8.3.1. System monitoring
8.3.2 Research priorities
8.3.3 Other research implications
Efficient implementation of the development strategies described herewith require that some routine data be collected. Hypotheses described in Section 7.2: A theory of local system dynamics can form the framework of testable ideas within which to collect information. Development agents need to have an even better understanding of cattle population dynamics, status of the grazing reserves, extent of cultivation and the problems the Boran have with managing grazing and water resources. Given that operating funds are commonly very limited, monitoring of trends should involve qualitative as well as quantitative methods. Quantitative monitoring could involve use of aerial survey once every few years, if research budgets allow. Alternatively, or as a supplement to aerial survey, road networks could be driven during important times of the year to provide objective assessment of land-use trends on a year-to-year basis. Quantitative monitoring could also include routine surveys of the volume of local markets, diversity of commodities sold and their prices. Qualitative monitoring, in contrast, could include having development agents attend important community meetings and record felt needs of the
Boran to see how these change over time. In all cases data should be collected to help development agents understand why certain trends are occurring.
By the late 1980s SORDU was focused on monitoring and evaluating range trend (Hacker, 1988a,b). The theory of local system dynamics forwarded in Chapter 7 has great implications for the interpretation of range trend survey results. If cattle stocking rate has the greatest effect on range utilisation, then it would be expected that the long-term pattern for basal cover and standing crop of herbaceous species would appear cyclic, rather than linear, despite variation caused by annual fluctuation in rainfall. During the drought-recovery phase of the early 1990s, basal cover and standing crop of herbaceous plants might have higher values compared to those collected in the high-density phase of the mid- to late-1990s. Measures of these attributes could increase again during the subsequent drought-recovery phase. Patterns would also be greatly complicated by variation in rainfall during either phase.
In sum, definite trend could be exceedingly difficult to detect; this is especially true if sampling precision is poor. To conserve operating resources, it could be useful to restrict data collection to high-density phases only. Monitoring bush encroachment is a bit more straightforward, but may also be subject to problems of interpretation. The theory of system dynamics predicts that establishment of woody seedlings is not a gradual phenomenon, but is more episodic in nature. Some years have an explosive increase in seedlings while in many other years establishment is negligible. The greatest rate of seedling establishment should occur during years of above-average rainfall in the high-density phase of the cattle population, with other ecological factors held relatively constant (Section 7.2.3.1: Range ecology).
The framework proposed in Section 7.2: A theory of local system dynamics is essentially a series of hypotheses concerning important interactions. These require testing using interdisciplinary methods. There may not be scope to do so in the forseeable future, however, given other urgent priorities for collecting information to help mitigate the effects of the current crisis in the southern rangelands.
It is forwarded that the most important research questions in the southern rangelands today involve social science and economics. In particular, it is important to know how the human population is coping with problems that have arisen from high population growth and perennial drought. For example, little is known concerning: (1) detailed dynamics of human population growth, including effects of changing social values and outside inputs on rates of birth and mortality; (2) whether there have been recent changes in settlement patterns and their implications for resource use; (3) who is emigrating out of the system, why and what happens to emigrants; (4) whether the traditional social order (the Gada) is able to cope with severe stress or not; and (5) implications of change for the most vulnerable groups in Borana society, namely women and children. It is also important to consider adaptive research regarding factors that could promote sustainable cereal cultivation. Needs for livestock research are relatively minor, except for a study dealing with alleged dilution of the Boran breed by highland stock.
It has been argued in Section 8.2.3.1: Overview of strategy that using banked livestock capital as a stimulus for local urban development is important for completing a development loop of mutual assistance between the Boran and town dwellers over the longer term.
Investigation is required as to whether feasible options exist for initiating small-scale industries in towns such as Yabelo, Negele, Mega and Moyale. Research is also needed to better quantify risks of banking versus holding wealth as livestock so that a portfolio management approach to maximise returns can be recommended.
A brief summary of the major implications drawn from research findings is presented here. There are implications here for basic and applied science. Perspectives are organised with respect to 28 important themes. For detailed discussion of research results, readers should consult sections of preceding chapters.
1) Equilibrial versus non-equilibrial systems: Currently there is a controversy concerning to what degree range-production systems exhibit equilibrial or non-equilibrial dynamics (Ellis and Swift, 1988; Westoby and Noy-Meir, 1989; Bartels et al, 1990; de Leeuw and Tothill, 1990; Behnke and Scoones, 1991). To recapitulate, equilibrial systems are characterised by more internal and negative feedback loops between vegetation and people. At certain stocking rates livestock productivity can be depressed due to stringent competition for resources; such competition may be exacerbated by consumer-induced declines in the productivity of the forage resources themselves.Livestock can affect major changes in plant species composition and vegetation structure.
And people can be more important than climate in affecting ecological trends; in this respect concepts such as carrying capacity are more relevant.
Dynamics of non-equilibrial systems, in contrast, are more strongly affected by climate. Internal interactions are less important in causing trends, and carrying capacity is thus a less relevant concept. One key implication of this distinction in system behaviour is that environmental degradation in non-equilibrial systems can be more attributable to variation in long-term patterns of climatic conditions such as rainfall. Degradation in equilibrial systems, in contrast, is more easily traceable to activities of humans and livestock. Both types of systems probably exist in East Africa (see Section 6.4.5: Equilibrial versus non-equilibrial population dynamics).
Indicators here suggest that the Borana system tends to exhibit equilibrial characteristics with stocking rate strongly mediating effects of annual rainfall on livestock and human populations. This agrees with more classical concepts of pastoral system dynamics (Pratt and Gwynne, 1977; Lamprey, 1983). Density-dependent effects on cattle productivity and mortality rates appear to occur. Cattle grazing probably facilitates bush encroachment, and thus system structure, by reducing risk of fire for woody seedlings. Grazing also encourages erosion on some landscapes. To infer that the system is equilibrial, however, should not imply that it is static. It is very dynamic within broad limits of stocking rates (i.e. within 10 to 30 head of cattle/km2). Annual stocking rates that vary so widely require different tactics to manage the system as a whole. Westoby and Noy-Meir (1989) contend that in contrast to equilibrial systems, non-equilibrial systems require opportunistic management. This perspective seems to equate an equilibrial with a static state. It is clear, however, that the equilibrial system of Borana also requires opportunistic management to deal with the long-term trend and phases in the interdrought cycle (see Section 7.4: Component interventions and system dynamics).
One problem seems to be how the term "equilibrium" is defined and interpreted. If an equilibrium implies that populations of plants and people vary little from year to year, it is difficult to conceive of any range environment that would conform to this. On a continuum of equilibrial to non-equilibrial system behaviour, the Borana system appears to exhibit a higher degree of equilibrial characteristics within broad limits of stocking rates. The cattle population is very dynamic, and the situation requires opportunistic management. Because the system tends to be equilibrial, more development options are possible than if it were non-equilibrial. Predictable pressures on the system can create opportunity for positive change.
The Borana system tends to be equilibrial due to: (1) the relatively high annual rainfall; (2) dominance of perennial herbaceous species; (3) soils and landscapes that have tendencies to erode under high levels of livestock use; and because (4) cattle and human populations are spatially confined (access to land by both is limited). Increased population growth, both within Borana society and among other neighbouring ethnic groups, appears to be the major contributor to increased spatial confinement. Development interventions which have served to improve access to permanent water source and reduce risk of disease to both livestock and people have probably helped to promote population growth. It is therefore likely that both population growth and development processes have contributed to encouraging a more pronounced, equilibrial behaviour of the Borana system in recent years But this assessment may also be a creation of the inability to better observe population dynamics over narrower spatial and temporal scales of resolution.
2) Effects of pastoralism on the environment: As alluded to above, the Boran have certainly had important effects on their environment. Roughly 40% of the study area is now occupied by dense woody vegetation while 19% has suffered soil eroision. It is very likely that both trends are attributable in large measure to cattle grazing. While some aspects of degradation may have happened recently, others have probably taken hundreds of years.
It was stated in Section 3.4.2: Environmental change, that high rates of population growth have recently compromised traditional patterns of resource use. The pattern is decribed as a "patch dynamic"; i.e. the Boran used a mixed-savannah region intensively for a number of years, bush encroachment occurred, the people moved elsewhere allowing the original vegetation to gradually recover as a result of rest from grazing, plus successional processes and bush fires. This cycle may have taken several generations to complete. The problem today, however, is that the people have been forced to become more sedentary because a high population density restricts options to move. The net result is a general increase in bush encroachment, with less opportunity for sites to rest and recover from grazing. Hence the only means to re-establish the traditional pattern is to dramatically reduce the human population.
While less than 5% of the study area was under cultivation through the mid-1980s, the prospect remains that cultivation could dramatically increase to offset risks of famine. If cultivation spreads unabated to shallow upland soils, it is foreseen that the danger for erosion over the short or medium term could greatly exceed that observed from decades of past use by cattle.
While it is more straightforward to assess grazing or cultivation-induced erosion as having negative impacts on system productivity, it is more difficult to generalise concerning effects of bush encroachment. Effects of woody species on the herbaceous community vary according to the type of woody species and site. Effects can be positive, neutral or negative. Overall, however, it is concluded that effects are eventually negative with respect to cattle production.
Given the relatively high rainfall, vegetation communities dominated by woody plants may represent the most stable condition for this system (Pratt, 1987a). If this is the case, the Borana Plateau may have been densely wooded prior to the arrival of pastoral peoples thousands of years ago (Lied and Morrison, 1974). Habitual burning by pastoralists may have established and maintained the mixed savannah which we see disappearing today. The ability to maintain this less stable ecological system may thus have been compromised by high population growth and government policy which prohibited range burning from 1974 to 1991.
While impacts from severe erosion could be taken as permanent, impact from bush encroachment could be reversible to a high degree. If population pressure can eventually be reduced by facilitating emigration of people out of the system, and if traditional management practices such as prescribed burning can be re-established, many bush-encroached areas could probably be recovered.
Effects of pastoralism on rangeland environments have recently received much attention (Winrock International, 1992). In a broader sense, however, it remains unclear which situation constitutes the greatest threat to Ethiopia, degradation of the highlands or the lowlands. It is thought that degradation in the highlands may be more critical. This is because the highlands have much higher densities of people and animals, higher rainfall, greater opportunities for expansion of cultivation and a larger prevalence of readily erodible landscapes (EMA, 1988).
Finally, it has been hypothesised that nutrient transfer by cattle from grazing areas to encampments via faeces is important in encouraging bush encroachment within the framework of the traditional patch dynamic. There are no data, however, to test this assertion and this could be a topic for future research.
3) Awareness of the Boran concerning environmental issues: Borana leaders are aware that their system is becoming overpopulated with people and that stocking rates of cattle are often high. They also equate heavy cattle grazing with bush encroachment and soil erosion. Proclamations made at the 1988 Gumi Gayu assembly were dominated by concern over resource use, and included measures to protect valuable trees, promote fodder banks, secure grazing for forra cattle herds and encourage better maintenance of water points. Many leaders appear to realise that their traditional way of life is coming to an end in several respects. They rarely offered, however, ideas as to how the traditional social order could better cope with stress and change. It has been mentioned that on some issues, the Boran need the help of government to regulate some aspects of resource use that the traditional system is unable to deal with.
4) Sustainability: This study was never intended to address concepts of system sustainability as they occur in the literature today (Vosti et al, 1991; Flora, 1992). Sustainability can be viewed from ecological, agricultural, social and/or economic perspectives. Precise definitions of sustainability remain elusive. It is reasonable, however, to offer some insights on sustainability of the Borana system based on our experiences here.
Concerning ecology, it is claimed that the cumulative effects of bush encroachment and erosion have indeed reduced land access, and thus carrying capacity, for cattle over the longer term. Informants have reported that the Borana system was dominated by more grasslands and open savannahs 30 years ago. We have no means, however, to determine what a reasonable baseline condition is and thus what the true extent of the impact has been. It is also debatable how important the loss of grasslands and savannahs is for system resilience compared to acute loss of grazing reserves due to human population growth in recent times.
For animal-based agriculture and economics, empirical models suggest that per capita food production in terms of milk and asset accumulation measured in herd size are both in a rapid downward trend. This is due to the increasing imbalance between the numbers of people and those of cattle. The acute nature of this trend dictates that this is the first "sustainability issue" to be dealt with in development strategies. Until people's needs can be better met over the short term, questions of long-term sustainability of the environment are less relevant.
It has also been forwarded that the Ethiopian Government has a stake in maintaining the social sustainability of the Boran system to promote the production of livestock for the rest of the nation. The immediate threats to the social order are famine and increasing poverty. Without a viable social order to coordinate labour to extract water from the deep wells in dry seasons, it is conceivable that the livestock system could suddenly become much less productive. A collapse in the social order could thus undermine system sustainability much faster than ecological degradation could.
It is important to note that the primary benefit of destocking, with compensation in the form of banked livestock capital, is in terms of system stabilisation and improvement of human welfare over the short to medium term. It is less justifiable at present to advocate destocking to preserve the environment.
5) Sources of system destabilisation: The crisis in Borana today, similar to that of other pastoral systems, is due to high rates of human population growth in an environment which is increasingly finite in terms of resources. The rate of population growth, however, is not excessive and is comparable to that for other semi-settled pastoral groups. The main problem appears to be lack of opportunities for people to emigrate outside of the pastoral sector. Lack of urban job opportunities is an important bottleneck, but so is the lack of education. It is unfortunate that efforts to develop the human resource in the southern rangelands have been far less than those employed to stimulate livestock marketing. The situation has probably been greatly exacerbated by the cultural and political isolation of the Boran from mainstream Ethiopian society. Farming peoples in the highlands are also experiencing population bottlenecks that continue to spur emigration in ever-growing numbers. It is supposed that farming people can emigrate much more easily to the urban sector than pastoral people like the Boran.
6) Change in human population growth: A review of secondary information suggests that the human population in Borana is growing at about 2.5% per year, with a doubling time of 28 years. This is in line with estimates of population growth for other semi-settled pastoralists. It has been speculated that the net rate of population growth is increasing, and sources of this increase could involve increased provision of food grain and health care, less adherence to traditional rules concerning behaviour and/or that the Gada cycle which imposes rules on reproduction is temporarily affecting only a smaller cohort of individuals in the population. The Boran believe, however, that any recent increase and there being no epidemic diseases is due to luck.
7) Biodiversity and the conservation of nature: As with sustainability, this study project was not originally intended to examine concepts of biodiversity as they occur in the contemporary literature. In perhaps the most comprehensive range-oriented review to date, West (1993) defines biodiversity as a multi-faceted phenomenon involving a variety of organisms, their genetic variability and the ecological units in which they occur. What are included as important components of biodiversity are largely shaped by the values of the observer; some argue, for example, that indigenous people be included as well. Despite problems with definitions and functional interpretation of the importance of biodiversity, West (1993) argues that concern for maintaining or enhancing biodiversity will increase among donors who fund projects in the developing world. In the developed world debate may increasingly centre on which species are functionally redundant, and whether or not species per se are the best currency to deal with. In the developing world, however, the problem is more basic and deals with a lack of information on types of organisms that occur and how they could be affected by land use.Given the acute nature of human crisis in the Borana system, it is offered that biodiversity is not a research priority at present. Our survey results provide only some preliminary information on the types of plants and animals present in the Borana system and one view of an appropriate schema of ecological site classification. Traditional uses of many plants by people and livestock have also been tabulated along with some possible trends regarding the abundance of woody species under heavy cattle grazing (see Section 2.4.1.5: Native vegetation, Section 2.4.1.6: Native fauna and Section 3.3.5.2: Household use of plants and pastoral perceptions of range trend). The region has a rich flora and fauna, but trends in abundance or diversity of these organisms as affected by human activity are not documented.
Future studies that try to assess biodiversity in the southern rangelands may first need to consider how woody encroachment affects biodiversity at the site and landscape levels of resolution. Attempts to enhance biodiversity could conflict with strategies to maintain the southern rangelands for cattle grazing. For example, although a high degree of woody encroachment may be unfavourable for cattle production, this may represent the best system state for biodiversity in terms of the number of plant species and a suitable habitat for wildlife.
Policy issues also bear on biodiversity, at least, with respect to highly visible species of wildlife. Informants report that there has been a decline in the abundance of large wild mammals in the past 30 years and this is thought to be due, in part, to the lingering effects of conflict between Ethiopian and Somali forces during the late 1970s. One impression is that the three government cattle ranches, which have existed since the late 1970s, appear to have served as refuge for wildlife because of low stocking rates and the exclusion of the Boran and their herds (see Section 2.4.1.6: Native fauna). It has been recently decided to return these areas to the Boran. This is an important gesture in light of severe resource constraints in the system, but it may bode unfavourably for wildlife. Similarly, pond development that increased access of the Boran to previously underutilised areas since the 1960s may have affected the local abundance and diversity of native organisms. Future strategies to maintain water development projects may consider effects on native organisms in addition to effects on the Boran production system as a whole.
8) Drought impact: Monitoring of drought impacts in 1983-84 suggested that: (1) people were only marginally affected in terms of famine-induced mortality; (2) many households in one region even reported births during the drought; and (3) the drought had differing effect on various sex and age classes of cattle. Still herds were decimated overall; the difference in the effect of drought on cattle versus people underscores the role of drought in exacerbating poverty. Roughly 45% of the milk cows, 90% of the calves, but only 22% of the mature males died (Donaldson, 1986). The higher survival rate of males is speculated to be due to their lower nutritional requirements compared to cows and their greater mobility. This result testifies as to the value of mature males in reducing the risk of cattle loss during drought. Patterns of cattle mortality suggested that deaths occurred in "waves"; it was hypothesised that the more productive animals died first, leaving a nucleus of less productive but hardier animals. This was also confirmed by cow-history analyses. This suggests that attempts to improve the local Boran cattle via breeding to enhance productivity would not be sustainable.
Finally, a recall survey in one region suggested that camels died to a similar extent as cattle during the 1983-84 drought. The advantage of camels, however, appears to be their continued milk production, which allows camel owners to sell milk during drought when prices are high. This is in stark contrast with the terms of trade of other livestock products for grain that fell dramatically.
9) Efficiency of pastoral production systems: As in other African pastoral systems, once the high stocking rate is factored in and milk is included as a major product for people, the efficiency of the traditional pastoral production of the Borana system exceeds that for commercial ranching in a comparable setting in terms of energy yield per person and per unit area. Commercial ranching produces more meat per head, however. This shows, in general, that commercial ranching is a poor alternative for a subsistence-oriented pastoral society. This anomally, however, has been amply demonstrated here and elsewhere in terms of the incongruity of production objectives and human demographics between the two modes of operation. The one problem with the pastoral system compared to modem ranching, however, is that the higher stocking rates of pastoral systems increase the danger of risks for environmental degradation and system instability in some situations.
10) Productivity of range livestock: It is commonly assumed that compared to unimproved animals in higher potential areas, range livestock have a lower level of productivity, whether caused by breed characteristics, climate or both. There is no strong evidence to support this view for the southern rangelands. Limited data on small ruminants suggest that their productivity is similar to that in other systems. Performance of unimproved Boran cattle under pastoral conditions appears similar to the lower end of the spectrum for unimproved animals reared on research stations elsewhere in Africa. Under low stocking rates and favourable rainfall, it is hypothesised that the productivity of the Borana cattle system can be extremely high. Compared to higher rainfall areas, range systems also have the advantage of a lower incidence of many important diseases (Sileshi Zewdie, SORDU veterinarian, personal communication). Provided adequate rainfall, the highest rates of production per head would have occurred in 1985-86 and again in 1992-93. Production is thus cyclic in nature as mediated by stocking rate.
A related issue is the precision of cattle production studies in different environments. It has been hypothesised that cattle productivity in the southern rangelands is markedly affected by rainfall and stocking rate. Research results thus need to be qualified with respect to these variables in any given year. One result of the failure to do this are studies that do include sufficient animals but that yield widely differing statistics for the same system in different years. Ideally, productivity should be evaluated under experimental conditions at an assortment of stocking rates and rainfall years. Mean figures could then be calculated as weighted averages that consider the relative frequency of different background conditions.
The one major production feature that requires intervention in the Borana system is the problem of calf mortality. This is an appropriate strategy because it builds on traditional values and requires a small amount of strategic resources that are locally available. Tactics would vary, however, according to the type of rainfall year and wealth of the producer household. Wealthy households appear to have more problems with calf diseases while poorer households appear to have more problems with calf nutrition. It is also proposed that calf mortality can be density dependent under high stocking rates. The most sustainable improvement in reducing of calf mortality would probably be achieved along with a concomitant increase in animal offtake.
The typical nutritional research concept in seasonal African environments is that ruminants are primarily limited by crude protein in dry seasons. This implies that other nutrients are less limiting. From Section 7.2: A theory of local system dynamics, it is offered that sequential changes in stocking rate during the interdrought cycle affect nutritional constraints in a stepwise manner. This is an ecological, rather than mainstream agricultural, view of nutritional dynamics.
Using the period 1985 to 1990 as an example, it is hypothesised that minor nutrients like minerals could have been a more pervasive constraint for cattle nutrition when the stocking rate was less than 10 head/km2 in 1985. When herds began to recover during 1986 to 1988, protein might have emerged as the principal problem in dry seasons. By 1989, however, herd owners reported that cattle were unable to regain condition even during the long rains even though precipitation was above average. From 1989 onwards, when the stocking rate was greater than 20 head/km2, energy was likely the problem because of forage competition. When energy is the problem in a risky rangeland environment, the best measure is some creative destocking to avoid system collapse. The point is that the model of protein limitation has only limited utility, even in situations where protein supplementation of mature cattle is feasible. This illustrates the value of an interdisciplinary view that involves animal production and ecology.
As with other pastoral systems, the central question in Borana seems to be how to stimulate more offtake of animals in such a way as to not compromise pastoral security based on animal assets. Improvement of cattle productivity overall should benefit from increased offtake. And to bank livestock capital (described above) are forwarded as the best-bet means to achieve greater cattle offtake
11) Compensatory growth of cattle: Claims that adult range cattle are unable to compensate for periodic deprivation of water or for nutritional deprivation as calves, are results falsified by controlled trials. In fact, the animals have exhibited considerable ability for compensatory growth under the experimental conditions employed. One implication of the calf growth trial is that as long as the calf lives, regardless of the level of early nutritional deprivation, the odds are that it will attain a level of lifetime productivity similar to animals that have not been deprived. Under conditions imposed in our trials, calves were eventually abe to compensate for milk "lost" to people as offtake. long-term growth patterns are thus controlled to a large extent by the environment. Risks of extensive weight loss by cows at some point during their productive lifetime are very high. This can effectively cancel out expensive inputs administered to the animal as calves. Calf feeding for sustainable improvements in growth is thus very risky and inappropriate. As long as cattle are marketed as matures, calf feeding for improved selling weights is irrelevant.
Calf growth rates were improved to the greatest degree by provision of improved forage and water in contolled trials. This suggests that in drier environments, water availability may be a significant constraint in the ability of animals to make use of improved forage supplies.
12) Conservation of indigenous livestock breeds: As reviewed in Section 1.2: The lowlands and pastoralism in a national perspective, the Boran cattle breed is important in terms of its productivity, durability and marketability for domestic use and export. There has been concern as to whether this breed is in danger of genetic dilution from inferior breeds from the southern highlands. This concern may have already spurred action, as one of the objectives of the Ministry of Agriculture in establishing a ranch at Did Tayura, north of Yabelo, was to promote the genetic integrity of the Boran through on-station breeding programmes.
The Boran trade their male cattle for highland cows during drought-recovery periods to speed up recovery of milk production and herd growth potential (Tafesse Mesfin, TLDP General Manager, personal communication). This underscores the value of male cattle for post-drought trade in the Borana system, and it also shows that cattle in the southern highlands are serving as a stable reservoir to add resilience to the performance of the range sector. The extent of the trade remains unquantified. It also remains for research to document whether the alleged genetic dilution is truly a long-term threat. If highland cows perform poorly under stress in the rangelands environment, it is possible that the long-term effect on the Boran gene pool is negligible (see Section 5.4.5: Cattle growth and implications for breed persistence). It is also unclear whether this trade is a recent strategy to help cope with increasing equilibrial dynamics of the production system or a practice that has gone on for many generations. If the former situation is true, preservation of the Boran breed may be yet another benefit of attempts to better manage and stabilise the pastoral system in response to drought.
13) Pastoral social organisation and resistance to change: It has almost become folklore in African development circles that pastoralists are very independent, conservative and fiercely resist outside efforts to improve their lot in life. There is no evidence, however, to support this position from the southern rangelands as observed during the 1980s. The Boran are open-minded to appropriate ideas and have in fact pioneered some of their own concepts in resource management. In this they could be an appropriate model for the introduction of new development concepts in semi-arid Africa. Part of this may be due to the persistence of their traditional social structure which can greatly facilitate communication with development agents, and it has aided project implementation in the past. Research that can enable outsiders to better understand the internal workings of the society and show how the social order can or cannot cope with new pressures would be very valuable.
Even though it has been relatively easy to work among the Boran in the past, they have become rightly suspicious of government from their experiences with Peasant Associations, national recruitment for the military and fulfilling mandatory quaotas for livestock sales in the 1970s and 1980s (see Section 1.4.3: The SERP and the Pilot Project). To what extent their full trust can be won back by the new government remains unclear.
14) Development constraints: One of the only intervention concepts that is entirely independent of external resources is hay making using local grasses and native legumes to provide improved feed for calves. Most of the others, either directly or indirectly, rely on some degree of external support. Granted the sustainability and appropriateness of such interventions are open to debate, the problem then becomes whether any substantial improvement in human welfare is possible without stronger linkages to the rest of Ethiopia. The Boran are willing to pay for innovations they consider valuable; so realistic demand for development for which they will contribute at least in kind is presumed to exist. Most of the production and development constraints enumerated in Chapter 7: Development-intervention concepts, are considered to lie outside of the pastoral sector. This discounts the notion that the main constraints to change are caused by inappropriate pastoral attitudes or behaviour. The inability of the nation to meet adequately many routine needs of food or other commodities is paramount. To the extent that national development initiatives improve the functioning of commerce and the creation of urban job opportunities, they should also have positive effects on Borana society.
15) Top-down versus bottom-up innovation concepts: Experience here suggests that top-down innovation concepts such as the pond scoop, the improved butter churn, pasture improvements using exotic herbaceous forages, drought fodder banks using Atriplex and Opuntia forages and stimulating cattle growth through early nutritional supplementation of nursing calves will not succeed. Because the Boran appear unwilling to risk valuable cattle to pull the pond scoop which seems an extended project to them; the butter churn may have been more relevant 30 years ago when per capita milk surpluses were more common; exotic herbaceous forages appear unproductive due to constraints of rainfall and/or air temperature; drought fodder banks are impractical given the scope of the problem of sustaining thousands of cattle during drought; and attempting to stimulate cattle growth over a three or four-year time horizon is simply too risky according to the Boran. In contrast, innovations devised on the basis of grass-roots knowledge of the community such as hay making, use of native legumes, cement cisterns, banking livestock capital and strategies to mitigate calf mortality appear far more appropriate given the cultural attitudes and their greater reliance on local resources. These innovations are thus much more likely to have their impact.
There has been a debate recently regarding an appropriate mix of "upstream" (i.e. high-tech) versus "downstream" research activities in sub-Saharan Africa (Winrock International, 1992). Whether it involves developing appropriate technology or gaining key insights for the interdisciplinary systems approach, this study project of the Boran system speaks for the value of downstream research in which the beneficiaries of development play a significant role. The value of upstream research in this project has been relatively played down.
16) Linkages among development entities and research institutions: In our experience, the collaboration between grass-roots development agencies and researchers has proven to be fruitful (see Section 1.4.7: Interaction between research and development and project impact). Creative development agents helped complete the loop between the Boran and researchers to create a more participatory farming systems research approach. One paradox is that while researchers are pre-occupied with creating technological interventions and generalisable system perspectives, they can be blind as to the complexity of the day-to-day lives of rural people and how this complexity constrains adoption of innovations. Development agents can help provide the eyes and ears for research. Today, the research establishment is under increasing pressure to demonstrate the impact of their efforts on people. Both in terms of helping understand real problems, extending appropriate technology more rapidly and the widespread dispersal of project results in the scientific literature, collaboration between research and development workers can be mutually beneficial. One drawback in this respect is the fact that researchers and development agents have different perspectives, education, values and goals. This makes truly mutualistic collaboration difficult, the focus is clearly agreed to be solving acute problems of people in a given setting. That was the source of our success at integration here.
17) Interventions that increase production versus those that mitigate risks: Whether the intervention is reclaiming drought grazing reserves, banking livestock capital, facilitating sustainable agropastoralism, educating pastoralists to leave the pastoral sector, forging durable links between pastoralists and farmers or assisting the smooth functioning of markets and viable terms of trade, the most important initiatives for the Boran today involve dealing with a crisis situation. This means expanding the social and economic options for the people to enable them to be more opportunistic in coping with a rapidly changing world. Such action gets its mandate, in part, from populations being in a very precarious position with respect to their baseline resources compared to the past, and embodies the philosophy of Sandford (1983a) who considered "efficient opportunism" to be the fundamental element of sustainable pastoralism. In this view technological improvements, are considered very minor in their potential impact and are said to serve more of an ancillary role. Today the questions should revolve more around how livestock assets can be used to improve the human condition rather than a singular focus on livestock productivity per set.
The irony remains, however, that most research and extension personnel are inculcated with the ideology that improved technology is the only solution. Risk mitigation appears more complex and beyond the domain of any one development agency. It is proposed that while individual agencies can each make a positive contribution to pastoral risk mitigation, the most effective approaches would involve interagency efforts that coordinate policies and procedures that deal major blows to the problem. The issue then becomes one of effective dialogue and policy formulation among institutions which increasingly affect the destinies of people like the Boran.
18) Forage improvements: Forage trials indicated that dual-purpose annuals such as cowpea (Vigna unguiculata) and lablab (Lablab purpureus) were the highest producers of seed and forage. It is contended that forage extension should focus on plants that help the people feed themselves first. Cowpea appears to offer the best possibilities as an intercrop with maize. For range improvements, the best bet is to stick with the most promising native species of grasses and browse. These have the advantage of proven persistence in the environment. Herbaceous exotics, in particular, appear to have their growth constrained by low rainfall and/or cool ambient temperatures. For calf forages, the best intervention by far is hay making using indigenous grasses.
19) Gender issues: From two independent surveys, it was somewhat surprising to find that 20 to 25% of household heads were women. This tendency may be more pronounced in pert-urban areas occupied by a higher proportion of poor and middle-class households (Holder and Coppock, 1992). Many of the technical intervention concepts forwarded in this volume should be targeted towards women in pert-urban locations. The social and economic status of women, and whether their roles are changing in Boran society, should be more a focus of research. If more young men leave the system, the spectre is that women and children will assume more labour responsibilities. Whether new duties for women will include more strategic and managerial tasks is unclear.
20) Impact of cement cisterns on labour and water use: Provision of local water tanks did not appear to reduce the time women spent collecting water in dry seasons nor did it seem to alter markedly the pattern of water usage by people in the household. The additional water provided by water tanks appeared to be directed towards calves. Uncertainty in the length of any given dry season means that water will still be used very conservatively. The fact that water tanks did not alter women's work schedules should not imply that they don't have other benefits in terms of improving the quality of life for women.
Despite results of oral surveys in which women always indicated that they worked long hours every day, direct observation of married Borana women on a 24-hour basis suggested that they spend around 30% of their waking hours in activities associated with leisure in the dry season (Coppock, 1992a). Other studies of time allocation in wet periods indicated that they have ample time to incorporate activities such as hay making into their schedules (Coppock, 1991). Ultimately, whether women would use more of their time to implement innovations will depend, to a large extent, on how they value leisure time in comparison to benefits perceived to result from innovations.
21) Livestock marketing: Analyses of market records from the early 1980s confirmed that cattle were the dominant species sold and mature male cattle were the most common age and sex group marketed. This is common elsewhere in pastoral Africa. Studies of livestock marketing behaviour as reported by the Boran suggested that many behave as "optimistic gamblers" (Coppock, 1992b). Far from being predisposed to sell cattle readily to buy food when a drought occurs, they struggle to endure considerable misery before they capitulate to a sale. Cattle thus appear to be held more as a lasting asset rather than an expendable resource. They hold out as long as they can and hope that the next rainy season will be adequate and thus allow them to avoid sales.
This explains drought-marketing dynamics in the southern rangelands. The attitude has negative implications for the stability of the system under drought perturbation because the collective effect of such behaviour can translate into disastrous herd losses for the community when more animals have to compete for less forage in drought grazing reserves. This behaviour of waiting to sell "until you have no choice" has negative implications also for seasonal strategies intended to improve terms of trade between livestock and grain. Although the post-harvest season is the best time to maximise favourable terms of trade, the Boran are reluctant to sell then because "they don't yet have a problem." Innovations such as grain storage and banking livestock capital thus face considerable cultural obstacles. Confronting these obstacles is the key to system transformation and development.
Adult male animals, followed distantly by cull cows, are preferably sold because a herd owner receives the greatest net return from their sale, given that production costs are minimal. Cash received is used to buy commodities plus replacement calves so that selling mature cattle can meet the two objectives of generating income and herd building. The Boran do not want to sell immatures except the poor because they have no choice.
Borana marketing behaviour and cultural values strongly suggest that the perverse supply rule (Sandford, 1983a) operates here, although this will vary with livestock species and household wealth. This is not to say that animals will not be drawn to markets that have higher prices, they will. It is to say, however, that households seek higher prices so they have to sell fewer cattle over the long term, and thus avoid herd depletion. Except for households actively trying to build their herd through trade by buying immatures, the majority of households have relatively fixed cash needs and that this is a disincentive for livestock commercialisation. Stimulating cash income among the Boran may be far easier using sheep rather than cattle markets; this is because the two species serve different economic functions in different households. Sheep are more of a cash crop while cattle are for milk production and a means to store wealth.
22) Livestock commercialisation: While it is expected that most households will engage in more livestock sales in the future, this will be essentially a coping reaction to pressure to secure more food grain. Unless markets are stable and offer grain at favourable terms of trade, such increased market dependence would be dangerous. Informants have, for example, expresed the fear that producers who are exclusively commercially oriented may gradually emerge from the ranks of the educated wealthy and/or from those with strong ties to the urban sector; some of these people may have been government export agents in the early 1980s and that is how they reamed the business. If commercial herds ever have to compete with subsistence herds for water and forage, this could be a source of social conflict in the future. Overall, it is hypothesised that traditional producers are unlikely to be transformed into purely commercial operators.
23) Evolution of agropastoralism: Demographic preconditions could now force a widespread shift to agropastoralism in the Borana system. The extent of this shift will, however, be limitted by a landscape that offers relatively few opportunities for sustainable cultivation. Rainfall may be less of a drawback while the patches of valley bottoms, swales and vertisols, where cultivation is viable, may comprise less than 12% of the study area. It remains unclear the extent to which competition for these valuable sites is a problem for the production system (Scoones, 1991). In a few documented cases, the Boran appear to be attempting to accomodate both cultivation and forage production on these sites even though they seem to be aware of the negative implications that extensive cultivation could have for the grazing system overall in terms of competition for key resources and environmental degradation.
McIntire and Gryseels (1987) noted the uncertain nature of forging a working linkage between crops and livestock in the semi-arid zone. Our observations support this contention because, at least in respect to the recent past, cultivation seems to be pursued when the livestock sector fails, and vice versa. Integration of cultivation with livestock production would probably happen only when the ratio of cattle to people becomes low enough for cultivation to be attempted on yearly basis. This suggests that agropastoralism would emerge as a result of chronic negative pressures on the system. Agropastoralism thus does not necessarily equate with "progress" and improvement of human welfare here. Because of constraints of arable land it is expected crop-livestock mixed production possibilities will remain relatively tenuous for most households in the forseeable future. This is particularly true if development strategies advocated earlier in this chapter succeed in managing the system in such a way to release pressure. If this occurs, the most desirable situation, especially in terms of promotion of environmental sustainability, would be for cultivation to remain as an opportunistic activity only. The future strategies for the system should thus keep a primary focus on sustainable extensive livestock production.
24) Evolution of dairy marketing: Observations here suggest that pert-urban dairy marketing evolves in response to increasing poverty and food deficits. Dairy products are renewable sale items that provide a regular income and allow producers to avoid selling from a small supply of animals. In the past producers used to sell dairy products occasionally to buy some discretionary items; today more of them sell to buy grain for survival. Wealthy producers sell more products in absolute terms if they reside closer to a market; in relative terms, however, dairy marketing is more important for the poor because they have little else to sell. Hence dairy markets provide ample reason for the poor to migrate to pert-urban locations where the favourable terms of trade allow them to sell an otherwise inadequate food-energy source (milk) to be sold to purchase a survival ration of grain. Dairy marketing is expected to be strongly influenced by the long-term system trend and by the interdrought cycle. When the poor sell more milk, it is likely to be taken from the calf. Calves of poor households near towns are therefore expected to be at increased risk of morbidity and death due to malnutrition.
25) Evolution of herd diversification: Trends suggest that herd owners seek to diversify into small ruminants to take pressure off the need to sell more cattle to buy grain. Cattle represent a collective family asset for which the producer needs to obtain a concensus that a sale is in the best interest of the household. Small ruminants, by contrast, are truly "small change" and appear to be sold whenever a need arises. It is unclear whether men or women control small ruminant production and sales.
26) Pastoral household diversity: The Boran, like other African pastoralists, are a diverse society so that the idea of the "average household" has little use in understanding the dynamics of the system or in prescribing blanket intervention approaches today. Similarly, households exhibit great regional variability in their access to natural resources and local variability in their access to urban markets. The system should be appropriately segregated into pert-urban and rural sectors, with the former consisting of areas within a radius of 30 km of major towns. Compared to people far from towns, those in pert-urban areas are expected, on average, to be poorer in livestock holdings, more market-oriented, more aware of a changing world in general and more likely dominated by female heads of households. One interesting point is that the "average household" may have been a more relevant concept 30 years ago when the society was not as diverse as it is today. Trends towards greater wealth stratification and peri-urban economic dependence may be exacerbated by increased competition for resources and equilibrial system dyanmics. Greater wealth stratification is evident from statistics that suggest that the wealthy, which may comprise around 18% of the population, control 65% of the cattle while the poor with 51% of the population may control only around 10%.
27) Urban-pastoral linkages: Over the long term, it is proposed that promoting the development of towns such as Yabelo, Mega, Negele and Moyale holds potentially great benefits for the Boran in terms of providing local markets and job opportunities. It appears, by contrast, that only a very few Boran have ever emigrated to distant urban centres such as Addis Ababa. This difference means that urban development initiatives which focus on the smaller towns and cities in the countryside, rather than on the major metropolis, would offer many more associated benefits to livestock producers like the Boran.
It is also proposed that banking livestock capital could add considerable impetus for funding rural development initiatives in small towns by providing funding in addition to helping the Boran better manage their personal assets. Interest in stimulating urban development among international donors has been spurred, in part, by the current crisis of widespread migration of rural people to the cities (Winrock International, 1992). One important question involves where the capital will come from to facilitate the growth of small businesses and urban enterprises. It is argued here that some of this capital could come from livestock otherwise used as traditional stores of wealth. Policies and incentives which promote banking livestock capital could be useful in a pan-African context and could have the ancillary benefits of relieving pressure on grazing lands, stimulating animal production per head and stabilising system dynamics. Research is needed, however, to study the economic risks involved for livestock producers in the context of portfolio management of assets.
28) Systems science, interdisciplinarity and education: It may be argued that systems research is inherently site specific (Winrock International, 1992). It is contended, however, that the systems perspectives described in this volume are founded on basic principles that can be applied to any livestock production system in Africa. One implication of the extended chapter discussions and literature reviews is that, while pastoral systems appear diverse superficially, they are remarkably similar in many fundamental respects. The diversity, in part, results from any one system being at a different point along a more or less similar continuum of change. For example, the Maasai of Kenya are probably a couple of generations ahead of the Boran in terms of coping with system pressure. The effect of their proximity to Nairobi has led to system dynamics proceeding in some directions that will not hold for the Boran. The Boran, in turn, may be "ahead" of other pastoral groups in Ethiopia that remain more isolated from the outside world.
The key for getting at appropriate possibilities for development impact is knowing the status of any given system at any point in time. While commodity work can point to what to do in a given situation, systems work conducted in parallel can show us when to intervene and why. Systems analysis is thus appropriate to facilitate impact of technology and management innovations. It may also be argued that the long time frame and the expenses involved make the future of systems work uncertain in Africa. While in part this is true, it is offered that, rather than merely funding more long-term work, the challenge is to extract principles of system function from existing studies, compare these and boil them down to a series of key interactions. These hypotheses can then be tested in an economical fashion in a variety of situations so that systems science can make progress. Verified principles can then be used more widely in the diagnosis of local problems and for their solutions within the framework of rapid appraisal of rural communities (Chambers, 1992).
The intellectual source for the systems approach we have employed is diverse; it has called upon material in population ecology, theories of agrarian change (Boserup, 1965) and classical farming systems research. It has not been founded on concepts dealing with ecological community structure, energy flow or nutrient cycling as have been invoked in other pastoral research projects (Coughenour et al, 1985).
The most useful computer models in our approach were of the simple empirical variety that illustrated some consequences of population dynamics on a finite resource base. Cheap and easy to construct, these models were used sparingly but provided good general insights. In contrast, models involving projected herd performance or economic outcomes of various production strategies (Upton, 1986b; Cossins and Upton, 1988b; von Kaufmann et al, 1990) helped formulate ideas and organise thinking, but these were ultimately assessed to have limited utility in predicting real-world outcomes or prescribing best-bet innovations. It is speculated that these problems arise from several factors that include: (1) limited depiction of the true complexity of ecological dynamics and human behaviour; and more specifically (2) the difficulty in portraying risk as viewed by rural livestock producers. More data are needed on biological, ecological and sociological interactions for inclusion into computer models.
The framework outlined in Section 7.2: A theory of local system dynamics may be unusual compared to previous systems theories because: (1) it presents a very dynamic view of system interactions; and (2) the most important interactions can only be understood if one sees that system forces act across disciplines. For one simple example, stocking rate influences milk production per head and per hectare and milk production influences the social and economic behaviour of households. The suite of these interactions can shift from year to year, and also depend on rainfall and other external events.
All this implies that to understand system dynamics, analyses of cause and effect must be interdisciplinary. It is proposed that this represents a wide open field of scientific investigation. In contrast to viewing cause and effect as interdisciplinary, independent scientists who conduct multi-disciplinary work in parallel probably tend to examine issues of cause and effect as isolated and by confining them within their narrow world view. These situations offer little opportunity to advance systems science.
Another case in point is the intervention involving banking livestock capital. This is an interdisciplinary intervention in terms of what to do, when to do it and how to do it. Again, banking livestock capital is as much a nutritional intervention for cattle as it is an economic intervention for people. Knowledge of livestock production and ecological concepts of carrying capacity and competition allows one to understand variability in risk over time and see when the intervention should be implemented; knowledge of traditional, social and economic values allows one to see why the intervention could be resisted; and knowledge of economic forces operating in the commercial banking sector allows one to see other economic tradeoffs involved, all pointing to the vital need for interdisciplinary and systems-oriented education for people involved in researching or developing systems and requiring capabilities to deal with problems concerning agriculture and natural resources.
