country. In Vietnam, two million farmers 521
have cut pesticide use from more than 3 sprays to 1 per season; in Sri Lanka, 55,000 522
farmers have reduced use from 3 to 0.5 per season; and in Indonesia, one million farmers 523
have cut use from 3 sprays to 1 per season. In no case has reduced pesticide use led to 524
lower rice yields (Evelleens et al., 1996; Heong et al., 1998; Mangan and Mangan, 1998; 525
Desilles, 1999; Jones, 1999). Amongst these are reports that many farmers are now able to 526
grow rice entirely without pesticides: 25% of field school trained farmers in Indonesia, 20-527
33% in the Mekong Delta of Vietnam, and 75% in parts of the Philippines. 528
529
If pesticides are removed, then fish can be reintroduced. In Bangladesh, an aquaculture 530
and integrated pest management programme implemented by CARE has completed 6000 531
farmer field schools, resulting in 150,000 farmers adopting more sustainable methods of 532
rice production on about 50,000 hectares. The programme emphasises fish cultivation in 533
paddy fields, and vegetable cultivation on rice field dykes. Rice yields have improved by 534
5-7%, and costs of production have fallen owing to reduced pesticide use. In addition, 535
each hectare of paddy yields up to 750 kg of fish, a significant increase in system 536
productivity for poor farmers with few resources (Rashid, 2001). 537
538
In Kenya, intercropping of local legumes and grasses with maize has been found to 539
reduce stem borer (Chilo spp.) attack through interactions with parasitic wasps. 540
Researchers call their redesigned and diverse maize fields vutu sukuma (push-pull 541
systems). More than 2000 farmers in western Kenya have adopted maize, grass-strip and 542
legume-intercropping systems, and have increased maize yields by 60-70%. The official 543
advice to maize growers in the tropics has been to create monocultures for modern
544
varieties of maize, and then apply pesticide and fertilizers to make them productive. Yet 545
this very simplification eliminated vital and free pest management services produced by 546
the grasses and legumes. Vutu sukumu systems are complex and diverse, and are cheap as 547
they do not rely on costly purchased inputs (Khan et al., 2000). 548
549
Another project in Yunnan, China has shown the value of mixtures of rice, both in 550
reducing disease incidence and increasing yields (Zhu et al., 2000; Wolfe, 2000). 551
Researchers working in ten townships on 5350 hectares encouraged farmers to switch 552
from growing monocultures of sticky rice to alternating rows of sticky rice with hybrids. 553
The sticky rice brings a higher price, but is susceptible to rice blast, which is generally 554
controlled through applications of fungicides. But planting mixtures in the same field 555
reduced blast incidence by 94% and increased total yields by 89%. By the end of two 556
years, it was concluded that fungicides were no longer required. 557
558
559
Impacts on Rural Livelihoods and Economies 560

13
561
Rural people’s livelihoods rely for their success on the value of services flowing from the 562
total stock of natural, social, human, physical and financial capital (Coleman, 1990; 563
Putnam, 1993; Costanza et al., 1997; Carney, 1998; Scoones, 1998; Pretty and Ward, 2001). 564
A number of examples can be extracted from the dataset to show that agricultural 565
sustainability projects and initiatives have been able to contribute to the accumulation of 566
locally-valuable assets. A selection of the impacts reported in these sustainable 567
agriculture projects and initiatives include: 568
i) improvements to natural capital, including increased water retention in soils, 569
improvements in water table (with more drinking water in the dry season), reduced soil 570
erosion combined with improved organic matter in soils, leading to better carbon 571
sequestration, and increased agro-biodiversity (cf Hinchcliffe et al., 1999; Watson et al, 572
2000; McNeely and Scherr, 2001; Pretty and Ball, 2001); 573
ii) improvements to social capital, including more and stronger social organisations 574
at local level, new rules and norms for managing collective natural resources, and better 575
connectedness to external policy institutions (cf Uphoff, 1999; Pretty and Ward, 2001); 576
iii) improvements to human capital, including more local capacity to experiment and 577
solve own problems; reduced incidence of malaria in rice-fish zones, increased self-578
esteem in formerly marginalised groups, increased status of women, better child health 579
and nutrition, especially in dry seasons, and reversed migration and more local 580
employment (cf Li Kangmin, 1998; Shah and Shah, 1999; Bunch, 2000; Regasamy et al., 581
2000). 582
583
The empirical evidence indicates that some improvements in agricultural sustainability 584
have had positive effects on regional economies. In the Ansokia Valley, Ethiopia, one 585
project increased annual food production from 5600 to 8370 tonnes in six years, at the 586
same time as the population increased from 36,000 to 45,000. The project turned around 587
an annual food regional deficit of -2106 tonnes to a surplus of 372 tonnes year-1. In 588
Bushenyi, Uganda, formerly experiencing substantial food shortages during the months 589
of October to December, one project so increased banana and cattle production that the 590
region could sells 330 tonnes bananas and 2.7 tonne of meat each week. In En Nahud, 591
Sudan, the 10,000 tonnes of additional food produced by 15,000 households were 592
consumed by local people. None found its way into national statistics. 593
594
There is also evidence that productivity can increase over time as natural and human 595
capital assets increase. If agricultural systems are low in capital assets (either intrinsically 596
low, or have become low because of degradation), then a sudden switch to `more 597
sustainable’ practices that have to rely on these assets will not be immediately successful. 598
In Cuba, for example, urban organic gardens produced 4200 tonnes of food in 1994. By 599
1999, they had greatly increased in per area productivity – rising from 1.6 kg m-2 to 19.6 600
kg m-2 (Murphy, 1999; Funes, 2001). Increasing productivity over time has also been 601
noted in fish-ponds in Malawi. These are typically some 200-500 m2 in size. Researchers 602
compared the performance of 35 fish-ponds over six years: in 1990 yields were 800 kg/ha, 603
but rose steadily to 1450 kg/ha by 1996. This is because fish-ponds are integrated into a 604
farm so that they recycle wastes from other agricultural and household enterprises, 605
leading to steadily increasing productivity over time as farmers themselves gain 606
understanding (Brummet, 2000). 607
608
609
Confounding Factors and Trade-Offs 610
611

14
What we do not yet know is whether moving to more sustainable systems, delivering 612
greater benefits at the scale occurring in these projects, will result in enough food to meet 613
the current food needs in developing countries, let alone the future needs after continued 614
population growth and adoption of more urban and meat-rich diets (Delgado et al., 1999). 615
But what we are seeing should be cause for cautious optimism, particularly as evidence 616
indicates that productivity can grow over time if natural, social and human assets are 617
accumulated (see also McNeely and Scherr, 2001). 618
619
A more sustainable agriculture which improves the asset base can lead to rural livelihood 620
improvements. People can be better off, have more food, be better organised, have access 621
to external services and power structures, and have more choices in their lives. But like all 622
major changes, such transitions can also provoke secondary problems. For example, 623
building a road near a forest can help farmers reach food markets, but also aid illegal 624
timber extraction. Projects may be making considerable progress on reducing soil erosion 625
and increasing water conservation through adoption of zero-tillage, but still continue to 626
rely on applications of herbicides. If land has to be closed off to grazing for rehabilitation, 627
then people with no other source of feed may have to sell their livestock; and if cropping 628
intensity increases or new lands are taken into cultivation, then the burden of increased 629
workloads may fall particularly on women. Also additional incomes arising from sales of 630
produce may go directly to men in households, who are less likely than women to invest 631
in children and the household as a whole. 632
633
There are also a variety of emergent factors that could slow the spread of agricultural 634
sustainability. First, practices that increase the asset base may simply increase the 635
incentives for more powerful interests to take over, such as landlords taking back 636
formerly degraded land from tenants who had adopted more sustainable agriculture. In 637
these contexts, it is rational for farmers to farm badly – at least they get to keep the land. 638
The idea of sustainable agriculture may also appear to be keeping people in rural areas 639
away from centres of power and `modern’ urban society, yet some rural people’s 640
aspirations may precisely to be to gain sufficient resources to leave rural areas. 641
Agricultural sustainability also implies a limited role for agro-chemical companies, who 642
would not be predicted to accept such losses of market lightly. It also suggests greater 643
decentralisation of power to local communities and groups, combined with more local 644
decision-making, both of which might be opposed by those who would benefit from 645
corruption and non-transparency in private and public organisations. Research and 646
extension agencies will have to change too, adopting more participatory approaches to
647
work closely with farmers, and so must adopt different measures for evaluating job 648
success and the means to promotion. Finally, social connectivity, relations of trust, and 649
the emergence of significant movements may present a threat to existing power bases, 650
who in turn may seek to undermine such locally-based institutions. 651
652
Further tensions arise over the balance between whether food production is more 653
sustainable if for local markets alone, or whether poorer farmers and communities should 654
be encouraged to access international markets, but with the result that food causes 655
greater transport externalities through long-distance travel. Moreover, farms with 656
increased productivity export increasingly large amounts of nutrients to be eaten 657
elsewhere, and it will be vital to ensure that replacement occurs at sustainable rates. 658
There will be some who dispute this evidence of promising successes, believing that the 659
poor and marginalised cannot possibly make these kinds of improvements. But we 660
believe there is hope and leadership in this evidence of progress towards sustainability. 661
What is quite clear is that these technologies and practices offer real opportunities for 662

15
people to improve their food production whilst protecting and improving nature. 663
664
665
Scaling Up through Appropriate Policies 666
667
Three things are now clear from this dataset about spreading agricultural sustainability: 668
i) some technologies and social processes for local scale adoption of more 669
sustainable agricultural practices are well-tested and established; 670
ii) The social and institutional conditions for spread are less well-known, but have 671
been established in several contexts, leading to very rapid spread in the 1990s; 672
iii) The political conditions for the emergence of supportive policies are least well 673
established, with only a very few examples of real progress. 674
675
As has been indicated earlier, agricultural sustainability can contribute to increased food 676
production, as well as make an impact on rural people’s welfare and livelihoods. Clearly 677
much can be done with existing resources. A transition towards a more sustainable 678
agriculture will not, however, happen without some external help and money. There are 679
always transition costs in learning, in developing new or adapting old technologies, in 680
learning to work together, and in breaking free from existing patterns of thought and 681
practice. It also costs time and money to rebuild depleted natural and social capital. 682
683
Most agricultural sustainability improvements seen in the 1990s arose despite existing 684
national and institutional policies, rather than because of them (Pretty, 1999; Pretty et al., 685
2001). Nonetheless, the 1990s have seen considerable global progress towards the 686
recognition of the need for policies to support sustainable agriculture. Although almost 687
every country would now say it supports the idea of agricultural sustainability, the 688
evidence points towards only patchy reforms. Only two countries, Cuba and Switzerland, 689
have given explicit national support for a transition towards sustainable agriculture – 690
putting it at the centre of agricultural development policy and integrating policies 691
accordingly. Cuba has a national policy for alternative agriculture; and Switzerland has 692
three tiers of support for practices contributing to agriculture and rural sustainability 693
(Funes, 2001; Swiss Agency for Environment, Forests and Landscape, 1999, 2000). Several 694
countries have given sub-regional support, such as the states of Santa Caterina, Paraná and 695
Rio Grande do Sul in southern Brazil supporting zero-tillage and catchment management, 696
and some states in India supporting participatory watershed and irrigation management. 697
A larger number have reformed parts of agricultural policies, such as China’s support for 698
integrated ecological demonstration villages, Kenya’s catchment approach to soil 699
conservation, Indonesia’s ban on pesticides and programme for farmer field schools, 700
Bolivia’s regional integration of agricultural and rural policies, Sweden’s support for 701
organic agriculture, Burkina Faso’s land policy, and Sri Lanka and the Philippines’ 702
stipulation that water users’ groups manage irrigation systems (Pretty, 2002). 703
704
A good example of a carefully designed and integrated programme comes from China. In 705
March 1994, the government published a White Paper to set out its plan for implementation 706
of Agenda 21, and put forward ecological farming, known as Shengtai Nongye or agro-707
ecological engineering, as the approach to achieve sustainability in agriculture. Pilot 708
projects have been established in 2000 townships and villages spread across 150 counties. 709
Policy for these `eco-counties’ is organised through a cross-ministry partnership, which 710
uses a variety of incentives to encourage adoption of diverse production systems to replace 711
monocultures. These include subsidies and loans, technical assistance, tax exemptions and 712
deductions, security of land tenure, marketing services and linkages to research 713

16
organisations. These eco-counties contain some 12 million hectares of land, about half of 714
which is cropland, and though only covering a relatively small part of China’s total 715
agricultural land, do illustrate what is possible when policy is appropriately coordinated (Li 716
Wenhua, 2001). 717
718
An even larger set of countries has seen some progress on agricultural sustainability at 719
project and programme level. However, progress on the ground still remains largely 720
despite, rather than because of, explicit policy support. No agriculture minister is likely to 721
say they are against sustainable agriculture, yet good words remain to be translated into 722
comprehensive policy reforms. Agricultural systems can be economically, 723
environmentally and socially sustainable, and contribute positively to local livelihoods. 724
But without appropriate policy support, they are likely to remain at best localised in extent, 725
and at worst simply wither away. 726
727
728
Conclusions 729
730
This empirical study shows that there have been promising advances in the adoption and 731
spread of more sustainable agriculture. The 208 projects/initiatives show increases in 732
food production over some 29 million hectares, with nearly 9 million households 733
benefiting from increased food production and consumption. These increases are not yet 734
making a significant mark on national statistics, as we believe there is a significant 735
elasticity of food consumption in many poor rural households. They are eating the 736
increased food produced, or marketing small surpluses to other local people. We cannot, 737
therefore, yet say whether a transition to more sustainable agriculture, delivering 738
increasing benefits at the scale occurring in these projects, will result in enough food to 739
meet the current food needs of developing countries, the future basic needs after 740
continued population growth, or the potential demand following adoption of more meat-741
rich diets. Even the substantial increases reported here may not be enough. There should 742
be cautious optimism, as the evidence indicates that productivity can increase steadily 743
over time if natural, social and human capital assets are accumulated. 744
745
Increased agricultural sustainability can also be complementary to improvements in rural 746
people’s livelihoods. It can deliver increases in food production at relatively low cost, 747
plus contribute to other important functions. Were these approaches to be widely 748
adopted, they would make a significant impact on rural people’s livelihoods, as well as 749
on local and regional food security. But there are clearly major constraints to overcome.
750
There will be losers along with winners, and some of the losers are currently powerful 751
players. And yet, social organisation and mobilisation in a number of contexts is already 752
leading to new informal and formal alliances that are protecting existing progress and 753
developing the conditions for greater spread. Improving agricultural sustainability 754
clearly will not bring all the solutions, but promising progress has been made in recent 755
years. With further explicit support, particularly through international, national and local 756
policy reforms, these benefits to food security and attendant improvements to natural, 757
social and human capital could spread to much larger numbers of farmers and rural 758
people in the coming decades. 759

17
Acknowledgements 760
761
We are grateful to three organisations for their support for this research: the UK 762
Department for International Development (DFID), Bread for the World (Germany), and 763
Greenpeace (Germany). We are grateful to Thomas Dobbs, Per Pinstrup-Andersen, 764
Hiltrud Nieberg, Roland Bunch and Vo-Tung Xuan for substantive comments on the 765
project report, together with feedback from participants at the St James’s Palace, London 766
2001 conference on `Reducing Poverty with Sustainable Agriculture’. Two anonymous 767
referees gave additional useful comments. We are indebted to 358 people directly 768
associated with sustainable agriculture projects who have given their valuable time to 769
send material, to complete questionnaires, to verify findings, and to advise on the wider 770
project. In the final project report, we thank them all by name (At URL 771
www2.essex.ac.uk/ces). 772
773
References 774
775
ACC/SCN. 2000. 4th report on The World Nutrition Situation. UN Administrative Committee on Coordination, 776
Sub-Committee on Nutrition. In collaboration with IFPRI. United Nations, NY 777
Altieri, M. 1995. Agroecology: The Science of Sustainable Agriculture. Westview Press, Boulder 778
Anderson, S., Gündel, S. and Pound, B. 2001. Cover Crops in Smallholder Agriculture: Lessons from Latin America. 779
IT Publications, London 780
Avery D. 1995. Saving the Planet with Pesticides and Plastic. The Hudson Institute, Indianapolis 781
BAA. 2000. Annual Review and Handbook. British Agrochemicals Association, Peterborough 782
Balfour, E.B. 1943. The Living Soil. Faber and Faber, London 783
Brummet, R. 2000. Integrated aquaculture in Sub-Saharan Africa. Environ. Develop. & Sust. 1 (3-4), 315-321 784
Bunch, R. Two Ears of Corn. World Neighbors, Oklahoma City 785
Bunch, R. 2000. More productivity with fewer external inputs. Environ. Develop. & Sust. 1 (3-4), 219-233 786
Bunch R and López G. 1996. Soil recuperation in Central America: sustaining innovation after intervention. 787
Gatekeeper Series SA 55, International Institute for Environment and Development, London 788
Carney, D. 1998. Sustainable Rural Livelihoods. Department for International Development, London 789
Carson, R. 1963. Silent Spring. Penguin Books, Harmondsworth 790
Carter J. 1995. Alley Cropping: Have Resource Poor Farmers Benefited? ODI Natural Resource Perspectives No 3, 791
London 792
Cernea M M. 1991. Putting People First. Oxford University Press, Oxford. 2nd Edition. 793
Chambers R. 1983. Rural Development: Putting the Last First. Longman, London 794
Cleaver K M and Schreiber G A. 1995. The population, agriculture and environment nexus in Sub-Saharan Africa. 795
World Bank, Washington DC 796
Coleman, J. 1990. Foundations of Social Theory. Harvard University Press, Mass. 797
Conway, G.R. 1997. The Doubly Green Revolution. Penguin, London 798
Conway, G.R. and Pretty, J.N. 1991. Unwelcome Harvest: Agriculture and Pollution. Earthscan, London 799
Costanza, R., d’Arge, R., de Groot, R., Farber, S., Grasso, M., Hannon, B., Limburg, K., Naeem, S., O’Neil, 800
R.V., Parvelo, J., Raskin, R.G., Sutton, P. and van den Belt, M. 1997. The value of the world’s 801
ecosystem services and natural capital. Nature 387, 253-260 [also in Ecol. Econ. 25 (1), 3-15, 1999] 802
de Freitas, H. 1999. Transforming microcatchments in Santa Caterina, Brazil. In Hinchcliffe, F., Thompson, J., 803
Pretty, J., Guijt, I. and Shah, P. (eds). Fertile Ground: The Impacts of Participatory Watershed 804
Development. IT Publications, London 805
Delgado, C., Rosegrant, M., Steinfield, H., Ehui, S. and Courbois, C. 1999. Livestock to 2020: the next food 806
revolution. IFPRI Brief 61. International Food Policy Research Institute, Washington DC 807
Desilles, S. 1999. Sustaining and managing private natural resources: the way to step out of the cycle of high-808
input agriculture. Paper for Conference on "Sustainable Agriculture: New Paradigms and Old Practices?", 809
Bellagio Conference Center, Italy, April 26th-30th, 1999 810
Devavaram, J., Arunothayam, E., Prasad, R. and Pretty, J. 1999. Watershed and community development in 811
Tamil Nadu, India. In Hinchcliffe, F., Thompson, J., Pretty, J., Guijt, I. and Shah, P. (eds). Fertile 812
Ground: The Impacts of Participatory Watershed Development. IT Publications, London 813
Diop, A. 1999. Sustainable agriculture: new paradigms and old practices? Increasing production with 814
management of organic inputs in Senegal. Environ. Develop. & Sust. 1 (3-4), 285-296 815
E.E.A. 1998. Europe’s Environment: The Second Assessment. Report and Statistical Compendium. European 816
Environment Agency, Copenhagen 817

18
Eveleens, K.G., Chisholm, R., van der Fliert, E., Kato, M., Nhat, P.T. and Schmidt, P. 1996. Mid Term Review
818
of Phase III Report. The FAO Intercountry Programme for the Development and Application of 819
Integrated Pest Control in Rice in South and Southeast Asia. FAO, Manila and Rome 820
F.A.O. 1999. Cultivating Our Futures: Taking Stock of the Multifunctional Character of Agriculture and Land. Rome 821
F.A.O. 2000. Agriculture: Towards 2015/30. Global Perspective Studies Unit, F.A.O, Rome 822
Funes, F. 2001. Cuba and sustainable agriculture. Paper presented to St James’s Palace conference Reducing 823
Poverty with Sustainable Agriculture, 15th January. University of Essex, Colchester 824
Hamilton P. 1998. Goodbye to Hunger: A study of farmers’ perceptions of conservation farming ABLH Nairobi, 825
Kenya 826
Heong, K.L., Escalada, M.M., Huan, N.H. and Mai, V. 1998. Use of communication media in changing rice 827
farmers’ pest management in the Mekong Delta, Vietnam. Crop Management 17 (5), 413-425 828
Hinchcliffe, F., Thompson, J. and Pretty, J.N. 1996. Sustainable Agriculture and Food Security in East and 829
Southern Africa. Report for the Committee on Food Security in East and Southern Africa, Swedish 830
International Agency for International Cooperation, Stockholm. 831
Hinchcliffe, F., Thompson, J., Pretty, J., Guijt, I. and Shah, P. (eds). 1999. Fertile Ground: The Impacts of 832
Participatory Watershed Development. IT Publications, London 833
Jones, K. 1999. Integrated pest and crop management in Sri Lanka. Paper for Conference on "Sustainable 834
Agriculture: New Paradigms and Old Practices?", Bellagio Conference Center, Italy, April 26th-30th, 1999 835
Kenmore, P.E, Carino, F.O., Perez, C.A., Dyck, V.A. and Gutierrez, A.P. 1984. Population regulation of the 836
brown planthopper within rice fields in the Philippines. J. Plant Protection in the Tropics 1(1), 19-37 837
Khan, Z.R., Pickett, J.A., van den Berg, J. and Woodcock, C.M. 2000. Exploiting chemical ecology and species 838
diversity: stem borer and Striga control for maize in Africa. Pest Management Science 56 (1), 1-6 839
Koohafkan, P. and Stewart, B.A. 2001. Water Conservation and Water Harvesting in Cereal-Producing 840
Regions of the Drylands. FAO, Rome 841
Landers J N, De C Barros G S-A, Manfrinato W A Rocha M T and Weiss J S. 2001. Environmental benefits of 842
zero-tillage in Brazil – a first approximation. In Garcia Torres L, Benites J and MartinezVilela A 843
(eds). Conservation Agriculture – A Worldwide Challenge Volume 1. XUL, Cordoba, Spain 844
Li Kangmin. 1998. Rice aquaculture systems in China. In Eng-Leong Foo and Tarcision Della Senta (eds). 845
Integrated Bio-Systems in Zero Emissions Applications. Proceedings of an internet conference on 846
integrated biosystems. http://www.ias.unu.edu/proceedings/icibs 847
Li Wenhua. 2001. Agro-Ecological Farming Systems in China. Man and the Biosphere Series Volume 26. 848
UNESCO, Paris 849
Lobo C and Palghadmal T. 1999. Kasare: A saga of peoples faith. In Hinchcliffe et al. (eds). Fertile Ground. IT 850
Publ., London 851
Mangan, J. and Mangan, M.S. 1998. A comparison of two IPM training strategies in China: the importance of 852
concepts of the rice ecosystem for sustainable pest management. Agric. Human Values 15, 209-221 853
McNeely J A and Scherr S J. 2001. Common Ground, Common Future. How ecoagriculture can help feed the world 854
and save wild biodiversity. IUCN and Future Harvest, Geneva 855
Murphy, B. 1999. Cultivating Havana: Urban agriculture and food security in Cuba. Food First Development 856
Report 12. Food First, California 857
Norse, D., Li Ji and Zhang Zheng. 2000. Environmental Costs of Rice Production in China: Lessons from Hunan and 858
Hubei. Aileen Press, Bethesda 859
N.R.C. 2000. Our Common Journey: Transition towards sustainability. Board on Sustainable development, Policy 860
Division, National Research Council. National Academy Press, Washington DC 861
Peiretti, R. 2000. The evolution of the No Till cropping system in Argentina. Paper presented to “Impact of 862
Globalisation and Information on the Rural Environment”, Jan 13-15th, Harvard, Cambridge, Mass. 863
Petersen, P., Rardin J.M. and Marochi, F. 1999. Participatory development of no-tillage systems without 864
herbicides for family farming. Environ., Dev. & Sust., 1 (3-4) 235-252 865
Petersen, C., Drinkwater, L.E. and Wagoner, P. 2000. The Rodale Institute’s Farming Systems Trial. The First 15 866
Years. Rodale Institute, Penn. 867
Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kunz, D., McNair, M., Crist, S., Shpritz, L., Fitton, L., 868
Saffouri, R. and Blair, R. 1995. Environmental and economic costs soil erosion and conservation 869
benefits. Science 267, 1117-1123 870
Pingali, P.L. and Roger, P.A. 1995. Impact of Pesticides on Farmers' Health and the Rice Environment. Kluwer 871
Academic Press, Dordrecht 872
Pinstrup-Anderson, P., Pandya-Lorch, R. and Rosegrant, M. 1999. World Food Prospects: Critical Issues for the 873
Early 21st Century. IFPRI, Washington DC 874
Popkin, B. 1998. The nutrition transition and its health implications in lower-income countries. Public Health 875
Nutrition 1(1), 5-21 876
Pretty, J.N. 1995. Regenerating Agriculture: Policies and Practice for Sustainability and Self-Reliance. Earthscan 877
Publications, London; National Academy Press, Washington DC; ActionAid, Bangalore 878

19
Pretty, J.N. 1998. The Living Land: Agriculture, Food Systems and Community Regeneration in Rural Europe.
879
Earthscan Publications Ltd, London 880
Pretty, J.N. 1999. Sustainable Agriculture: A Review of Recent Progress on Policies and Practice. United Nations 881
Research Institute for Social Development (UNRISD), Geneva 882
Pretty, J.N. 2000. Can sustainable agriculture feed Africa? Environ. Dev. & Sust. 1, 253-274 883
Pretty, J.N. 2002. Agri-Culture: Communities Shaping Land and Nature. Earthscan, London (in press) 884
Pretty, J.N. and Ball, A. 2001. Agricultural Influences on Carbon Emissions and Sequestration: A Review of Evidence 885
and the Emerging Trading Options. Centre for Environment and Society Occasional Paper 2001-3, 886
University of Essex 887
Pretty, J.N. and Hine, R. 2000. The promising spread of sustainable agriculture in Asia. Natural Resources Forum 888
(UN) 24, 107-121 889
Pretty, J.N. and Ward, H. 2001. Social capital and the environment. World Development 29 (2), 209-227 890
Pretty, J.N., Brett, C., Gee, D., Hine, R., Mason, C.F., Morison, J.I.L., Raven, H., Rayment, M. and van der Bijl, 891
G. 2000. An assessment of the total external costs of UK agriculture. Agric. Syst. 65 (2), 113-136 892
Pretty, J.N., Brett, C., Gee, D., Hine, R., Mason, C., Morison, J., Rayment, M., van der Bijl, G. and Dobbs, T. 893
2001. Policy challenges and priorities for internalising the externalities of modern agriculture. J. 894
Environ. Planning & Manage. 44 (2), 263-283 895
Putnam, R.D., with Leonardi, R. and Nanetti, R.Y. 1993. Making Democracy Work: Civic Traditions in Modern 896
Italy. Princeton University Press, Princeton New Jersey 897
Rashid A. 2001. The CARE Interfish projects in Bangladesh. Paper presented to St James’s Palace conference 898
Reducing Poverty with Sustainable Agriculture, 15th January. University of Essex, Colchester 899
Reicosky, D.C., Kemper, W. D., Langdale, G.W., Douglas, C.L. and Rasmussen, P.E. 1995. Soil organic matter 900
changes resulting from tillage and biomass production. J. Soil and Water Conservation 50 (3), 253-261 901
Reicosky, D.C., Dugas, W.A. and Torbert, H.A. 1997. Tillage-induced soil carbon dioxide loss from different 902
cropping systems. Soil and Tillage Research 41, 105-118 903
Reij C. 1996. Evolution et impacts des techiques de conservation des eaux et des sols. Centre for Development 904
Cooperation Services, Vrije Univeriseit, Amsterdam 905
Rengasamy S, Devavaram J, Prasad R, Erskine A, Balamurugan P and High C. 2000. The Land Without a 906
Farmer Becomes Barren (thaan vuzhu nilam thariso). SPEECH, Ezhil Nagar, Madurai, India 907
Sanchez, P.A., Buresh, R.J. and Leakey, R.R.B. 1999. Trees, soils and food security. Philosophical Trans. Roy Soc 908
London B 253, 949-961 909
Sanchez P A and Jama B A. 2000. Soil fertility replenishment takes off in east Southern Africa. Paper for 910
International Symposium on Balanced Nutrient Management Systems for the Moist Savanna and 911
Humid Forest Zones of Africa. Cotonou, Benin. October 9, 2000 912
Scoones, I. 1998. Sustainable Rural Livelihoods: A Framework for Analysis. IDS Discussion Paper, 72, Univ. of 913
Sussex, Brighton 914
Shah P and Shah M K. 1999. Institutional strengthening for watershed development: the case of the AKRSP in 915
India. In Hinchcliffe F, Thompson J, Pretty J, Guijt I and Shah P. (eds). Fertile Ground: The Impacts of 916
Participatory Watershed Development. IT Publications, London 917
Smaling E M A, Nandwa S M and Janssen B H. 1997. Soil fertility in Africa is at stake. In: Buresh R J, Sanchez 918
P A and Calhoun F (eds). Replenishing Soil fertility in Africa. Soil Science Society of America 919
Publication No 51. SSSA, Madison, Wisconsin 920
Smil, V. 2000. Feeding the World: A Challenge for the 21st Century. MIT Press, Cambridge, MA 921
Smith, L.C. and Haddad, L. 1999. Explaining Child Malnutrition in Developing Countries: A cross-country 922
analysis. Research Report 111 (March 2000), IFPRI, Washington DC 923
Smith, P., Powlson, D.S., Glendenning, M.J. and Smith, J.U. 1998. Preliminary estimates of the potential for 924
carbon mitigation in European soils through no-till farming. Global Change Biology 4, 679-685 925
Sodhi P.S. 2001. Livelihood improvements in the KRIBCHO project. Paper presented to St James’s Palace 926
conference Reducing Poverty with Sustainable Agriculture, 15th January. University of Essex, Colchester 927
Sorrenson, W. .J, Duarte, C. and Portillo, J.L. 1998. Economics of no-till compared to conventional systems on 928
small farms in Paraguay. Soil Conservation Project MAG-GTZ, Eschborn, Germany 929
Steiner, R., McLaughlin, L., Faeth, P. and Janke, R. 1995. Incorporating externality costs in productivity 930
measures: a case study using US agriculture. In Barbett, V., Payne, .R and Steiner, R (eds). 931
Agricultural Sustainability: Environmental and Statistical Considerations. Wiley, New York, p 209-230 932
Swiss Agency for Environment, Forests and Landscape. 1999. The Environment in Switzerland: Agriculture, 933
Forestry, Fisheries and Hunting. Berne, Switzerland. 934
Swiss Agency for Environment, Forests and Landscape and Federal Office of Agriculture. 2000. Swiss 935
agriculture on its way to sustainability. SAEFL and FOA, Basel. 936
Thrupp, L.A. 1996. Partnerships for Sustainable Agriculture. World Resources Institute, Washington DC 937
U.N. 1999. World Population Prospects – The 1998 Revision. United Nations, NY 938

20
Uphoff, N. 1999. Understanding social capital: learning from the analysis and experience of participation. In
939
Dasgupta P. and Serageldin I. (eds). Social Capital: A Multiperspective Approach. Washington DC: 940
World Bank 941
Uphoff N (ed). 2002. Agroecological Innovations. Earthscan, London 942
von der Weid J M. 2000. Scaling up and Scaling Further Up. AS-PTA, Rio de Janeiro 943
Waibel, H. and Fleischer., G. 1998. Kosten und Nutzen des chemischen Pflanzenschutzes in der Deutsen 944
Landwirtschaft aus Gesamtwirtschaftlicher Sicht. Vauk-Verlag, Kiel 945
Watson, R.T., Noble, I.R., Bolin, B., Ravindranath, N.H., Verardo, D.J. and Dokken, D.J. (eds). 2000. IPCC 946
Special Report on Land Use, Land-Use Change and Forestry. Approved at Intergovernmental Panel 947
on Climate Change (IPCC) Plenary XVI (Montreal, 1-8 May, 2000). IPCC Secretariat, c/o World 948
Meteorological Organisation, Geneva. 949
Wood, S., Sebastien, K. and Scherr, S.J. 2000. Pilot Analysis of Global Ecosystems. IFPRI and WRI, Washington 950
World Bank/F.A.O. 1996. Recapitalisation of Soil Productivity in Sub-Saharan Africa. Washington D.C. & Rome 951
Wolfe, M. 2000. Crop strength through diversity. Nature 406, 681-682 952
Zhu, Y., Chen, H., Fen, J., Wang, Y., Li Y., Chen, J., Fan, J., Yang, S., Hu, L., Leaung, H., Meng, T.W., Teng, 953
A.S., Wang, Z. and Mundt, C.C. 2000. Genetic diversity and disease control in rice. Nature 406, 718-954
722 955

21
956
Figure 1a. Cumulative proportion of farmers by project size according to region (dotted 957
horizontal lines represent 50%, 75% and 95% of total farmers) 958
959
0.0
0.2
0.4
0.6
0.8
1.0
0.001 0.01 0.1 1 10 100 1000 10000
average area/farmer (ha) per project
proportion of total farmers
Latin America
Asia
Africa
960

22
Figure 1b. Cumulative proportion of total area by project size according to region 961
(dotted horizontal lines represent 50%, 75% and 95% of total area) 962
963
0.0
0.2
0.4
0.6
0.8
1.0
0.001 0.01 0.1 1 10 100 1000 10000
average area/farmer (ha) per project
proportion of total area
Latin America
Asia
Africa
964

23
Figure 2. Frequency of occurrence of each type of improvement mechanism by projects, 965
farmers and area (data from SAFE-World research project) 966
967
0 20406080100
Intensification of
single component
(i)
New productive
element (ii)
Better use of water
and land (iii)
Per ha food crop
yield improvements
(iv)
Other (plantation
and fibre crops)
Mechanisms
Proportion of total (%)
Projects (%)
Farmers (%)
Area (%)

24
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
0 2000 4000 6000 8000 10000
yields before/without project (kg/ha)
relative yield change after or with project
maize
sorghum/millet
beans/soya/peas/groundnut
rice
wheat
potato/sweet pot/cassava
cotton
vegetables
no change
Figure 3. Relative crop yield changes in agricultural sustainability projects/initiatives 968
(89 projects) (data from SAFE-World research project) 969
970

25
Figure 4. Relative change in yield grouped by crop type (mean and +/- s.e.m) (data from 971
SAFE-World research project) 972
973
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
maize
wheat
rice
sorghum/millet
vegetables
beans, soya, peas
roots
cereals
grand total
relative yield change after/with projec
t

26
Figure 5. Change in annual household food production with sustainable agricultural 974
practices and technologies (data from SAFE-World research project) 975
976
977
0
2
4
6
8
10
012345
hectares per household
marginal increase in household
.
food production (t/yr)