44. Spontaneous range enclosure movements reverse the usual relationship between government planners and the pastoral populations which are the targets of their planning. Planners and project designers generally assume that pastoralists must be instructed, coerced or inticed through appropriate incentives to respond to government development efforts. With spontaneous range enclosure, however, the impetus for change comes from below, from the livestock-keepers themselves, and while final costs and benefits may be difficult to predict, it is clear that these local-level initiatives will produce fundamental social, technical and economic changes. In these cases it is the local communities which have taken the initiative, and planners, administrators and technicians must respond appropriately to changes they have not initiated, and processes which they do not fully control.
45. Policy responses to range enclosure can take three different forms:
(i) Because of their negative backwash effects on non-participants, enclosure movements tend to polarize public opinion. In this environment one of the most useful functions administrators can perform is to arbitrate between those who favour and those who oppose enclosure. Aside from their coercive power, which will be discussed below, the authority of administrators as arbitrators rests on their ability to argue persuasively based on reliable information and a clear understanding of the issues. The preceding discussion has outlined the questions which applied research would have to answer in order to sustain the credibility of administrators as impartial arbitrators.(ii) Administrators may also attempt to control enclosure by suppressing, encouraging or simply regularizing the process. In this case emphasis will lie on routine administrative action designed to establish a degree of central control over local-level events. The following discussion examines the possibilities and problems of regulating enclosure through the direct exercise of administrative authority.
(iii) Finally, government intervention may be directed at both ameliorating the negative effects and exploiting the potential benefits of enclosure. In this instance emphasis will shift from government regulation to the provision of inputs, services or improved agricultural techniques which will encourage voluntary compliance with government's development objectives. The final section of the paper will suggest ways that enclosure movements can be indirectly managed through the provision of critical inputs.
The limits of regulation
46. The ability of government agencies to regulate range enclosure directly depends on their capacity to manipulate the factors which cause enclosure, as illustrated by the Sudanese and Somali cases. Unfortunately, the demonstrated capacity of administrators to control events in extensive pastoral areas is, and usually has been, limited, whether the setting be contemporary Africa or historical periods of ranch development in industrial nations. The limits to administrative control arise, in part, from the nature of the pastoral populations which administrators confront, populations which are often mobile, self-reliant and locally organized. The financial limits of realistic administrative initiative may also be set by the low productivity and revenue generating potential of dry pastoral areas. Faced with comparatively low tax revenues relative to the high costs of governance, administrators may be forced to limit their activities to those interventions which have the highest chance of success. There follows here a point-by-point examination of the extent to which administrators can expect to control the different factors which cause enclosure.
Commercial fodder and livestock production
47. African governments are committed to increasing all kinds of commercial agricultural production. Given this consistent policy, administrators are in a position to indirectly encourage fencing by encouraging the growth of commercial agriculture and business interests at the local level. Attempts to retard or suppress enclosure will, however, run counter to the stimuli provided by general government policy and long-term economic pressures, and it is doubtful whether such policies would eventually prevail.
Overstocking and drought
48. The failure of two decades of range management projects has demonstrated the administrators have little ability to control stocking rates in order to dampen the effects of periodic drought. Overstocking on common ranges and drought are therefore likely to recur and to continue to provide incentives for enclosure in many areas.
49. One of the possibilities raised by range enclosure is that it will induce herd owners to control their own herd sizes in order to conserve their own rangeland. However, herd owners will only be inclined to limit their herd sizes when most of the range around them has already been subdivided and appropriated by others. In the short term, range enclosure will exacerbate the problem of overstocking on the open range by withdrawing parts of the range from communal use, and by forcing more livestock into the remaining area. The pressure for further enclosure will therefore increase as individuals watch the commonage shrink and attempt to grab their piece of it before it is too late. In this way enclosure movements build within themselves pressures for their own expansion, pressures which administrators may find very difficult to effectively counteract.
Borehole development
50. Technicians and planners have often drawn up water development guidelines which prevent the clustering of boreholes. Nevertheless, in few areas of tropical dry Africa has it been possible to implement these comprehensive plans consistently. Part of the problem is attributable to the scarcity of suitable drilling sites and the irregular distribution of these potential sites. Political and business interests also routinely influence the number and locations of new bores. Because of the difficulties of implementing water development plans, the irregular maintenance of existing wells has often an important part in preventing overgrazing by reducing the total number of functioning wells (Sandford, 1983). Improved servicing of wells could exacerbate the overgrazing situation and create further pressure for enclosure. Thus, while water management and planning cannot be ruled out in theory as a means of regulating enclosure, practical impediments will likely limit the effectiveness of this means of control.
Local political institutions
51. Many independent African governments have been hostile to tribal political organizations in pastoral areas because of the threat these organizations pose to central authority and national unity. Attempts to replace indigenous groups with government - sponsored organizations such as group ranches have not proved notably successful (Oxby, 1981). Administrators responsible for range and livestock development may therefore be in a position to encourage enclosure by discouraging traditional forms of competition over the use and ownership of the range. They are much less likely to be able to discourage enclosure by promoting effective local institutions which are capable of owning, managing, and defending group interests in communal land.
Land law and taxation
52. Perhaps the greatest opportunity for government control lies in its ability to manipulate the social, legal and financial costs associated with enclosure. Even here, however, the room for manoeuvre would appear to be limited. The enclosure movements discussed in this paper were all technically illegal in terms of their respective national land laws, as these laws were interpreted by local administrators. Administrators may have the capacity to encourage enclosure through new legislation, or by failing to enforce existing legislation, but anti-enclosure land laws alone do not appear to be a sufficient deterrent to fencing, if other factors encourage it and if officials are unwilling or unable to use force. Official recognition of enclosures and their taxation may prove more effective.
53. The preceding discussion suggests that administrators have at their disposal the means to encourage the fencing of rangeland, but that they have little capacity to oppose fencing effectively, should they want to do so. It would appear that spontaneous enclosure movements are generated by forces which planners only marginally control, and that these movements possess a momentum of their own. The management of enclosure movements should not, therefore, be restricted to attempts to regulate and control them. Administrators should also consider how best to exploit the development potential of these movements which are likely to persist whether or not they are officially condoned.
