Somalia - Agricultural sector review (Vol. 3 of 3) : Annexes 4-7

Report No. 2881a-SO Somalia Agrcultural Sector Review FILE COPY (In Three Volumes) Volume IlI: Annexes 4-7 lune 29, 1981 Eastern Africa Regional Office Northern Agriculture Division FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its conterts may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit - Somali Shilling (So.Sh.) US$0.1589 5 So.Sh. 1.00 US$1.00 5 So.Sh. 6.295 WEIGHTS AND MEASURES 1 hectare (ha) 2 10,000 m 1 square kilometer (km ) 100 ha 1 metric ton (ton) 1,000 kg ABBREVIATIONS ADC - Agricultural Development Corporation ENC - National Trading Company FYDP - Five Year Development Program GDP - Gross Domestic Product LDA - Livestock Development Agency MLFR - Ministry of Livestock, Forestry and Range MOA - Ministry of Agriculture MNP - Ministry of National Planning NBB - National Banana Board NES - National Extension Service NRA - National Range Agency ONAT - Farm Machinery and Agricultural Services Organization SPC - State Planning Commission TYDP - Three Year Development Program FISCAL YEAR January 1 - December 31  FOR OFFICIAL USE ONLY - 207 - ANNEX 4 SOMALIA AGRICULTURAL SECTOR REVIEW The Spcial Impact of Agricultural Development Table of Contents Page No. I. INTRODUCTION .............................................. 1 II. SOCIAL AND CULTURAL RESOURCES ............................. 1 A. . General .............................................. 1 B. Population ........................................... 2 C. Unifying Factors ..................................... 3 D. Traditional Socio-Political Structure .... ............ 4 E. Indigenous Systems of Production and Exchange .... .... 5 III. SOCIAL TRANSFORMATION ...................................... 10 A. The Role of the State ................................ 10 B. The Growth of State Organizations .................... 14 C. Redistributing Population ............................ 16 D. Summary and Conclusions .............................. 18 IV. PROBLEMS AND PROSPECTS IN SOCIAL DEVELOPMENT .... .......... 22 A. General .......... .................................... 22 B. Self-Reliance and Self-Sufficiency .... ............... 22 C. Scarce Resources and Equitable Development .... ....... 23 D. Popular Participation in Development .... ............. 24 E. Institutions ............ ............................... 25 F. ;Missing" Institutions ............................... 26 TABLES 1: Population, Area and Population Density by Region 2: Population by Settlement Type, Region 3: Estimated Average Number of Livestock per Agricultural Household 4: Earnings Differentials in Government Employment 5: Earnings Differentials in SNAI Sugar Factory MAP: Regions and Districts (IBRD 15413) This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  - 208 - ANNEX 4 Page 1 SOMALIA AGRICULTURAL SECTOR REVIEW The Social Impact of Agricultural Development I. INTRODUCTION 1.01 The objective of this annex is to review and analyze the impact that recent development policy has had upon Somali society. Since 1970 at least, when the Second Charter of the Revolution articulated the goal of social justice through the philosophy of scientific socialism, the Somali Democratic Republic has been clearly committed to a social policy of development with equity. Since over eighty percent of the population is directly engaged in agricultural production, an analysis of the social impact of agricultural development can measure the degree to which overall Somali social policy is being achieved in practice. The first part reviews the country's social and cultural resources, and the basic elements with which the development strategy must work. 1.02 The second part examines the role of the state and its institutions in transforming Somali society, and describes the consequences of political and social forces on rural producers. The final section focusses on problems in, and prospects for, the achievement of social objectives and contains our recommendations for government action to resolve some of the dilemmas posed by the gaps between the theory and the practice of societal development in Somali conditions. This annex makes no judgment on the validity of Somalia's social goals, but only reviews how far government is now actually going to realize the goals it has set for itself. II. SOCIAL AND CULTURAL RESOURCES A. General 2.01 The concept of "human resources" as generally used is far too narrow to encompass the human factors that are so important in the development process. Society is much more than the sum of its individual members' energies, skills, and potentials that can be harnessed for new endeavors. Rather, it is a complex structure of groups, activities, interests, and relationships. This structure must be understood, just like any other structure such as a soil or hydrogeological formation, before any decision on how best to mobilize it for new undertakings can be taken. Existing patterns in the distribution of social groupings, in the strategies by which groups attempt to achieve their  - 209 - ANNEX 4 Page 2 own goals, or in the relationships they have with one another can be obstacles to particular types of change or they can be the building blocks for new ideas and structures. An adequate description of the social and cultural "landscape" is thus a necessary first step in assessing social policy. B. Population 2.02 The data yielded by the national census of 1975 is currently being processed. The provisional figures which have so far been released permit a general overview of population distribution, but do not yet allow detailed statements of age and sex ratios, the total labor force, sizes and types of settlements, or the like. It is expected that the completed analysis will be available for the purposes of the next full round of development planning now getting under way. 2.03 The current revised figure for the total 1975 population is 3,722,000. 1/ Accepting this figure, and further assuming a net annual growth rate of 2.5 percent estimates vary from 2.3 to 2.7 percent, the population in mid-1979 would have been over 4.1 million. Of the 1975 totals, 59 percent were classified as "nomads," 22 percent as "settled farmers," and 19 percent as the "inonagricultural" population. Population by Region 2.04 Somalia is now divided into 16 administrative regions, as indicated on Map 1. For general planning purposes, these regions are sometimes grouped into six macro-regions which correspond to broad differences of geography. The area and population total and density for each region are shown in Table 1. 2.05 In general the lower rainfall areas of the northeast and central rangelands carry the lowest and 1-^st dense populations. Two areas bear the highest densities: the first is formed by the two major river valleys of the south and the better-watered Bay Region between the rivers; the second is the northwest, the primary area of intense commercialization of livestock. In rough terms of economic geography, the country can be characterized as two zones of relatively concentrated activity connected by a large area which is sparsely utilized. It is to be expected that one effect of the recent paving of the highway linking the zones will be an intensification of economic activity in the intervening zone. 1/ This total is 230,000 more than the provisional figures which appear in earlier government and IBRD documents. Since those figures were released, Lower Juba Region has been divided into Lower Juba and Middle Juba Regions, and the total for the two, as now given, is 193,000 more than the total for the former single region. The new figures also give Middle Shebelli Region 37,000 more people than before. In any case, the census was carried out in the disturbed conditions of the longest drought in modern history, and overcounting may have occurred.  - 210 - ANNEX 4 Page 3 Population by Settlement Pattern 2.06 Three categories of settlement type were distinguished in the 1975 census (Table 2). The "non-agricultural" population of the census was defined as the population of the regional and district capitals. This category, therefore, does not strictly coincide with 'urban population', since some towns are not capitals. Furthermore, non-agricultural workers in areas outside the capitals are not included while farmers and herders who happen to live in the capitals are. The "settled farming" population means people in permanent villages. This category, therefore, includes crop farmers (except as above), but it also includes many people who may live in permanent settle- ments but who derive the principal part of their income from pastoral activities. In Somalia there is a continuous gradation from non-livestock-owning farmers or plantation workers to sedentary but full-time pastoralists. This category thus includes a large but unspecified number of people who are classified in some documents on Somalia as "semi-nomads." The "nomadic" population includes all those enumerated outside permanent settlements. It is evident from the table that the nomads are not restricted to the rangelands of the central and north-eastern macroregions, but form a fairly high proportion of the population of those regions where crop farming is practiced. In only three regions, Benadir (Mogadishu itself), Lower Shebelli and Bay (where substantial numbers of people have settled), do nomads constitute less than 60 percent of the total population. Nomads are the constant feature of the Somali social landscape, settled people (62 percent of whom live in four of the sixteen regions) the special case. 2.07 The large proportion of nomads poses special problems for Somalia (as for its sister "pastoral republics," Mongolia and Mauritania). Census- taking, planning, provision of social services and general administration are all complicated by the mobility of the population. The government has embarked upon a general policy of settling nomads, but realizes that this can only take place as sufficient resources, which nomads now have to move to exploit, eventually become available at permanient sites. In fact the frequency with which pastoral Somalis change residence, and the regularity of their migration routes, vary by area and over time. Some are almost sedentary but live in dispersed homesteads, while others move frequently and opportunistically over a large territory. Most spend at least four months of the long dry season (jilaal) at or near a permanent water point. Even when they move, they range within broad areas associated with their descent groups. Nomadic "anarchy" is an exaggeration; planning can proceed once detailed knowledge of environmental constraints on movement is assembled. C. Unifying Factors 2.08 Somalia-s development resources include the relative unity that exists in religion, language, culture and historical outlook; as far as these are concerned, the Somali people have always been a nation. Only recently, however, have most of them been in a single state. This concensus manifests itself concretely: civil servants, for example, can be posted anywhere in  - 211 - ANNEX 4 Page 4 the country without feeling alien in language or culture, and can therefore begin working immediately; written communication is clear and well-understood by all; the vision of all Somalis as underdogs in the partition of their historic lands by more powerful foreigners underpins a loyalty to the nation which has endured over time and has been mobilized by successive national leaders from earliest days. The uniformity in even relatively esoteric subjects like the folk systems of astronomy, measurement, and edaphology, can enable extension workers, researchers, or curriculum designers to make quick progress with their clients unencumbered by conflicting ideas or foreign concepts. D. Traditional Socio-Political Structure 2.09 The creation and expansion of colonial administrations in the first half of this century wrought profound changes in the organization of Somali society. The independent state of Somalia has further altered the basic social and production processes, and will be examined in Part II below. But the traditional structure still exists, although it is being transformed and weakened by the growth of the modern state. That structure represented a relatively successful adaptation to environmental and social constraints at a low level of technological complexity. It enabled Somalis to occupy a vast and often inhospitable land, and it provided basic subsistence with social and physical security. It still provides many Somalis with their most significant relationships to other people and to the resources necessary for their daily life. It is this structure that the state seeks to develop, change, or replace. 2.10 The fundamental idiom in which the traditional world was understood was that of kinship. Membership in one of the six broad patrilineal descent groups (technically, clan-families) gave each Somali an ultimate claim to a place in the social order. Belonging meant sharing the overall rule of customary law and having a stake in the particular resources of land, water and power that each clan-family defended as the patrimony established by the clan's founder. Four of these clan-families were largely pastoral by occupa- tion -- the Darod, Dir, Isaq, and Hawiye -- while two were primarily agricul- tural -- the Digil and Rahanwin. Outside this structure in what is now the Somali Democratic Republic lay only the Arabized communities of the coastal trading towns, some Galla-speaking pastoralists and farmers, and Bantu-speaking cultivation groups who pushed up from the south into the riverine areas and were largely incorporated into the agricultural clans of the Somali. 2.11 The name of one clan-family provided the individual with a basic "address" (as l.M. Lewis points out), a location in the social structure. Each lower-order segment of one's genealogy identified one more specifically, and the relations between segments at greater or lesser genealogical distance provided guidelines to each person on how to behave toward others. This identification was important in part because small groups were frequently physically interspersed with others from many different clans. For while  - 212 - ANNEX 4 Page 5 each of the large clans and clan-families was identified with from one to three or four large blocks of territory, the exigencies of the environment meant that individual groupings might range widely into areas beyond the zone ordinarily most securely occupied by themselves and their close kin. 2.12 Among the pastoralists it was the dia"-paying group, however, that was the most effective mobilizer of ongoing loyalties and obligations. Formed from among close kinsmen (but not necessarily including all of them), this group entered a collective formal contract (or treaty, "heer", in Somali) to pay and collect compensation for injury and death. By sharing political risk and using physical force when necessary, the dia-paying group provided its members basic social and physical security. The group acted only when necessary, however, and from day to day there was no centralized political authority over the actions of individual herding units. The moral force and influence of Islamic religious leaders helped to bind the otherwise centrifugal social forces, while wealth, intelligence and tact could earn a man considerable influence within his lineage. 2.13 Among the cultivators of the south, control over access to the limited resources of land and water enabled clan and lineage leaders to wield more power than among the pastoralists where these resources were more widely dispersed. Chiefly hierarchies were therefore more concretely established, and the statuses of stranger, client (which status was formally abolished in 1960), and full clansman were more rigorously distinguished. Nonetheless, cultivating villages regularly comprised people of several different lineages, and village leadership was in the hands of the elders of the constituent clan segments. Villages as such were not aggregated into wider political units, but a solidarity grew across lineage segments within the village through water point construction, village maintenance, and agricultural task sharing. 2.14 Access to other major factors of production, labor and capital were also regulated through the system of kinship. Islam, and specifically Somali custom, guided the choice of marriage partners, in part on the basis of the wider political alliances they would encourage. Rules allowing polygamy and strictly guaranteeing female chastity before and after marriage permitted pastorialists to disperse the family labor force as necessary to provide for the different needs of various classes of livestock. Inheritance patterns kept lands and livestock in family hands, usually in the male line. E. Indigenous Systems of Production and Exchange Pastoralism 2.15 The vast majority of the Somalia labor force is engaged in pastoral activities. For the majority of the population, pastoral production is the major subsistence mode. Most Somali cultivators also keep substantial herds, and even town-dwellers usually own a few animals. Few Somali own no livestock at all. Pastoralism is labor-intensive as well, in that it involves constant,  - 213 - ANNEX 4 Page 6 if not especially strenuous activity to care for the herds. Some animals, goats for instance, are easier to herd than others, and less labor is required during certain parts of the year. In general, however, animal husbandry keeps all family members busy from an early age (in crop-growing families, children of nine to twelve are often responsible for the sheep and goats). The slack periods associated with crop farming (when educational outreach, for example, can proceed) are much less pronounced among herding communities. 2.16 Pastoral production is carried out as a household enterprise, with the senior male of the nuclear or composite household as herd manager. Individual animals may be owned by household members other than the head, and those owners may thereby have a subsidiary role in major husbandry and offtake decisions. Some animals may also be in the herd on loan or on contract, and may not be alienable by the herd manager. Calculations of potential offtake from a herd must take into account the differential alienability of individual animals within it. 2.17 The mix of livestock species that one family holds varies according to local ecosystemic factors such as disease vectors or the type of pasture available, to family fortunes and points in the domestic cycle, and, increas- ingly, to demand factors from the commercial sector. Table 3 gives the average family herd by livestock types and by regions of the country from the 1975 census, but the figures (a) are likely to derive from relatively inacurrate counts, and (b) certainly mask wide variations among families in both the numbers of animals owned and in the mix of species in individual herds. Table 3 demonstrates, at the most general level, the higher numbers of sheep and goats in the non-agricultural areas of the north and the concentration of cattle toward the better-watered sodth. According to the census, it is the intermediate climatic zone in the central rangelands that has the largest family herds in the country. 2.18 Grazing management is cmnplex. It requires quick responses and great flexibility to take advantage of the sporadic and fleeting resources of water and grass in the arid lands where most pastoralists live. Somalis have adapted to their particular ecosystem in three primary ways: (a) By keeping goats and sheep plus either cattle or camels (rarely both), they make efficient use of the vegetation of individual and adjacent pastures and minimize the disease risk; (b) By oscillating between dispersed grazing in the two rainy seasons to harvest the vegetation of areas as far from permanent water sites as is practicable, and more concen- trated grazing in the dry season around the permanent water, they in effect practice a rotational system which reserves pasture for dry season needs; and  - 214 - ANNEX 4 Page 7 (c) By subdividing the herd into segments, with the bulk of the camel herd (halweyn) tended far away from the rest of the family for much of the year by the young men of the family, and sometimes also by keeping wives as sub-herd managers in different locations, they further minimize risk at the cost of dispersing household members. 1/ 2.19 Herd offtake in traditional conditions before the current era of intense commercialization (para 3.07) is unknown. Somalia did not have a culturally-elaborated emphasis on bull animals, and it may be presumed that male cattle and camels in excess of the number of potential breeders and workers were removed from herds early to increase the milk available for human consumption. In a pilot household survey carried out in 1975 by the Somali Central Statistical Department, sex ratios of the different animal classes held by settled farming and nomadic families were enumerated as follows: Sex of animals held by 104 households Middle Shebelli Region, 1975 Settled HouseholdsN------ ----- omadic Households------ Males Females M/F Ratio Males Females M/F Ratio Camels 5 2 1:4 190 404 1:2.1 Cattle 65 247 1:3.8 348 1,776 1:5.1 Goats 98 347 1:3.5 715 4,227 1:5.9 Sheep 19 58 1:3.1 414 2,802 1:6.8 The accuracy of these figures is subject to reservation and they are difficult to interpret given the drought losses that could have occurred in the two years before this survey. It can be seen, however, that in virtually all cases, the offtake of males is quite high and that the notion sometimes held that pastoralists build up their herds without considering the consequences is obviously mistaken. 1/ This strategy can complicate the task of extension education, since different decision-makers for any given sub-herd are in different places; but it is, of course, an entirely rational way to manage the family herd and the range resources.  - 215 - ANNEX 4 Page 8 Cropping 2.20 There are households engaged in farming in Somalia, mostly as dryland farmers in the area between the Juba and Shebelli rivers, espe- cially in Bay Region. In addition, flood recession farming and traditional small-scale irrigation are practiced on the banks of the two rivers. The statement is sometimes made that farming is a recent introduction in Somalia; in fact it has been practiced for at least four centuries. 2.21 As elaborated in Annex 5, dryland farms average about five hectares. With technology based on the hoe, the area actually cropped is primarily a function of labor availability at the peak period of demand during weeding. Farm families average between five and six people. Those farm families with more labor available are likely to hold more livestock and to devote that labor to animal husbandry: it appears that throughout the period for which there are any records, the use or sale of livestock products has given more benefit than additional investments in croplands. There are few data on the size of holdings per active crop worker or on the use of hired labor or sharecropping clients. A summary of the data available on farm size, labor demands and yields is given in Table 4. In these circumstances it is impossible to discuss whether the growers of most of the non-subsistence crops, or those who innovate with new crops, animal or tractor plowing, or schooling for their children are from any particular layer or type of the dryland farming families. 2.22 In most areas of dryland farming the main crop is sorghum, of which there are several local varieties. Pulses may be interplanted with sorghum, and sometimes millet and maize are also grown. The total area cultivated in any one year is divided into several plots in different areas. Given the local conditions of highly scattered rainfall without land scarcity, this practice, and that of keeping plant densities low, maximize the farmer's chances of having at least some of his planting come to full harvest. Field preparation is minimal except when the brush of a fallowed area needs clearing. Seeds are dibbled or broadcast and covered by slight ridging. Weeding is the most time-consuming of agricultural tasks, and thus proves the major bottleneck to expansion of the area under cultivation. No inputs of insecticide, herbicide, or fertilizers (not even deliberate manuring by the local herds prior to planting) are used. Since local varieties are interplanted and have different maturities, harvesting is of individual heads of grain by the use of knives. The crop of the second, shorter rainy season (Der) is less likely to succeed than that of the longer Gu, but if the year is a good one, farmers store as much as they can -- up to three years worth has been observed and farmers report stored grains lasting up to ten years in shallow pits lined with leaves, covered with branches, and sealed with soil. In all cases, crops are the personal possession of the household that grows them, although fields may revert from the household to the wider lineage group if they are permanently abandoned.  - 216 - ANNEX 4 Page 9 2.23 Traditional irrigation is done in two ways: by successively planting the alluvial soils along the rivers as the flood recedes, relying on residual moisture and rainfall; and by the construction of irrigation channels from runoff streams or from the rivers. In both cases, for obvious reasons of water use optimization, fields are smaller, closer together, and have higher planting densities than in rainfed areas. Maize is the staple crop, but beans, fruit, cotton, and sugarcane are also grown. Technology is similar to that of the dryland areas. Labor demands are higher for the preparation of the generally heavier soils and for the maintenance of irriga- tion works. On the other hand, the greater potential for cash incomes from the wetland crops, as well as the greater prevalence of tsetse fly along the rivers, correlates with the generally smaller livestock holdings of irrigation farmers. Again, data on the mix of variables such as types of labor availa- bility (adult, child, hired, etc.) size of holdings, and returns to labor from different crops are not enough to indicate what the probable effects of such factors as price changes or reliable input supplies would actually be in any of these farming systems. 2.24 Other farming systems are even less well described in the literature than the above and are less significant. These indigenously developed systems include: ex-plough cultivation in the Northwest Region; temporary cultivation of many small areas in abundant rainfall conditions (wherein nomads may opportunistically scatter seed, protect the plot with a thorn fence and return later to harvest); runoff (tug) diversion in Sanaag and other northern regions; date palm cultivation in the northeast; and vegetable gardening (including, recently, a rapidly growing number of cassava farms) around Mogadishu. The agriculture practiced by riverine peoples who spend large amounts of their time working on various plantation schemes is also unstudied. 2.25 Social factors in indigenous agriculture include the mechanisms which distributed workload and risk and therefore guaranteed a minimum of security except during general calamities. Land clearing, and the construction of wells, water harvest tanks (uar), irrigation channels, and minor public works were all often carried out by community collaboration. Membership in the group, whether acquired by birth or adoption, implied both acceptance of the tasks to be shared and the right to claim help in time of need. Clan and lineage organization and membership entailed a guarantee of security of tenure and the obligation to share in the compensation that might be assessed against one's group for damages caused by any of its members. Whether a strong principle in favor of generous redistribution of grain surpluses existed is not known: in some areas of Africa this factor helps account for the situation in which individuals, rather than reinvest in crops or hire labor, invest instead in livestock, which can be loaned out or "hidden" in distant herds.  - 217 - ANNEX 4 Page 10 III. SOCIAL TRANSFORMATION A. The Role of the State 3.01 The indigenous fabric of Somali society and culture is still intact. Indigenous cultural and economic systems have enabled Somali people to exploit a harsh environment and to exert their influence over a physical area quite large by comparison with those of most other Eastern African ethnic groups. The southwestward expansion of Somali pastoralists and cultivators has ensured the "Somalization" of the peoples from the south who pushed up into the coastal and riverine settlements of the southern part of the country. Somali pastoralists are increasing in numbers in neighboring Kenya, and Somali entrepreneurs have migrated both south and northward across the Gulf of Aden. The resilient internal processes Somalia-s culture have continued to produce poets, religious and political leaders. The 80% of Somalis who live in rural areas still provide for most of their needs either by themselves or out of resources available in their own local areas and from their neighbors and kin. 3.02 Nonetheless, there have also been profound changes in the society. Most have been brought about by the growth of the state political, administra- tive, and development apparatus in this century. It is not the task here to trace the details of Somali colonial history. But to understand the forces which underlie the development effort in Somalia it is necessary to analyze some of the effects of the growth of the state. Decline of Traditional Political Structures 3.03 Successive governments, the colonial administrations and the indepen- dent republic especially, have deliberately sought to substitute their political authority for the older authority of the clan structures. As the state has monopolized the use of force, the relevance of the clans and the contractual dia-paying groups in guaranteeing physical security has declined. Every successive government has exerted a firmer control of the instruments of force and had a stronger determination to eliminate the influence of clan structures at least in the domains of politics and of conflict resolution. There is debate as to exactly how far clan loyalties have been eroded. One authority points out that there are about as many candidates for the 1969 elections in the country as there were dia-paying groups, and that even today, despite official discouragement, clan ties provide the individual's primary identity. If clan still provides a social "address," however, it does not provide firepower: the government now makes effective use of its powers of coercion to ensure that its policies are carried out, while other bureaucratic agencies, for example pension schemes for civil servants or insurance companies, have increasingly taken over the payment of various kinds of compensation. In the settled villages, moreover, the appointment of village chiefs is now firmly under the control of state authorities. The position at the moment is that day-to-day matters of local interest are still being worked out mainly by local political processes, but the fundamental matrix of policy is now in the hands of the state.  - 218 ANNEX 4 Page 11 Urbanization and Migration 3.04 The expansion of the colonial and post-colonial state structure and the accompanying economic changes have had a dramatic effect on the pattern of urbanization in the country. First, the raisons d-etre of the coastal settle- ments have been transformed. In the past they were the entrepots of the dhow trade, sited about a day's sail apart, where a harbor and fresh water were available, and where access to the hinterland was relatively easy. While in some areas political influence spread from the port over the hinterland, for the most part the interior was politically autonomous, and the relationship with the coast communities was cast in commercial terms only. In the modern period, a few coastal settlements have added major administrative and economic functions (Kismayo, Berbera, and especially Mogadishu), but most others have gone into steep decline (e.g., Adale, Mait, Zeila) as the dhow trade declined and as economic and administrative functions were focussed along inland commercial routes. Second, this new pattern of urbanization concentrated activities in a few locations. Hargeysa has grown from a tiny settlement to being the country-s second largest city by reason of its favored situation as the colonial administrative center for the British protectorate in the north. Above all, Mogadishu has almost achieved a heavy urban dominance as political, economic, industrial, administrative, higher education and cultural activities have been centered in the capital. With 10 percent of the country-s total population, Mogadishu is now experiencing the difficulties of providing jobs, amenities and services that might better be handled with greater decentraliza- tion of urban activities. 3.05 The growth, first of the colonial administration, second of the commercialization of agriculture, and third of the independent Somali public sector has provided wholly new types of occupation. Wage employment has become something that herders and crop farmers all over the country perceive as a real possibility for at least some of thc4r children. So much so that some nomads are spending as much as So. Sh. 4,000 a year to board a child of primary school age near a village school. Some of the aspiration for wage employment will be realized, of course, abroad on the Arabian Peninsula. Most children may not finish school and may, if they find employment at all, find it at levels that require only muscle-power, not academic skills. Be that as it may, the current situation is that increasing numbers of rural families, even those not forced to migrate by drought or ill fortune, are sending one or more children to the urban areas to diversify their sources of family income. In remains to be seen whether their expectations that sustained remittances will be forthcoming and will repay the investments in schooling and the opportunity costs of lost family labor will be fulfilled or not. Nevertheless, urbanward migration is now taking place on a major scale in Somalia and must be understood as an expression of confidence in the relative ability or urban employment to generate supplementary family income.  - 219 ANNEX 4 Page 12 Intervention in Production Systems 3.06 From the time of the earliest appearance of colonial interest in Somali territory, there has been intervention in the conditions of agricultural production. At first this interest was commercial, and in the livestock sector the influence of the state has been mainly limited to conditioning the trade flow in livestock. In the crop sector, influence has been more direct. The state has recently intervened with increasing intensity for ideological, strategic, and political reasons. The overall consequence of this history of state involvement in agriculture has been the emergence of an embryonic system of social class, as will be elaborated below. 3.07 The Commercialization of Livestock. The earliest interest of the British in the northern Somali coast was as a source of meat for their expand- ing garrison at Aden. Trade in pastoral products (and those collected in the wild) from both coasts is ancient, being first recorded in detail 650 years ago. Trade of live animals, ghee, hides and skins was entirely in private hands, organized by brokers between pastoral producers themselves and merchant exporters, with some market dues going to the chiefs of the port communities. In the twentieth century, the trade has grown enormously; the oil boom periods of the Arabian peninsula first in the 1950's and then in the mid-1970's have funded huge new demands. In Somalia, production has remained with nomadic families; there has been no substantial change in the organization of produc- tion itself. The primary orientation has changed, though, from subsistence to market pastoralism. Commercialization has undergone this significant intensi- fication while remaining virtually entirely in private channels. Private traders have invested in water storage units to localize production, in a transport fleet to move animals to market and port and in hay production activities (building enclosures in good rangeland) to support the international movement of the live animals. Collective efforts have been made to gain increased government support and to organize commercial transactions in their best interests. Government has responded to the importance of the livestock sector for overall government revenues by improving port and road infrastruc- ture, by providing animal health supervision, by organizing a payment transfer system that enables the exporter to buy and ship imports with part of his proceeds abroad, and by otherwise not intervening in the commercial sector in the way it has for commodities traded by less influential, sometimes non-Somali, merchants. Acquisition of the River Valleys 3.08 Those Somalis who were historically fortunate enough to acquire the lands along the two rivers have now all but lost them to large-scale private and public interests. Such lands were traditionally claimed by clans or village groups, but relatively little was actually used on a continuous or intensive basis. As in many other countries, the colonial powers held that lands not actually being intensively used were fair game for government purposes. Thus, starting in 1908 with an Italian scheme for concessionary agriculture to help pay for the colonial administration and to supply tropical products to the metropole, large tracts of the best-favored riverine lands  - 220 - ANNEX 4 Page 13 were acquired for large-scale crop production. The first of these were for private banana plantations and the Jowhar sugar estate, but the right of eminent domain has been asserted by each successive government to requisition land for large-scale crop schemes. In some early cases there was acknowledge- ment of traditional tenure, since the land was purchased from clan elders (who may not have had any clear right in customary law to alienate it). Mlore recently compensation has only rarely been paid or sought, with the population apparently "content to profit from the labor opportunities and increased circulation of wealth which the projects offer" (Lewis 1967:22), or at least not inclined to protest very sharply. 3.09 As irrigation canals were built away from the rivers themselves, additional lands were taken over for plantation crops. Some of it was developed, then abandoned, and sometimes repossessed as the economics or technology of various crops fluctuated. There has been no attempt to account for how much irrigable land, and of what quality, remains in the hands of villagers. It may be assumed that the 209 banana plantations on the two rivers, together with the existing irrigation projects and state farms on both rivers, occupy virtually all of the high-quality lands. In all, the 1977 Hunting Report records that about 35,000 ha were under controlled irrigation (some 27,000 ha on the Shebelli and 7,500 on the Juba). While some of these lands were being occupied, at least temporarily, by smallholders, most were held by large-scale farmers (many of them the Somali successors of Italian concessionaries) and the Government. A further 47,000 ha at least were identified for large-scale irrigation development: indeed, as the Hunting Report notes, "the great bulk of investment in the rural sector is to be devoted to the construction of new irrigation schemes" (p. 33). If this development were technically possible (see Annex 2), it would complete the takeover of even bottomland (presently flood-irrigated) from traditional producers. The data on the present ratio of small holdings to large is sketchy at best. In any case the central point of social relevance is not this ratio, whatever it might be, but the overwhelming trend not just to controlled irrigation technology but to the permanent alienation of lands from their traditional owners. Early Outreach to Small Producers 3.10 There were a few attempts in the past to improve conditions for small producers, but these were pale even by comparison with other African colonies. In the former British colony they were very limited, consisting of the creation of small veterinary and medical programs, the construction of a handful of schools and a few roads and boreholes, and the beginnings of a system of land registration for the agricultural holdings of the northwest and the forest gum production areas to the northeast. In the south, the offer of wage employment on what were to be models of progressive concessions turned frequently into the forcible impressment of labor. The building of a chain of schools and medical facilities evidenced the early Italian motivation to "uplift" the population, but the agricultural experiments run at Genale and Villa d'Abruzzi were solely directed at plantation crops. After World War II, governments slowly extended health, education, and administrative services. For the first time there were experimental programs in farmer credit, water  - 221 - ANNEX 4 Page 14 catchment construction, grazing rotation, farm mechanization, animal traction in the south (it was innovated indigenously in the northwest), as well as the beginnings of agro-industry. But until independence there was no attempt at all to cover the countryside with any service or administrative demand: there was not even the collection of a head tax or an animal tax that might have brought government into contact with the mass of the population. B. The Growth of State Organizations 3.11 Since Independence, and especially since the 1969 revolution, the activities of the state in the countryside have accelerated considerably. To understand the reasons for the particular programs undertaken and their consequences, it is necessary to see them in the context of the evolving nationalist and socialist commitments of the country. The Socialist Commitment 3.12 The declaration in 1970 of the socialist nature of the Somali revolution, reflected the country-s growing consensus that the continuation of policies begun during the colonial period could not break the chains of its poverty. It also declared a will to effect change in the countryside after a decade of drift in development terms. These three strands of thought -- ending neo-colonial subservience, ending national poverty and changing the face of the countryside -- have dominated development policy since. Ideology and Education 3.13 The colonial experience had engendered a strong sense of national injury among Somalis. The national territory had been divided, allocated and reallocated among foreigners with no consultation of Somalis. By comparison with her colonial neighbors, Som.ia had been clearly neglected. At the beginning of the 1970's the new government determined to overcome this sense of disinheritance, this marginalization. It resolved a long debate by adopting the Roman script for the national language. It embarked on two major literacy campaigns, the first in 1973 in the cities and then in 1975 throughout the country. Even though the major drought had struck and the literacy workers were diverted to the relief effort, within a few months 854,000 Somalis, nearly half the adult population, had become sufficiently versed in their language to earn certification of their literacy. For the first time, a mass mobilization in the countryside had been undertaken, and it worked. In the circumstances, the boost to national self-esteem may have been as important as the technical result: since there were few materials available to sustain the new reading skills, many people (one survey in 1977 said 50 percent) soon relapsed into non-literacy. But the demonstration of education as a tool for participating in the nation was effective. Primary school enrollments, only some 70,000 in 1974, rose to nearly 200,000 in 1976, and more and more rural parents were beginning to sacrifice the value of their children-s labor, and (for nomads) to meet the cost of boarding, in order to keep the children at school.  - 222 - ANNEX 4 Page 15 3.14 Scientific Socialism in Development. Government s determination to eliminate the vestiges of colonialism and to force the pace of economic growth has given rise to a number of interrelated policies. Among these the most notable have been: - the nationalization of the banks and the creation of state export and import trading monopolies for many items to control elements of commerce felt to have been bleeding the country; - the holding down of wages in the public sector in order to try to prevent the use of government service to amass wealth. The wage scale has not been amended since 1969 except for the lowest-salaried workers (although allowances have been changed), and inflation has seriously reduced the purchasing power of salaries. Appendix I of this Annex outlines public sector compensation levels. Reaction to this squeeze has taken two forms: out-migration to jobs on the Arabian peninsula (at salary levels wlhich Somalia could probably not match anyway), and, at least as importantly, a massive diversion of attention from official duties to pursue business ventures outside official duties. - rapid expansion of the state sector in crop agriculture, and especially in irrigation. Annex 6 describes the Crash Programmes and State Farms and the present and proposed state or joint-venture projects in irrigation. All of these stem from convictions, stated in explicit policy or not 1/, that crop enterprises offer the best hope of rapidly increas- ing national output and of substituting domestic products for imports, that state organization of these enterprises should be the most direct way of erfecting these increases since technical and labor inputs can be fully controlled, and that irrigation more fully guarantees yields of high- value crops. Most of these convictions are shared by other countries of far different political persuasion. They grew firm before the turn to scientific socialism and before the drought, events which merely deepened Somalia-s commitment to state production and irrigation to guarantee the country's food security. Yet, faith in this direction has not been rewarded, as production has stagnated. 1/ For example, the Ministry of Commerce and Industry, in a pamphlet describ- ing the Crash Programmes, states as their objectives: (a) creating job opportunities, (b) promoting transformation of the land tenure system, (c) increasing food production for the country, (d) orienting youth to modern farming, (e) motivating cooperation and spirit among youth. These goals speak to legitimate national needs more fully than to the needs of existing farmers.  - 223 - ANNEX 4 Page 16 the creation and spread of cooperatives in agriculture, trade, and industry. The extensive organization into cooperatives was clearly meant to reach small producers and to help achieve the socialist transformation within the country. The Cooperative Law (No. 40) of 1973 envisaged western-style service cooperatives only as a prelude to full collectivization of production. Service, or "multi-purpose", cooperatives and group farms were "lower type" cooperatives, while the transition to "higher type" (collective) cooperatives "shall be the final stage" for the "future socialist pattern of the economy." 1/ In the absence of effective extension or a strong tradition of cooperative labor, the inducement to form cooperatives and group far-xs in the crop sector (and grazing associations or traders' cooperatives elsewhere) was to be the provision of free or highly subsidized government services including, notably, tractor plowing. Government intervention has had technical and morale problems, however. As a result, attempts to organize and manage collective responsibility have failed; most noticeably on the group farms which were to have been small-scale collectives of volunteers working surplus land with government support. The Cooperatives Department has reverted to a policy of total individual responsibility for plots within the farm. No full-scale collectives of the third stage have yet been organized and doubts have been expressed as to whether they ever will be. C. Redistributing Population 3.15 A final aspect of policies to break the poverty of the countryside has been the strategy of resettling the population. It is important to note that this social strategy had been proposed before the drought, and is only part of socialist strategy as a secondary matter. The policy arose for two reasons. First, it is difficult and costly to bring administrative and social services to a dispersed and mobile population. This fact had early suggested to the Somali government, as it has to many others, the cost efficiency of getting nomads into large, settled communities. The commit- ment to large-scale irrigation development also implied the growth of a labor force for irrigation well beyond that available from neighboring villages. No overall calculations of labor force recruitment to the total range of projects was ever done (with the result that there has been serious competition for labor and some shortages even of unskilled labor on the 1/ Quotations from Law No. 40 of October 4, 1973 as printed in Agricultural Cooperatives in Somalia: Handbook, Ministry of Agriculture, Mogadishu, 1977.  - 924 - ANNEX 4 Page 17 Juba and in the Northwest). But the ILO/JASPA study estimated (p. 43) that 27,800 laborers would be necessary for the agricultural projects in the 1974-1978 Plan period. 3.16 It was in this setting that the drought struck in 1974-75. Suddenly the attention of the nation was riveted not only on the need to care immediately for those thousands of pastoral people bereft of their livestock, but also on the prospect that all future population increase in the pastoral sector should be absorbed outside pastoralism to avoid any further charge on the rangelands. The creation of the Settlement Development Agency in 1975 (Annex 6) was the concrete response to the emergency situation, but the policy of attempting to redistribute population has taken diverse forms. - The Crash Programmes were to provide employment and training and started by recruiting from the urban unemployed, whose numbers had already become disturbing to the new government by 1970. The five Revolutionary Youth Centers, created in 1974, are further institutions for training and education, mainly at younger ages, of orphaned or unattached children. - The Sedentarization of nomads is generally understood to be national policy, though it has never been explicitly elaborated. The 1974-1978 Five-Year Development Programme (FYDP) called nomadic pastoralism a "tradition-ridden mode of production" (p. 21) and called for its "transformation into a more secure and productive one" (p. 38). It may be fortunate that the policy has not been pursued for its own sake, because worldwide experience has shown that premature or forced sedentarization of pastoral peoples simply does not work short of massive coercion. Creating the conditions under which pastoralists do not have to search for water, forage, minerals, craft materials, market opportunities, and other necessities provides the only basis for encouraging their settlement. - The resettlement schemes were the response to the drought emergency. In all, 105,000 people were taken from nearly 20 refugee camps and resettled in the three large agricultural settlements of Dujuma, Sablale, and Kurtun Warey (90,000 total) under Settlement Development Agency responsibility, and the four fishing communities of Eil, Adale, Brava, and El liamed (15,000 total) under Coastal Development Authority responsibility. As Table 4 indicates, the population (especially of adult males) in the agricultural settlements has declined steadily and is now at 56 percent of the original total. Moreover, their performance as agricultural production organiza- tions has been extremely poor to date, three years after their founding. The population of the fishing settle- ments has held up, but the number of boats actually  - 225 - ANNEX 4 Page 18 working has declined to about 15% of those originally available. Even if these communities eventually become economically self-supporting, the huge start-up costs that will have been involved obviously preclude the replication of this technique of resettlement to any major degree. 3.17 Nonetheless, the ILO/JASPA employment study predicted that from 1977 to 1981, Somalia's population would grow by 414,000, of whom the 186,000 in the pastoral sector should probably be moved into other pursuits in order to protect the range (ILO/JASPA, pp. 25-26). If government accepts this position, and to the extent that resettlement is not spontaneous, it is clear that new strategies will have to be devised to accomplish whatever resettlement task government takes upon itself over the long term. D. Summary and Conclusions The Growth of Class Differentiation in Rural Somalia 3.18 Traditionally, Somalia was a land where class differences were unmarked. The classic work on northern nomads is I.M. Lewis, "A Pastoral Democracy" (1961), so titled because of the strong political and economic egalitarianism of virtually all household heads. In the agricultural areas there was somewhat more hierarchy of political power than among the nomads, but still general equivalence of access to resources within each clan-s territory. 1/ Lewis also spoke of townsmen as the extension of lineage groups in the countryside rather than as an urban class apart (1961:94), and said that Somalia's lack of resources had produced, even during colonialism, "little diversified class structure" (p. 267). While there may be some debate about the past as Lewis has portrayed it, there can be little doubt that sharp differences of income have emergeu. Tables 5A and 5B show the extent of these differences for the public sector as a whole and for a single agro-industrial enterprise, the sugar estate at Jowhar. The wages paid to civil servants and unskilled laborers at all the State Farms and other agricultural enterprises and at the project construction sites are roughly equivalent to those in this single example. Unfortunately the size distributions of different shares of rural wage income cannot be calculated, since the total numbers of workers at different wage levels is not in any one record. 3.19 Differentials in the rural wage sector are only one small element of a nascent class structure. Cash income from the sale of labor or of surplus produce accounts for only a part of the total income of rural farming families. The larger share is still from subsistence production. Thus a part-time plantation worker who in addition raises most of his food and sells some of his crop surpluses may be better off than a full- time skilled worker whose sole income is his salary. The distribution 1/ I. M. Lewis, "From Nomadism to Cultivation," in May Douglas and Phyllis M. Kaberry, eds., Man in Africa, London: Tavistock, 1969, pp. 59-77.  - 226 - ANNEX 4 Page 19 of total income has never been researched, but it is still possible to speak of an emerging class system. Differentials of power, of access to strategic resources, and of opportunity are now cumulating in Somalia in ways that make the lives of people in different classes substantially distinct. While some individuals may overlap two or more categories, it is now possible to speak of at least the four following classes in rural Somalia: (a) the agents of the state, (b) rural wage workers, (c) the "mobilized" rural producers, and (d) the residual masses. 3.20 The agents of the state are those middle and higher level salaried workers whose jobs put them into rural areas during at least part of their career. They are the managerial, technical, and skilled workers on state enterprises, including foreign assistance projects, state farms, agro-industries, and district and regional administration. Their jobs are relatively secure and do not depend on their productivity in economic terms or even their day-to-day health in the way that a herdsman's or a farmer's income does. Though their numbers are small, they control the fine detail of life in the rural areas. They decide which laborers may be hired, they have access to transportation, determine what local suppliers will be favored or whose service needs will be met first, administer regulations and justice and supervise a variety of government operations from purchasing grain to distribu- ting food-for-work. They have ready access to communications channels and are well versed in their use, so their own needs are easily articulated. Like their urban counterparts, they have time and resources to engage in business concerns outside work. Because they know the bureaucracy, they can more easily provide their children with the help and education they will need to retain high status. Given that control of crucial means of production and redistribution of the social product is in the hands of the state, these people, its local representatives, are all-important gate-keepers to government resources. Simply by being there, they become major local political figures. To the extent that the government role in the countryside increases, they become ever more significant. A parallel group in the private sector in the rural areas, made up largely of merchants but also including some plantation owners, is in the same class position. It is defined by its public sector segment because its existence as a class, as we have seen in discussions of livestock commercialization and the nationalization of certain enterprises, is so heavily dependent upon public policy. 3.21 Rural laborers. This class is like the first in owing its existence to the activities of the state, and in depending on cash for many or most subsistence needs. It consists of construction and plantation workers and the lower-level staff of agro-industries. Much of it is made up of part-time laborers who, in the main, retain claims on animals or land in the subsistence/ produce marketing sector. Little is known of their total earnings or their methods of managing jobs with their own production enterprises. Obviously it is not a transitional group, for the size and permanence of this class has grown with the enormous expansion of state enterprises in the last few years, and it will grow further if present policies continue. With some education and an increasing dependence on wages, its interests in such secular services as health and education are similar to those of the class  - 227 - MINEX 4 Page 20 in more active control of the wage sector. But it will realize its aspirations for steady employment or for better opportunities for its children only if the state itself fares well, because unlike the first class, whose jobs are quite secure, the number and duration of jobs for this class fluctuates widely. Petty traders also probably belong to this class, since their livelihoods also depend on the circulation of cash in the rural economy. For all of the members of this class, political power is not a question: often their jobs are far away from their true homes, and the very servility of their status on the job lends them no confidence in their positions. On the other hand they are certainly not a rootless or landless class, either. 3.22 The "mobilized" rural producers. This class has arisen through its members' own initiative and contacts with the state and its structures. They are the clientele of development projects and the administrative apparatus: the members of cooperatives, the attenders of adult educ-.rion classes, the local political leaders of small communities, the members of livestock grazing associations, the adopters of new technology introduced by agricultural and fisheries agents, the nomads trained as fishermen or farmers in the resettle- ment programs. Through some dynamic of their own resourcefulness, their good connections with the first class, or their fortunate location, they are the ones receiving a high share of state services. They do not rely so much on the state as the first two classes: rather, they benefit from it. As its beneficiaries they can also be called upon to support it, and they form the bulk of the membership of the mass social organizations that regularly rally in regional headquarters and provide the hard core of the supporters of the party. Because they earn their livelihoods independently of the state, they are more isolated from changes of broad policy than the rural laboring class. They, too, probably send high proportions of their children to school as a wager on the urban economy, but their children are more firmly rooted in ongoing community life than the more frequently transplanted members of the first two classes. As irrigated land or improved range is made available to small producers, they emerge with the lion's share of the best resources. Hlistorically, for example, nomadic sedentarization has usually led to increased social differentiation -- those mobilized for quick exploitation of new opportun- ities will come out on top. This class is not necessarily a "progressive" class, however. Traditional society encouraged quick appropriation of available resources, the difference being that those resources (like fresh grass) were only temporary themselves. A class mobilizing its own energies can also be the most religiously orthodox or most loath to share any newfound opportunities. 3.23 The residual masses. It remains a fact of national life in Somalia that a large majority of rural crop and livestock producers are beyond the reach of the agents of government. The prices they get for their products, the availability of goods they buy and the circumstances of their access to facilities depend heavily on government policy, but they are only passively constrained by those policies. The primary goal of their production is for their own subsistence, though their choice of subsistence techniques is affected by market conditions. Many cannot afford to give up their children's labor to have them attend the tuition-free schools, or if they are nomads cannot afford the boarding costs. By definition they are the last to be  - 228 - ANNEX 4 Page 21 exposed to development projects, and before the benefits reach them, the projects may run out of resources. In the nomadic areas they are the ones, for example, now losing range areas to fodder production, being reoriented by the siting of wells or ranches without much consultation, and being disrupted by the continuing difficulty in obtaining unrestricted access to the Haud summer pastures. Whether nomads or farmers, they are the most remote from government, but probably the most directly affected by natural forces. So they form the class from whom drought refugees have flowed. While they own not inconsiderable standing wealth in the form of land or animals, they are the "poorest of the poor" in terms of their chances of improving their lot. Whether they will ever move into the ranks of the "mobilized" producers depends on the cost-effectiveness of government strategies in extending services to cover their needs before other classes consolidate their positions. Thus government has clear choices to make about the future of this so far neglected class. Summary 3.24 Any analysis of class differences necessarily relies on qualitative factors beyond sheer quantities like the distribution of income. It is thus inevitably more subject than discussions of numerically-based issues. In raising questions of the emerging patterns of differential influence, oppor- tunity, and income we cannot claim to be definitive. We have merely attempted here to stimulate further consideration of, and research on, such matters. The value of class analysis is to focus the attention of all the interests involved on the ways in which policy implementation (and other social forces that may be beyond government's influence) are affecting the society in broad terms. The classes we have identified are not necessarilly fixed, nor are they in overt conflict. Since we have not discussed urban class interests or the class structure of the whole nation, much analysis remains to be done. In the meantime there can be no doubt that class interests have emerged and to threaten to become more divergent in the future.  - 229 - ANNEX 4 Page 22 IV. PROBLEMS AND PROSPECTS IN SOCIAL DEVELOPMENT A. General 4.01 Somali development policy is clearly and laudably devoted to the development of the Somali people, and not simply to abstractions like the growth of the economy or of production at large. This commitment to the masses of the citizenry has been expressed in both the statements and the actions of the Government and of the Somali Revolutionary Socialist Party. As general policies are translated into strategies for dealing with particular problems, however, there is the danger of diluting the original goals. Some policy values may be in competition with others for priority. Strategies for achieving certain goals may work against the achievement of other goals. It is in the spirit of constructively monitoring the social effects of the agricultural development strategies that are now being implemented in the country that the following issues are raised. 1/ B. Self-Reliance and Self-Sufficiency 4.02 The concepts of self-reliance and self-sufficiency have different implications at different levels of the society. Traditionally, for example, individual rural households were self-sufficient for most of their routine needs and self-reliant for the remainder by exchanging their labor, products and loyalties with others