Johnson (1969) identifies the combination of animals herded and the role that agriculture assumes in a pastoral group's economy as being the most influential factors determining migration. The first and most obvious response to drought is to move the animals to areas where there is still pasture and water. This is probably the major drive for the expansion of pastoralism, especially in the case of the eastward expansion of the Fulše across the West African savannahs. In the precolonial era, pastoralists were limited principally by disease and, more occasionally, insecurity. In the twentieth century, these took second place to the occupation of land by cultivators and the consequent presence of boundaries that impede free passage.
The migration of pastoralists to areas of higher productivity alleviates stress on less productive or exhausted land. Conversely, if the movement of pastoralists is restricted, land that is already marginal becomes even more overused. Johnson (1975) observes that, when pastoralists make long journeys, stock deaths increase and pastoralists must weigh likely losses from the migration against comparable losses were they to stay on suboptimal land.
The creation and maintenance of corridors reinforces cooperation between the agriculture and pastoral sectors. However, corridors that are too long or too narrow tempt hungry animals to graze on the crops on either side; pastoralists have to use more labour to keep their herds under control, and the potential for aggravating the conflict between cultivators and pastoralists is apparent.
A major exception to this occurred in Central Asia during the Soviet era when an extensive military infrastructure, which controlled pastoral movement, was based on unrealistic cost structures. Another curious exception was Israel after 1967, when extreme militarization of the Negev led to major constraints on Bedouin ruminant production through the registration of herds, prohibitions on traditional migrations and, surreally, campaigns against black goats (Abu-Rabia, 1994).
A long-term recovery strategy and insurance against the impact of future droughts is changing the species in the herd. Although cattle are prestigious and highly valued in the market, they are vulnerable to drought in comparison with camels and goats. The relatively high rainfall in the 1960s encouraged pastoralists across the Sahel to switch from camels to cattle, even among populations such as the Tuareg (in Mali), who have historically been identified with camel culture. The droughts of the 1970s demonstrated that this was an unwise strategy, and drought recurrence in the 1980s underlined the point.
These types of changes in herd composition can also be applied within species. In West Africa, cattle breeds that specialize in grass are more prestigious than those that can digest a high proportion of browse. However, where low rainfall or high grazing pressure has changed the species composition of the landscape so as to favour shrubby vegetation, the herder with cattle that can tolerate a higher proportion of browse in their diet will survive better. In a concrete example, Fulše herders in Nigeria, faced with rapidly vanishing grass in the semi-arid zone, have switched their herds from the Bunaji breed, which depends on grass, to the Sokoto Gudali, which can digest browse far more easily (Blench, 1999a).
Strategies relating to species diversification vary; there are advantages in owning a variety of species so that, whatever climatic events occur, there will be survivors. For example, the multispecies herds that are typical of Mongolia and the Andes may well be a reflection of extreme climatic variability. However, maintaining such herds is a luxury that only the wealthier can afford. Herds of different species are generally split up, most commonly into browsers and grazers, so that the available forage can be exploited most effectively.
Within species, herd diversification takes place during a drought. Productive animals, particularly females, receive priority treatment, while the bulk of the herd is sent to find pasture further afield. This allows milk to be obtained from the subsistence herd, or from relatives, while the rest of the herd does not exhaust the grazing (Dahl and Hjort, 1976). Larger animals, particularly camels, although resilient in a drought, will die in numbers after a critical point. After drought, smaller stock reproduce more rapidly, allowing the herd to recover and serving as capital which can be exchanged for larger animals later on. Rebuilding a herd of camels, for example, is a slower process.
In the pre-modern era, predation on pastoral herds was a major concern of virtually all pastoralists and a constant demand on herding labour. The expansion of agriculture and the spread of modern weapons in the early twentieth century largely eliminated predators from whole ecosystems, for example wolves, lynxes and leopards in circum-Mediterranean systems, and hyenas and lions in West Africa. In the Soviet era, military-style collective hunts against predators in Central Asia and Siberia substantially reduced the impact of predation, although elimination was never practical. Predation remains a significant threat in the Andes, where pumas and foxes often take young animals (Göbel, 1997).
However, external changes are affecting views of predation and, thus, attitudes towards the wholesale elimination of predators. Many species, such as wolves, bears and snow leopards, are now seen as endangered - and are therefore the object of conservation efforts - rather than as a nuisance to be eliminated. Projects have been established in Mongolia, for example, to encourage herders to conserve snow leopards by accepting the losses and providing alternative sources of income to compensate for them. In Namibia, where more traditional livestock ranches are increasingly interspersed with wildlife enclosures, predators such as leopards and cheetahs are increasing, partly because a certain level of predation is accepted as necessary to the health of a wildlife stock. Similarly, in North America and Scandinavia, the reintroduction of wolves into national parks has caused considerable controversy, as they inevitably pass beyond the boundaries of the park and kill livestock outside. Such losses are not acceptable to livestock producers and the rather mythic status of wolves has allowed producers to articulate panic messages that are somewhat out of proportion to the wolves' actual depredations.
The consequence has been a policy war between these competing interests, which has been played out in front of an interested media. There is little or no doubt that environmentalist and conservation concerns will win; the strength of these lobbies across the developed world is constantly growing and they are well funded and articulate. When it is picturesque, wildlife can often generate more hard currency for national economies than pastoralists can, and this is what counts with policy-makers. Moreover, it has become clearer that there are ways of developing interlocking wildlife and pastoralist systems that allow both systems to flourish; such types of co-conservation are beginning to appear in East Africa (Bourn and Blench, 1999).
A problem that is rarely addressed by livestock services, but that weighs heavily in the investment decisions of livestock owners, is the prevalence of theft. It is not worth investing in quality animals if it is likely that they will be stolen. Owners will not pay for supplementary feeds if the only effect of fattening animals is to increase their attractiveness to thieves.
Livestock raiding has something of a romantic history, and in some parts of the world, such as the Kenya-Ethiopia borderlands or Madagascar, the successful rustler gains prestige (e.g. Fukui and Turton, 1977; Todd, 1977; Bollig, 1990; Turton, 1991). There is, moreover, an argument that chronic raiding helps to maintain an ecological balance (e.g. Sweet, 1965). However, more commonly, theft is practised by urbanized individuals who have links with the market system and can sell their haul rapidly to butchers. Cattle and sheep are the animals that are most commonly stolen, because they have relatively high market value and can be driven away on the hoof. Livestock theft is problematic in administrative terms; national authorities are often unsympathetic to pastoralists and perceive the expenditure of policing resources on lost sheep as futile. The consequence of this is that, not only do herders need to allocate considerably more resources to guarding animals, but also they tend to arm themselves and treat stock raiders to summary justice. This, in turn, tends to aggravate the authorities still further, as they then see pastoralists as the founders of unregulated militias.