Researching the Future of Pastoralism in Central Asia's Mountains: Examining Development Orthodoxies

Abstract This paper synthesizes research findings on contemporary mountain pastoralism in Kyrgyzstan and Tajikistan, based on a longer review characterizing mountain agropastoralism in Central Asia. We focus here on the principal issues that have been emphasized over the past two decades in policy, programs, and projects regarding pastoralism in Central Asia's mountains. We conclude that this emphasis has largely been driven by two unproven orthodoxies about The extent and causes of pasture degradation; and The need for decentralization and pasture land privatization. The paper proposes that new research should critically assess these orthodoxies through more empirical and long-term field research. This will yield practical applications to improve conditions for Central Asian mountain pastoralists and their environment. Pursuing measures for addressing pasture degradation will require determinations of whether, where, how, and why degradation and desertification are occurring. Detailed field research is also called for on the processes and effects of decentralizing the power to allocate and manage pasture resources from national and regional state authorities to local communities, as well as on the long-term effects of privatizing pasture land.


Introduction
Societies living in mountainous areas of Central Asia face particular adversities as a result of the geopolitical and economic conditions of post-Soviet states. Current adaptations by mountain societies were reviewed in an extensive report on pastoralism and farming in Central Asian mountains (Kerven et al 2011). The present paper is a shortened and partial synthesis of that review, aimed at two notable themes that emerged from our overview of research on Central Asian mountain agropastoralism over the past 20 years. These themes are (1) pasture degradation and (2) promoting participatory approaches to communal pasture management.
This paper considers only mountain pastoralism in Kyrgyzstan and Tajikistan, where we found that the bulk of relevant studies have been conducted. We have not sought to cover all variables that affect the lives of mountain pastoralists in Central Asia (eg the role of religion and ancient cultural beliefs). We propose an agenda for biophysical and socioeconomic research on mountain pastoralism in Central Asia.

Old and new research directions
The emphasis of research has radically changed since the end of the Soviet period in 1991. Contemporary studies of pastoralists and the context of pastoralism in Central Asia have moved far from the practical concerns of Soviet social and biological scientists. Scientists in the USSR were looking long and hard at the ground, the plants, and the animals. Directed by state planning committees, their obligation was to increase and stabilize production output within a strict ideological parameter of development.
In contrast, much post-Soviet research on agricultural production systems in Central Asia is rooted in Western development models, activated through international (mostly Western) funding channels that support short-term research and development programs, and variously aimed at bolstering civil society, biodiversity conservation, sustainable land management, and market value chains. The post-Soviet swing in emphasis has meant less basic biological research and few rigorous investigations regarding the present-day status of Central Asian mountain pastoralism. At the same time, the quality and funding of national research institutions has eroded throughout the transition period since 1991 (Kerven et al 1996).
There is, however, a considerable body of documentation-in development project reports and in Russian-language scholarly workson ethnohistory, livestock breeding, pasture vegetation and soils, and other specialized fields relevant to understanding Central Asian mountain pastoralism. We do not summarize Soviet-era research on this topic, though this is recommended for a more complete understanding of the basic issues (Kerven et al 1996).

Characterizing mountain pastoralism and transhumance in Central Asia
Most of Central Asia has a semiarid or arid climate. Thus, the mountains are attractive to pastoralists because they usually receive more precipitation than the plains and valleys (Mamytov 1987;Russian Nature 2011). The natural mountain vegetation offers alternative nutritional qualities for livestock (Kerven 2003). Slopes can be used to build gravity-fed irrigation channels to water food and fodder crops. Springs and streams provide water to people and livestock. The cooler mountain climate in summer means a more pleasant environment for people and their livestock.
But there are also severe drawbacks to making a living on mountain land in Central Asia. Higher levels of precipitation result in deep snowfalls in winter, which can cut off villages for long periods. Some livestock breeds cannot forage under deep snow and are not physiologically equipped to cope with intense cold periods. At higher altitudes, the short frost-free period results in a limited growing season for food and fodder crops, as well as natural pasture (Khukmatullo et al 2005). Transport is impeded by steep and dangerous terrain, and routes may be blocked by avalanches and rock falls. Remoteness and inaccessibility can lead to social isolation, as well as political and economic marginalization, as discussed later in this paper.
Given the attractions and disadvantages of mountains for making a living, from prehistoric times in Central Asia, people adopted the practice of transhumance: They spend part of each year with livestock in the mountains, when environmental conditions are optimal, and the rest of each year somewhere else-in adjacent lower valleys, distant plains, or even cities (van Leeuwen et al 1994).
Natural pastures-that is, unimproved by planted species or techniques-constitute the principal land area in the two countries, where mountain pastures are by far the greatest source of livestock forage (Table 1). Mountain pastoralism is a significant contributor to the gross domestic product in Kyrgyzstan and Tajikistan (Peyrouse 2009: 5).
Most pastures are located at altitudes between 1000 and 3500 masl, in intermontane valleys and mountain slopes, while one quarter are found at elevations greater than 3500 masl. Climatic conditions therefore differ according to slopes with different aspects, enclosed basins, and exposed plateaus (Russian Nature 2011; Sedik 2009). Annual precipitation varies from more than 1500 mm in the Gissar Range of Tajikistan to less than 100 mm in the Eastern Pamir (Khukmatullo et al 2005;Conservation International 2012). Much precipitation falls as snow in autumn, winter, and spring. Snowmelt in summer provides drinking water for humans and livestock, and for crop irrigation, in environments that are otherwise often very dry.

Pasture degradation
Degradation of land has been the predominant issue for many researchers and development agencies concerned with mountain regions of these two countries. This is evidenced by the proportion of reports and projects on the subject, noted in our review (Kerven et al 2011). Pasture degradation-its definition, causes, effects, extent, and amelioration-was also the most debated topic at the Bishkek Symposium titled ''Pastoralism in Central Asia: Status, Challenges, and Opportunities in Mountain Areas'' in June 2011.
The concept of degradation and its particular applicability to Central Asian mountain pastures is not simple. There are multiple and nonstandard criteria of degradation regarding pastures (Briske et al 2005). The initial problem is to decide what is to be measured (Behnke and Scoones 1993). A few examples are loss of plant and wild animal diversity from some previous measured state, retreat from botanical climax, rise in toxic and unpalatable plants to livestock, loss of topsoil and humus, increasing bare ground with no vegetation cover, and permanent loss of an economic good, in this case an irreversible decline in livestock production. Botanists, range ecologists, socioeconomists, pasture agronomists, livestock production specialists, and agricultural policymakers are unlikely to all agree on the critical indices of degradation.
In the last decades of Soviet planning for agriculture, Central Asian pasture scientists steadily and insistently challenged the orthodoxy that humans could always conquer nature by intensifying production and relentlessly increasing livestock output on the pastures (Kerven et al 1996;Alimaev and Behnke 2008). They pointed out the ecological ceilings that, if surpassed, resulted in critical environmental damage and loss of economic productivity and warned against greater development of irrigated fodder and food crops on steep mountain slopes (Mamytov 1987;Zotov and Adenov 1992). Since the 1990s, this early warning has expanded into a crescendo of concern by international donors and nongovernmental organizations (NGOs).
But in less than 1 decade after 1991, the entire Soviet mode of production in Central Asian pastoral regions was destroyed, and livestock numbers plummeted (Kerven 2003;Pomfret 2006), leaving the locus, causes, and degree of pasture degradation all radically altered (Coughenour et al 2008). Meanwhile, the ability of most private pastoralists to invest in alternative pasture management methods has been severely limited.
To date, studies in the region have indicated determinants of pasture degradation that are fundamentally different from the damage to land caused by the high input and N Changes in pasture tenure regulations and de facto use, resulting in poorer people's livestock having limited access to better-quality pastures surrounding key resources, as well as allowing richer people to claim better pastures; and N Global climate changes, leading to warming tendencies and changes in precipitation amount and timing in the region's mountains.
The following sections summarize the literature on these drivers.

Reduction of livestock seasonal mobility
Seasonal transhumance among plains, mountain valleys, and distant upper meadows has been greatly reduced in the post-Soviet period (Rahim and Maselli 2008;Robinson and Whitton 2010). Following dissolution of the state livestock farms ending state-controlled pasture use, pastures farther from mountain settlements have become underutilized, while the more accessible pasture areas are overutilized. This is reported for Kyrgyzstan by Schillhorn van Veen (1995), Ludi (2004), Farrington (2005), and Undeland (2005). In Tajikistan, studies show uneven seasonal grazing utilization over space (Domeisen 2002;Hangartner 2002;Haslinger et al 2007;Sedik 2009;Wirz 2009;Vanselow 2011). Grazing pressure is particularly severe during winter, and poorer households are obliged to graze their livestock continuously around villages and to destock. Efficient seasonal utilization of remote pastures requires expensive transport by vehicle and additional labor (either family or hired; Figure 1); moving livestock to remote pastures has therefore become an option mainly for richer households (Hangartner 2002;Farrington 2005).

Livestock feed shortages
By the later Soviet period of planned livestock production, state investments in irrigation, mechanized transport, and other infrastructure permitted heavier livestock pressure on pastures, making livestock reliant on plentiful and highly nutritious winter feed (Schillhorn van Veen 1995;Fitzherbert 2000;Ludi 2004).
In the early 1990s, fodder yields collapsed in both countries due to the privatization of arable land and the lack of cash investment, fertilizers, and working machinery. In the mountain areas, arable land is usually scarce and nowadays not irrigated; in the Soviet era, it was used for fodder cultivation or as hay land. Most rural households now, however, use irrigated mountain land to cultivate potatoes and vegetables for their subsistence needs (Eriksson 2006;Ronsijn 2006;Akramov and Omuraliev 2009). Furthermore, labor migration has led to abandonment of some arable land (Wolfgramm et al 2010), with consequent decline in availability of fodder crops, residues, and hay stored for winter. Cessation of concentrate imports from other former Soviet republics further MountainNotes worsened this winter feed scarcity (Fitzherbert 2000). The inefficient processing and storage of hay also aggravates the winter feed scarcity (Figure 2), with estimates that this leads to a loss of energy and nutrients of up to 40% (World Bank 2007).

Poverty of privatized livestock owners
Grazing and feeding livestock requires inputs of cash and labor, in addition to suitable land. When these are in short supply due to poverty, livestock owners must still try to keep their few livestock alive and reproducing. They do this by feeding them whatever is available, wherever they find it, and at the lowest cost. This has been one of the principal reasons more accessible pastures were overgrazed in the post-Soviet period (Kerven et al 2006(Kerven et al , 2008.
The consequences for livestock of poor nutrition over winter are higher adult mortality, lower fertility and birth rates, and increased risk of disease. For mountain households that cannot afford to obtain sufficient quality and quantity of winter feed, this leads to a cycle of poverty, because their flocks and herds cannot grow due to low reproductive rates and more animals must be sold to support the remaining few (AKF 2004(AKF , 2005(AKF , 2006. Many mountain households fail to enter rural commodity markets and to move beyond subsistence production. Emergence of secure and profitable value chains for livestock products is hampered due to remoteness, poor raw material quality, absence of disease controls, and insufficient price information (Ajibekov 2005;Nä scher 2009;Steimann 2011). In Kyrgyzstan, a large share of rural households depends on small amounts of state welfare eg child allowances and oldage pensions. There is also increasing loss of mutual trust and aid among rural households, a further cause of impoverishment (Kuehnast and Dudwick 2004;Farrington 2005;Sabates-Wheeler 2007;Steimann 2011).

Changes in livestock species kept
Each species of livestock has specific impacts on pastureland through their foraging habits and preferences for different palatable vegetation species. The species composition of livestock has changed considerably in the past 20 years (Table 2). In 1991, Kyrgyzstan's mountains contained 2.5 to 3 times the number of sheep that are now kept. However, the number of goats in private flocks has more than doubled (FAO 2011(FAO , 2012. In Tajikistan, there has likewise been a great increase in the recorded number of goats, but unlike Kyrgyzstan, the numbers of sheep have risen slightly, again after a steep decline in the 1990s (FAO 2011(FAO , 2012. The rise in goat numbers relative to the other livestock species is related to the impoverishment of mountain villagers. Goats compared to sheep are more prolific, cost less to buy, and are easier to herd in these hilly environments (Kerven et al 2009). All these factors have made goats more attractive for poorer households. Local government authorities in these countries comment that they are worried that the increase in goats may be causing pasture degradation through overgrazing. However, no field research has been conducted in this region to test the impacts on pastureland of the rising goat numbers versus sheep and cattle.

Rural outmigration and lack of herding labor
The lack of economic opportunity in these mountain areas has led to massive, mostly male, outmigration to national urban centers and to international destinations, mainly southern Kazakhstan and Russia (Olimova and Figure 3). The shortage of adult men has increased the cost for shepherding livestock farther from settlements. There is also a constant increase of livestock numbers partially financed through remittances (Eggenberger 2011). These changes may be aggravating the pressure on pastures around settlements that is already occurring due to other factors discussed here.

Changes in pasture tenure regulations
There has been a fundamental shift in the legal forms by which pastures are held. The application of new pasture tenure laws has had considerable effects on how pastures are used and thus the potential for overuse. In Kyrgyzstan, legal changes instituted in 2002 based pasture use on territorial leases, to be obtained by individuals or groups from local administrations (Undeland 2005;Liechti 2008). The pasture lease system was complicated and had the unintended effect of creating generalized open access of pastures. This led to overuse of more accessible pastures, as less wealthy villagers were effectively excluded from more desirable but more remote pastures (Jones 2003;Lerman and Sedik 2009;Steimann 2011). In 2009, a new law in Kyrgyzstan abandoned the lease system and instead transferred all administrative authority over pastures to so-called grazing committees at the local  community level. However, comparatively wealthy and wellconnected households can often afford to secure de facto exclusive access to a large pasture area through the construction of a barn on the winter or spring-autumn pastures (Steimann 2011). Such informal-yet not illegal-exclusion from formerly common property grazing areas can lead to further grazing pressure on the residual grazing areas that remain open access and unclaimed. Similarly, in Tajikistan individuals can obtain use rights to pastures through negotiation with the state administration at a district level (Peyrouse 2009;Sedik 2009;Rowe 2010). The current legislation is unclear about when and how pastures may be privatized or leased by individuals. Large herd owners increasingly tend to privatize remote seasonal pastures, while smaller owners lose access to these pastures and are eventually forced to overgraze other, more accessible areas (Ludi 2003;Robinson and Whitton 2010).

Climate change and pasture degradation
Climate change trends and projections in Central Asia could have important implications for future pasture degradation. Annual average temperatures are steadily rising in Kyrgyzstan and Tajikistan at a rate similar to or greater than average global temperature rise (Aizen et al 1997;Giese et al 2007;Savitskaya 2010). Projections for mountain areas are difficult because of the inherent climate variability of mountains. Climate model projections for Central Asia, however, include warming much greater than the global mean (3.7uC by 2100 compared to an 3uC globally), maximum warming in summer months, and a greater temperature increase in highelevation areas (Christensen et al 2007).
Pasture productivity, hay fields, and fodder crops are strongly influenced by climate conditions. The 2007 International Panel on Climate Change report concludes with a high level of confidence that Central Asia is very vulnerable (highest rating) to land degradation from climate change impacts (Cruz et al 2007). Research from Kyrgyzstan and Tajikistan concludes that drying associated with higher air temperatures could cause a significant reduction in the productivity of certain pastures (GoTJ 2008;GoKR 2009). While warming temperatures will result in a longer growing season that may benefit certain pasture plants and fodder crops, increased drying and precipitation variability (including drought) are likely to negatively affect pastures in particular (Tebaldi et al 2006). Climate change projections for warmer and drier summers are significant for agriculture, yet extreme climate events are equally or more important factors (Lioubimtseva and Henebry 2009). Livestock production is particularly sensitive to drought, while aridity is already a limiting factor in much of this region (Lioubimtseva and Henebry 2009).

Conclusions and research priorities
The most prominent research and development topics on Central Asian pastoral regions over the last two decades have been pasture management and mismanagement and, linked to this, land degradation (Kerven et al 2011). Much money has been spent by international donors, the United Nations, and international NGOs on short-term research and projects to improve pasture management. However, many of the reports we reviewed were superficial, derivative, and nonempirical. Nevertheless, they have been influential in attracting still more donor funds to combat ''degradation.'' Our first conclusion is that further research is needed to confront and test these ''environmental orthodoxies'': A great amount of ''development policy'' has often been driven by simplistic, and even scientifically unsupported, assumptions, [for example] the collection of environmental orthodoxies embedded in the ''Theory of Himalayan Environmental Degradation.'' The sheer simplicity and intellectual attractiveness of this particular orthodoxy has ensured its survival despite its effective scholarly rejection. (Ives 2001: 132-144) The promulgation of environmental orthodoxies suggests there may be a political economy of ''degradation discourse'' arising from the intertwined interests of researchers, NGOs, donors, and governments. Such a discourse can create incentives for researchers and research organizations to disseminate scare stories about the disastrous conditions of the land (and now the climate) that provide governments with justification to press for certain changes and additional funding. One of the difficulties for researchers is to be objective and independent yet obtain funding.
Donor and national governmentsupported projects have attempted all kinds of pasture management schemes, convinced that they were needed to halt degradation and desertification and to improve pasture productivity. (For a few examples, see World Bank 2003, 2007CARNET 2005;UNDP 2007;Ji 2008;UNDP and GEF 2008;ADB 2009;ADB 2010;UNEP 2011;UNU-EHS 2012.) Nevertheless, conclusions about whether, where, how, and why degradation and desertification are occurring, and what methods could be used to tackle these processes, have been based less on updated field-based evidence and more on untested orthodoxy. ''Overgrazing'' or ''overstocking'' is often cited in these reports as causing land degradation. But new field studies assessing the causes, effects, characteristics, and implications of grazing and pasture degradation in Central Asian mountains are applying careful measurements that raise questions about any simple correlation of overgrazing and land degradation (Bimü ller et al 2010).
The first research priority is to inventory the impact of the many pasture development projects and new pasture tenure legislation in the mountains. What was the uptake of the various pasture improvement methods demonstrated and advocated? Did mountain villagers who depend on the pastures apply these methods, and did the villagers obtain project results that were useful to them? What were the results? Are the benefits replicable without external technical funds and assistance? Who benefitted? What were the costs? Who bore the costs?
A second research priority is to assess the scientific basis for the interventions proposed and promoted to the pasture users. One reliable assessment of the reported findings is whether they refer to research results published in scientific, peer-reviewed international sources. A check through the bibliographic search engine Web of Science reveals that since the late 1980s there have been no scientific articles published with new data on pasture degradation in Tajikistan and only one article on effects of deforestation on soils in Kyrgyzstan. The same pattern is found by searching Google Scholar.
The reports about pasture mismanagement, degradation, and the need for rehabilitation in Kyrgyzstan and Tajikistan often repeat previous assumptions and preconceptions, usually without offering fresh data. There is a clear need to do more in-depth field work, followed by modeling, on the multiple interacting causes and feedback effects of changes in the soil, vegetation, climate, and animal populations-both livestock and wildlife-to understand the biophysical impacts of the profound changes in land management over the past 20 years.
Only when we have new and reliable data will we be able to say whether any practical measures can be taken to improve pasture management and still benefit the land users. Without sound data that tests the current assumptions, there is a risk that land users-farmers and pastoralists-will continue to be blamed for despoiling the land through bad management. This can provide a rationale for governments and their donor supporters to redistribute land through privatization, as is occurring in Tajikistan with World Bank support, or to exclude pastoralists from their land on the justification that this helps conserve vegetation cover, soil, biodiversity, river headwaters, etc, as is being implemented by the Chinese government in the mountainous pastoral regions neighboring Central Asia (Harris 2010;Xinchun 2011).

Community participatory pasture management
Another set of research priorities stems from the current policy trend to decentralize pasture management from the national to the local level to make it more ''participatory.'' In Kyrgyzstan, the World Bank's efforts to establish standardized communal pasture user committees seem to be influenced by rather simplistic ideas. Jacquesson (2010) argues that this new Kyrgyz pasture law rests on the longstanding misconceptions of ''clan,'' ''custom,'' and ''tradition'' as social institutions that would allow for a level playing field, thus ignoring the often unequal relations among local herders. After 1991, romanticized notions invoking ''nomadic traditions'' have become increasingly popular, not least because they have been politically useful to the government of Kyrgyzstan. Empirical evidence shows, however, that nowadays the mountain pastoral communities are anything but homogenous and are instead characterized by striking disparities in terms of wealth and power. Consequently, we would welcome more mutual exchange among donor agencies, development practitioners, and researchers, requiring a critical dialogue about assumptions and priorities and longterm scientific monitoring of the implementation and effects of particular development interventions. Unfortunately, though, development projects often operate with a much shorter time horizon than field research projects.

MountainNotes
The towering mountains of Central Asia hold a great appeal to certain people-among others, geologists, botanists, wildlife biologists, conservationists, anthropologists, hikers, bikers, ecotourists, development workers, glaciologists, geographers, climatologists, and livestock scientists. But after the researchers, development agents, and tourists have come and gone, the mountain dwellers remain. They deserve a longterm commitment to understanding their problems and assisting with their efforts to find solutions.