5
Northern Nigeria: Land Transformation Under Agricultural Intensification

Michael Mortimore

INTRODUCTION

Objectives

The objective of this case study is to review, in descriptive (rather than explanatory) terms, some evidence on the nature and rate of land transformation processes in two northern Nigerian farming systems. The term "land transformation" (Wolman and Fournier, 1987) is preferred to the more familiar "land use change" because it embraces a broader range of change processes relevant to the population-land equation. A suggested classification of these processes follows:1

  1. Land use change (the transfer of land from one class of management to another).

  2. Land investment (or divestment) in:

    • soil/water conservation structures,

    • irrigation/drainage structures,

    • management structures (especially enclosures),

1  

Much more needs to be done before the data and methods can be considered to be satisfactory. In particular, land investment cannot yet be analyzed in quantitative terms, so the discussion of it here is very brief.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop 5 Northern Nigeria: Land Transformation Under Agricultural Intensification Michael Mortimore INTRODUCTION Objectives The objective of this case study is to review, in descriptive (rather than explanatory) terms, some evidence on the nature and rate of land transformation processes in two northern Nigerian farming systems. The term "land transformation" (Wolman and Fournier, 1987) is preferred to the more familiar "land use change" because it embraces a broader range of change processes relevant to the population-land equation. A suggested classification of these processes follows:1 Land use change (the transfer of land from one class of management to another). Land investment (or divestment) in: soil/water conservation structures, irrigation/drainage structures, management structures (especially enclosures), 1   Much more needs to be done before the data and methods can be considered to be satisfactory. In particular, land investment cannot yet be analyzed in quantitative terms, so the discussion of it here is very brief.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop productivity management (especially fertilization), and tree planting and protection. Land productivity change in: soil chemical properties, soil physical properties (including moisture), soil biological properties, farm trees—physiognomy (including density and volume), and farm trees—floristic composition. The term "intensification" is taken to mean the following three interrelated processes: increased frequency of cultivation; labor intensification per hectare; and capital intensification per hectare. The evidence reported here supports the hypothesis that under conditions of land scarcity, population growth on smallholdings drives agricultural intensification, land investments, and productivity enhancement, which are together consistent with an objective of sustainable resource management. The Study Area In the semiarid zone of northern Nigeria (Figure 1), rainfall diminishes northward with the length of the growing season. A vast plain, with local relief rarely exceeding 100 m, slopes gently toward the north and east from about 500 m above sea level at Kano. The soils are predominantly derived from former dune sands and are commonly deep and freely draining. They are low in organic matter, nitrogen, and phosphorus (Jones and Wild, 1975). Restricted areas of interdunal soils have finer textural characteristics, wetter moisture regimes, and sometimes superior organic content. Natural vegetation (now increasingly modified) consists of open thorny savanna woodland and grassland, much influenced by fire, with occasional occurrences of natural grassland on frequently flooded sites or on recently vegetated dunes. Degradation of the soils and vegetation is believed (in official circles) to be widespread, but there are few data (Mortimore, 1989b). The population densities are very uneven, notwithstanding an appearance of homogeneity in the major environmental variables. There has been no accepted population census since 1963.2 Population density was last mapped for 1952–1953 (Prothero, 1958). A more recent proxy for population 2   The results of the 1991 census were not available at the time of this writing.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop NORTHERN NIGERIA MAP NO.1 Figure 1 Northern Nigeria: average annual rainfall, 1968–1987, showing the locations of study areas. HDC = High-density case; LDC = low-density case. SOURCE: Mortimore (1989b).

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop density is the percentage of land under cultivation, which was mapped in three classes from a side-looking airborne radar survey in 1977 (Federal Department of Forestry, 1977). This pattern, reduced and generalized, is shown in Figure 2. Such a classification disguises the continuum that exists, from over 80 percent cultivated to zero in uninhabited forest reserves. The lack of census data, or reliable population estimates, for recent years is a handicap in attempting to interpret the process of population change through time. This chapter presents two case studies that occupy contrasting positions on this continuum. The first is a high-density case (the inner Kano Close-Settled Zone), where rural densities are at least 350 persons/km2, and the second is a low-density case (the Manga Grasslands), where they are of the order of 100 persons/km2. The high-density area had an average annual rainfall (1968–1987) of 680 mm and the low-density area averaged 400 mm. More detailed interpretations of ecological and socioeconomic change in these areas are available in Mortimore (1989a, 1993) and Mortimore et al. (1990). Methods For estimating land use change over periods of a decade or more, sequential black-and-white air photography has been found to be the only practicable data source, notwithstanding differences in scale, quality, and in the dates of air surveys (though they were always conducted during dry seasons). Very little analysis of land use change has been attempted using the vast air photo resources available in northern Nigeria (but see Field and Collins, 1986). Official land use statistics are based on side-looking airborne radar surveys of 1977 (Federal Department of Forestry, 1977), and for technical reasons, neither this imagery nor earth satellite imagery is fully compatible with air photography. It cannot, therefore, be used as part of a series to evaluate land use change. The sustainability of a farming system can be assessed from either an economic or an ecological standpoint, but the conservation of the soil resources is a precondition for the longer term continuity of the system. Soil surveys and inventories, however, do not normally address the question of medium-term changes in soil chemical and physical properties, and the need for such monitoring has been neglected (Young, 1993). A recent study has proposed a set of indicators of soil quality and ways of measuring changes caused by management (Larson and Pierce, 1991). The Central Nigeria Project (Land Resources Development Centre, 1979) carried out extensive soil investigations in the Kano Close-Settled Zone in 1977. Soil pits were dug in catenary sequences at sites representative of major land systems. Profile descriptions, with detailed site and soil de-

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop Figure 2 Northern Nigeria: intensity of cultivation in 1977, north of latitude 12°N. SOURCE: Mortimore (1989b).

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop scriptive data, were filed, and samples were taken from representative horizons. Selected samples were analyzed.3 For the present study (Mortimore et al., 1990), a selection of catenas, within a radius of 40 km from Kano, was made to represent land systems in the high-density area. New samples of surface soils (0–20 cm) were obtained, by auguring, from sites on the same land facets, and as close as possible to the original soil pits. The new samples were analyzed alongside material from the original samples, using standard analytical procedures. Both original and new samples were obtained from cultivated upland sites. For evaluating change over 13 years (1977–1990), the results were first averaged by catena, then aggregated for each land system, and finally for the whole area of study.4 No soil data were available for the low-density area. Studies of the farm tree population in the high-density area used aerial photography of medium scale and three sequential dates, interpreted with stereo equipment and magnified up to 15 times. The air photo work was supplemented with ground surveys in randomly located quadrats, which included species identification, height estimations, and girth measurements of all woody plants above 10 cm in girth at breast height. POPULATION CHANGES5 The High-Density Case The censuses of 1952 and 1962 suggested a net rate of increase of 2.6–2.9 percent a year (Green, 1970; Mortimore, 1974). Rural rates must have been somewhat lower, because urban rates were 6.7–8.9 percent; 10.5 percent of the Province's population being then urbanized. A demographic survey in Malumfashi (southern Katsina State)—an area of similar ecology—gave a crude rate of natural increase of 2.9 percent for a population of 42,493 in 1973 (Bradley et al., 1982). The general fertility rate (GFR) was 221 per 1,000. In the Nigeria Fertility Survey a decade later (1981/1982), the GFR was reported to be 183 in Kano State, somewhat lower than the Malumfashi estimate.6 This implies a lower rate of natural increase. Continuing 3   These samples were stored, cataloged, and located on large-scale maps in the Kano Environmental Data Bank at the Department of Geography, Bayero University (McTainsh and Stokes, 1980). 4   The results of this experiment in longitudinal soil monitoring are regarded as provisional, in view of the small number of samples analyzed and the limitations of the method. 5   In the absence of census data for the last three decades, population change has to be inferred from fragmentary information. 6   The Nigeria Fertility Survey defined the GFR as the number of children born per 1,000 women aged 15 to 49 in the 5-year period before the survey. The figure may be low owing to inaccurate recall.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop high rates of infant mortality and a deterioration in public health care delivery during the past decade are widely believed to have affected rural areas in the semiarid zone of Nigeria. It is quite possible, therefore, that the rate of natural increase is well below the national average (estimated to be 3.3 percent in the World Bank's projections based on the early 1980s [World Bank, 1984]).7 Migration is difficult to measure, even at the microlevel. There is also a great deal of seasonal and short-term circulation, which fluctuates from year to year. Translated into density, the available information suggests an average density of 235 persons/km2 in the Close-Settled Zone (Ungogo District) in 1962,8 and in three villages surveyed in 1964, 353 persons/km2 (Mortimore and Wilson, 1965). Given a cultivated ratio of 84 percent, the amount of arable land per head was 0.24 ha in these villages and 0.36 ha in the District. Any subsequent estimate of density is speculative. If the rate of natural increase after 1962 were assumed to be 2.9 percent (as in Malumfashi in 1973), and migration were zero, the average density in Ungogo District in 1988 would have been 495 persons/km2 —an increase that is improbable on the basis of field observation. If, on the other hand, out-migration is assumed to have averaged 1.5 percent, a resulting net increase of 1.4 percent would have generated a density of 347 persons/km2 by 1990. This would allow 0.24 ha of arable land per head, only 68 percent of what was available in 1962. The Low-Density Case Very little information is available for the low-density study area. It can be safely assumed that the rate of natural increase does not exceed 2.9 percent (the Malumfashi estimate). Out-migration is probably considerable, even when seasonal and short-term circulation is excluded. Net population growth may therefore be as low as I percent. Some support for such an assumption is provided by the example of Dagaceri village, where the number of households did not change significantly between 1975 and 1986 (Mortimore, 1989a:98). Dagaceri lies outside the Manga Grasslands, but is ethnically, culturally, and ecologically comparable. Market towns, on the 7   In the inner Kano Close-Settled Zone, growth rates apparently fell from 2.2 percent in 1931–1952 to 1.2 percent in 1952–1962, probably as the result of out-migration. Hill (1977:91–92) estimated an average net increase of only I percent from 1932 to 1971 in Kumbotso District. 8   The cancelled census of 1962 is considered, on the basis of limited field checks, to be more reliable in Kano than the published census of 1963.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop other hand, may grow rapidly. There is also some local redistribution of the population, including occasional resettlement, on new sites, of entire villages. According to the 1952 census, the districts of Bornu Province in which the Manga Grasslands then lay had population densities in the range of 20–50 persons/km2. The average density now is thought not to exceed 100 persons/km2. The Demographic Basis of Intensification Rising population densities—at opposite ends of the range—provide the potential for labor intensification in the farming systems. A longitudinal review of the evidence available follows. A comparison between the two areas, whose present densities differ by a factor of 4, provides an opportunity to test the hypothesis that population density is a predictor of agricultural intensification. Such an investigation is, however, complicated by ecological and historical differences—no two areas can be strictly comparable. The high-and low-density areas represent different points on the rainfall gradient, have different soils and topography, and are differently located with respect to precolonial political and commercial centers. LAND USE CHANGE The High-Density Case In 1964, 79.6 percent of the area was under upland (rainfed) cultivation, and an additional 4.2 percent was irrigated lowland (Mortimore and Wilson, 1965). Cultivated upland was under annual cropping with sorghum, millet (Pennisetum typhoides), cowpeas (grown in mixtures), and groundnuts. (grown in mixtures or alone in rotations). Farmyard manure was applied at an average rate of 4.1 tons/ha, but ranging from 0 to 6.5 on individual holdings. The average size of fields was 0.5–1.0 ha, and of holdings, 1.76 ha (range: <1 to >10 ha). Land preparation (for groundnuts only), planting, weeding (three times), and harvesting were accomplished with labor-intensive hand hoes and machetes. Fields were permanently divided by living boundaries of thatching grass, useful shrubs, and trees. Farm trees were distributed irregularly in the cultivated fields at densities of 12–15/ha, and produced marketable output, construction and craft materials, and food. Fallows only occurred when a household was unable to cultivate its holding fully, for example through illness. Such holdings provided both the basis of household subsistence and a marketable surplus (groundnuts). There were active markets in land and labor, with rising prices for both.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop Shrub-or grassland-occupied low-lying sites were considered unsuitable for cultivation on account of flooding and were heavily grazed and cut over. Some erosion occurred on the banks of streams. This land use system had evolved over a period of many centuries and was quite stable. Aerial photography done in 1981 indicates that, after 18 years, the only significant change to have occurred was the transfer of some cultivated land to residential and commercial development. Average rainfall, and possibly flood risk, had diminished by over 25 percent since the 1960s. An analysis of land use change in an area of 332 km2 between 1965–1966 and 1981 confirms this conclusion (Table 1). A reduction in the area of flood plain, as interpreted from the photographs, after 1971 was due to the drying out of the soils during several successive years of drought. The expansion of cultivation, therefore, reached its conclusion several decades ago, with fallows virtually eliminated on the upland. Fieldwork in 1990 suggests that small areas of shrubland, notwithstanding their poor quality, had been added to the cultivated upland. Multicropping is still restricted to lowland sites (though pumps have increased the lift available) and by the water supply. This stable land use pattern, however, disguises important changes in the farming system (Mortimore, in press). (a) Until inorganic fertilizers became available, there was a long-term decline in yields. Data limitations do not permit this decline to be measured. (b) There is better evidence of a long-term decline in the average size of holdings through subdivision on inheritance (for example, see Mortimore, 1967). (c) There is incontrovertible evidence of manifold increases in the prices paid for land. (d) House TABLE 1 Percentage Land Use Change, High-Density Case, 1965–1981 Land Use 1965–1966 1971 1981 Cultivated upland 87.12 86.73 85.46 Cultivated and uncultivated floodplain 8.64 10.08 5.95 Settlement 1.21 1.71 4.85 Eroded area 2.70 0.88 2.22 Others 0.33 0.60 1.52 Total 100.00 100.00 100.00 Area mapped (km2) 332 365 435 Scale of photography 1:40,000 1:40,000 1:25,000   SOURCE: Mortimore et al. (1990).

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop hold grain insufficiency is socially extensive and may plausibly have increased. (e) Inputs of farmyard manure increase with population density (Hill, 1986) and inorganic fertilizer inputs increased from the 1960s to the 1990s. (f) Labor hiring has increased in cost and probably also in occurrence. (g) An increasing percentage of farmers use ox-ploughs, though the percentage varies between districts (from 28 to 69 percent according to official figures). The proportion was very small in the 1960s. Although a labor-saving technology, the ox-plough is associated with intensification (Boserup, 1965; Pingali et al., 1987). (h) Groundnuts, after being destroyed by rosette disease in 1975, ceased to be a major crop, and a high-yielding cowpea introduced by the World Bank's integrated rural development authority was taken up in the 1980s. (i) Livestock ownership (most commonly small ruminants, but including some cattle) continues to be very widespread and highly valued. There is no evidence of livestock ownership declining, though numbers fluctuate in response to the primary perception of animals as realisable assets, in particular when drought occurs (Mortimore et al., 1990). The Low-Density Case The upland subsystem consists of homogeneous, freely draining sands in a hummocky terrain of stabilized dunes, easily remobilized when the vegetation is removed. This vegetation is grassland, dominated by the annual, Cenchrus biflorus, which provides good year-round grazing. Settlements are sited on the upland, often in association with overgrazed and degraded patches. The lowland subsystem contains all the natural woodland (dominated by the dum palm, Hyphaene thebaica) and, where depressions intersect the water table, seasonal or permanent saline lakes. They support all the irrigated and most of the rainfed cultivation. Rainfed millet (Pennisetum typhoides), intercropped with cowpeas, provides the basis of household subsistence. Apart from irrigated plots, cultivated land is rotated in cycles of several years' annual cultivation, followed by rest periods of 10 years or more. There is no systematic use of farmyard manure, though livestock may be coralled on the fields. Field and holding sizes are at least double those of the high-density area, and planting densities, one-half or less. Planting is done in unprepared land, two weedings are accomplished with a labor-saving hoe (the ashasha), field boundaries are usually unmarked, and farm trees are extremely sparse. Fallows are occupied by regenerating dum palm bush and grasses. The data in Table 2 show that between 1950 and 1969 the land use system was relatively stable,9 with the exception of the loss of some wood 9   When the study was done, no more recent photographs were available. Updating to 1991 will shortly be carried out with new photographs.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop TABLE 2 Percentage Land Use Change, Low-Density Case, 1950–1969 Terrain Type Land Cover Class 1950 1969 Upland   67.8 71.8   Grassland 66.3 70.3   Woodland 0 0.1   Cultivation 0.3 0.5   Settlement 0.3 0.1   Mobile sand 0.9 0.8 Lowland    32.2 28.2   Grassland (sparsely wooded) 2.9 3.7   Open woodland 13.1 10.7   Dense woodland 4.3 3.3   Cultivation (lightly wooded) 9.9 7.9   Lake beds 2.0 2.6 Total    100.0 100.0 Area mapped (km2)   1,500 1,500 Scale of photography   1:30,000 1:40,000   SOURCE: Mortimore (1989a). land. This was converted to grassland. The cultivated area appears not to have increased.10 The small extent of change in the areas of cultivation and settlement conflict with expectations based on a high assumed rate of population growth during the period. Field studies indicate that the most visible change to occur since 1969 was an increase in the extent of mobile sand, which by 1986 had come to occupy up to 20 percent of one or two badly affected localities. This increase is attributed to the effects of drought and reduced rainfall more than to management, since the evidence available on livestock numbers does not indicate a substantial increase, and cultivation is not generally associated with dune remobilization (Mortimore, 1989a). Land Use Change and Population Growth Land use change in the high-density area is no longer significant (except for the expansion of settlements and the effects of hydrological change). 10   The apparent decrease should be treated with caution. The figures are subject to error and too much should not be read into small changes in small categories.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop TABLE 5 Illustrative Rates of Nutrient Application   Nutrients/ha (kg) Treatment N P K Mg Ca Organic manure           2.5 ton/ha 280 280 168 393 725 393 6 ton/ha 672 402 942 1,742 942 12 ton/ha 1,344 804 1,984 3,480 1,884 Crop residues           Field residues, 1.3 ton/ha 3 162 55 NA NA Roots, 0.7 ton/ha 1 132 67 NA NA Total, 2 ton/ha 4 296 122 NA NA NOTE: N = nitrogen; P = phosphorus; K = potassium; Mg = magnesium; Ca calcium; and NA indicates data not available. SOURCE: Mortimore et at. (1990). cium, and 0.7 percent magnesium (Essiet, personal communication). Smaller quantities of some nutrients are added to the soil in the form of residues and roots that are swept and dug up at the end of the dry season and burnt. The average weight of field residues in a 5 × 5 m quadrat was found to be 3.3 kg, and of roots of sorghum and millet, 1.8 kg. These values are equivalent to 1.3 and 0.7 tons/ha respectively, containing up to 0.5 percent nitrogen, and significant quantities of phosphorus and potassium (up to 3,800 ppm and 1900 ppm, respectively). Table 5 shows illustrative rates of nutrient application (based on fragmentary data). The inputs illustrated here take no account of animal droppings in the fields, leaf litter, or burnt prunings and other farm detritus; neither do they take account of variability in the nutrient content of different types of organic manure. Soil Management and Population Density Evidence has been presented that the topsoils on cultivated upland in the high-density area are stable with respect to most of the standard diagnostic physical and chemical soil properties. Although the values tend to accord with low levels of fertility, a comparison with uncultivated soils suggests that the soils are by no means degraded. The maintenance of fertility is the prime objective of smallholder soil management. If, as this conclusion implies, soil fertility decline can be stabilized under a high and increasing population density, then agricultural intensification is compatible with ecological sustainability in the farming system.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop By contrast, in the low-density area soil fertility levels (perhaps comparable to those of the high-density area) are maintained by long fallow cycles, and population growth has not yet forced a transition to more intensive methods of fertility maintenance, even though about 70 percent of the area is usually excluded from farming use. FARM TREES Under a regime of annual cultivation, natural woodland is gradually transformed, by the clearance, selection, protection, and planting of trees, into farmed parkland. The trees provide, among their many functions, browse for small ruminants in an integrated system of crop, livestock, and tree husbandry. Under such conditions, the density, volume, and regenerative status of the tree stock is a measure of sustainability in the system. Under less intensive systems, the tree stock responds to population growth in different ways, because natural vegetation is more abundant. The High-Density Area Studies of farm tree populations in two areas (west and east of Kano), using sequential aerial photography and ground surveys, generated the data summarized in Table 6. Notwithstanding the differences between the values obtained for the two areas, the table shows that the farmed parkland has been sustained over three decades including two major drought cycles (1972–1974 and 1983–1994). In the western area, the density of trees actually increased between 1972 and 1981. In both areas, the ground quadrats surveyed for the wood volume estimates (in 1985 and 1990) exceeded, on average, the densities obtained from aerial photography (1981). This may be due to a further increase in the numbers of trees or the inclusion of saplings not visible on the photographs. The girth classes of trees (Table 7), taking all species together, indicate that a large proportion of the tree population belongs to the smaller classes, showing that regeneration is taking place (though the distributions are strikingly different in the two areas). However, some species are doing better than others. The Low-Density Area Trees are very sparse on rainfed farms. Woodfuel, construction timber, and other products are obtained from natural woodlands in the uncultivated lowlands. Trees are sometimes regarded as a nuisance on farms as they may harbor bird pests. They have not, therefore, been integrated into the

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop TABLE 6 Tree Densities and Wood Volume, High-Density Area Method Year West Year East Air photo interpretation         Sampled area (km2)   2.03   5.3–6.4 Density (trees/ha) 1965–66 NA   7.2   1972 10.0   7.3   1981 12.3   7.3 Ground survey         Sampled area (ha) 1985 94 1990 31 Density (trees/ha)   22.0   11.5 Wood volume (m3/ha)a   12.4   19.6 Size of area (km2)   170   311 NOTE: NA indicates data not available. a The estimates of wood volume are notional, being based on girth at breast height and using the volume of a cylinder, a formula that has not been tested empirically in this ecological system. SOURCES: Data from J. Nichol, in Cline-Cole et al. (1990); Mortimore et al. (1990). farming system (with the exception of the date palm, which is grown on plantations on well-watered lowland sites). Neither have livestock, which spend the greater part of the year on natural grassland, only visiting the farms when crop residues are available. Farm Trees and Population Density In northern Nigeria (and elsewhere in West Africa), the transition from bush fallowing to annual cultivation (normally associated with increasing population density) leads to an increase in the size of trees and in timber volume per hectare (Cline-Cole et al., 1990; Pullan, 1974). In the high-density area, the density and regenerative status of farm trees has been maintained over a period of three decades including two major drought cycles. The value and multiple uses of farm trees have ensured this outcome, notwithstanding inflating woodfuel prices in nearby Metropolitan Kano. Increased integration of farm forestry with crop and livestock husbandry is essential to the process of intensification and is consistent with an increasing density of population. In the low-density area, a lower level of integration and an underdeveloped farm forestry component reflect a lower population density (a more

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop TABLE 7 Percentages of Girth Size Classes of Farm Trees, High-Density Area   Area   Class (GBH in m) West East 0.0–0.19 32 NA 0.2–0.39 17 NA 0.4–0.59 10 NA 0.6–0.79 7 NA 0.8–0.99 8 NA Subtotal 74 48 1.0–1.9 NA 33 2.0–2.9 NA 11 3.0–3.9 NA 4 ≥4.0 NA 4 Subtotal  26 52 Total 100 100 NOTE: GBH girth at breast height. NA indicates data not available by class. SOURCES: J. Nichol, in Cline-Cole et al. (1990); Mortimore et al. (1990). abundant supply of land), more natural woodland (on lowland sites), and also ecological constraints affecting the growth of trees in upland areas. DEMOGRAPHIC VERSUS NONDEMOGRAPHIC FACTORS The evidence reviewed so far leads to the conclusion that population growth, and high population density, are compatible with sustainable resource management by smallholders. A relationship between population growth and agricultural intensification in northern Nigeria could be inferred long before Ester Boserup's elegant statement of her hypothesis in 1965.12 A review of the farming system of the high-density case (the Kano Close-Settled Zone), during the last three decades, concluded that it is sustainable as a system (Mortimore, in press). We may now add that intensifi- 12   The relationship is implicit if not explicitly stated in early colonial annual reports and district assessment reports for Kano Province, for example, and even in the accounts of nineteenth-century explorers. For an early formal statement, see Grove (1961).

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop cation appears to promote sustainable management of soil and farm tree resources. Ecological variation, while complicating the spatial pattern, lends further credence to the demographic hypothesis in the sense that highly productive land (usually lowland) is relatively scarce in relation to less productive land (upland), and thus in greater demand. Scarcity has the same effect as a high population, and so consequently many of these areas have been invested in and intensively used, even under relatively low population densities. The link between population growth and sustainable intensification is mediated by other factors that can only be briefly touched on here. Land appropriation (by governments, institutions, and capitalist farmers) proceeded apace in Nigeria under the impetus of oil revenues and the making of personal fortunes in the 1970s and 1980s (Watts, 1987). Access to land is facilitated by a land law that accords separate status to customary and statutory tenure, almost unrestricted powers to state governments' land offices to effect transfers, and by subeconomic rates of compensation paid to customary claimants. The removal of increasing amounts of land, cultivated or uncultivated, from the stock available to the growing population of smallholders, may be expected to accelerate the effects of population growth on the use of what remains. There is also a process of purchase, consolidation, and enlargement under customary tenure, which operates in favor of an emerging class of farmer entrepreneurs (Labaran, 1987), which is also reported on irrigation schemes. In short, a process of competitive appropriation is increasingly affecting the supply and distribution of agricultural land. A further limitation of the demographic hypothesis of land use change arises from the fact that population growth does not translate directly into increased inputs of agricultural labor. At the level of the household, differences in access to land cause unevenness in family labor inputs. The hiring market either corrects or accentuates such unevenness, depending on the distribution of operating capital and of household poverty within the community. Capital may be spent on hiring labor or on saving it (e.g., by buying ox-plows). This technology has contradictory effects in densely populated areas, saving labor in land preparation but creating extra demand for it in hand-weeding operations. Actual family labor inputs are also influenced by their opportunity costs in alternative income earning activities, at home or away; by the sexual division of labor; by labor withdrawal for education; and by random indisposition. Among these, the opportunity costs seem to be crucial. Thus the extent of a household's dependence on homegrown food cannot really be described. It is a function of the structure of alternative opportunities open to the individuals within it. If off-farm activity offers a better perceived

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop return than labor invested in subsistence production, the link between population growth and land use is no longer inevitable. Markets are not new to northern Nigeria. The Kano Close-Settled Zone produced grain for the precolonial city of Kano, and the commercial value of lowland was recognized in a special tax. Farmers might earn incomes as laborers, craftsmen, or traders, especially during the dry season. The precolonial economy of the Manga Grasslands was based in part on the production and sale of potash and salt from saline lake beds. Livestock were also bought and sold. Markets were linked by long-distance trading networks, and both local and travelling traders operated in rural areas. Export crop production (groundnuts and cotton in particular) added to the demand for land, because households maintained a subsistence priority, though by the 1960s, land-poor producers in Kano had to sacrifice a part of their grain output to participate in the groundnut market. Since the demise of the groundnut as an export crop (in 1975), its place has been taken (to a large extent) by grain production for urban markets. In lowland irrigation in particular, market development has an important impact. In a study of fadama (lowland) use in nearly 19,000 km2, Turner (1977) found that distance to markets had a strong correlation with the proportion cultivated. But the next most important variable was settlement density, a proxy for population density. The penetration of markets into the farming sector has received added impetus from inflationary food prices, especially during the last decade. Markets for produce, inputs, working capital, land, and labor, all affect agricultural activity to an increasing extent, even in areas remote from towns. Market growth and population growth were linked in Boserup's (1965) statement of the demographic hypothesis. Land transformation is tied to both. The extent to which it correlates with changing population density (apparently independent of market influence) is attributable to the continuing priority accorded by peasant smallholders to household subsistence production. This, it is well known, is related to risk. Risk comes from the unpredictable operation of ecological factors (especially rainfall) and of the political economic environment. Our study areas are marginal with respect to both. CONCLUSIONS Two farming systems in semiarid Nigeria—high- and low-density cases—that have population growth rates probably lying between 1.5 and 3.0 percent per year, represent two different points in the process of land transformation. These points have cultivated percentages of over 80 in the high-density case and about 10 in the low. Smallholder investment in land improvement is relatively far advanced in the high-density case and has barely begun in

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop the low (except for moisture-rich lowland sites). Soil fertility is being managed on a sustainable basis on the permanent fields of the high-density case, whereas in the low-density case, long fallows are relied on. The farm tree component of the high-density system—well integrated with crops and livestock—is also being managed sustainably, whereas farm trees do not play a significant part in the low-density system, except on lowland sites. It is concluded that population growth, and high population density, are compatible with sustainable resource management under smallholder conditions. The evidence on soil fertility and farm tree management in the high-density case is derived from small samples, restricted to one farming system, and the conclusions derived from it are provisional. Nevertheless, they challenge the view, commonly held, that population growth necessarily puts destructive pressure on smallholder farming systems, and especially the rapid rates of growth that have been experienced in the last two decades. Growing population density may take expression, in a farming system in which smallholders aim to produce a large proportion of their subsistence, in either an intensification or a degradation pathway. What Lele and Stone (1989) call ''autonomous intensification'' has been proceeding in northern Nigeria for decades, and there is a case for "policy-led intensification" to consolidate the gains, and to minimize the possibilities for degradation. But the threshold for a transition from a degradational to an intensification pathway (the possibility of which was recognized three decades ago by Prothero, 1962) is not clearly understood owing to the shortage of empirical studies. Figure 3 (which is illustrative) suggests that irreversible degradation (frequently asserted in the literature about semiarid Africa) may be more accurately portrayed as a low-level equilibrium, under which economic yields, however scanty, continue. The high-density case, based on the evidence presented here, represents either stable intensification (no degradation and no yield improvement) or improving intensification (sustainable improved yields). Obviously the second is a desirable policy objective. But to induce intensification in the low-density case would be premature because the factor ratios (principally labor to land) are not yet appropriate for intensification, population growth being much further behind. Sustainable resource management and stable or even slowly improving economic yields may not assure stable or improving incomes. This consideration has led many observers to question whether the smallholder intensification that was adequate in the past can respond to recently accelerated rates of growth, and if not, whether a farming system will not relapse into a degradational mode (see, e.g., Lele and Stone, 1989). This view rests on the assumption that smallholder households are and must continue to be self-sufficient in food production, an assumption that underlies the concept of human carrying or supporting capacities. At the national level, such an assumption may be a policy directive.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop Figure 3 The transition from degradation to intensification in a farming system. The assumption of self-sufficiency is not appropriate at the level of the farming system, at which the link between the population and the land is not immutable. In particular, the myth of a full-time farming peasantry, exclusively dependent on the produce of the smallholding, should be discarded. Migration and circulation reduce the demand for food and generate income. Food crops may be sold as well as market ("cash") crops. Agricultural incomes may be used to finance education or employment while off-farm earnings may be invested in agriculture or land improvement.13 13 There are interesting parallels between the experience in the Kano farming system and in the Machakos District, Kenya, where rapid population growth has accompanied a transition from extensive to intensive farming and very substantial investments in land and water conservation during the last three decades (the preliminary output from this study is available in a series of working papers published by the Overseas Development Institute, London).

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop Wolman and Fournier's (1987:42) recommendation that "incentives need to be created for the farmers to remain on the land and make it produce" can be restated: "incentives should be created for smallholders to invest in the land to make it produce"—for even in the semiarid zone, the technical possibilities for increasing productivity have not been exhausted. But the risks to equity of unimpeded capitalization in land are considerable. Unless rights of access to land can be guaranteed (as was attempted in Kenya by land adjudication), the free operation of the market will have a negative effect. The new large landowners in northern Nigeria are not generally remarkable for their interest in intensification, conservation, or sustainability. ACKNOWLEDGMENTS The work reported in this paper was partly done under contract to the Federal Agricultural Co-Ordinating Unit, Ibadan, Nigeria, under the terms of the World Bank's Agricultural Sector loan to Nigeria, and the permission of the head of unit to publish it is gratefully acknowledged. REFERENCES Boserup, E. 1965 The Conditions of Agricultural Growth. London: Allen and Unwin. Bradley, A.K., S.B.J. MacFarlane, J.B. Moody, H.M. Gilles, J.G.C. Blacker , and B.D. Musa 1982 Malumfashi Endemic Diseases Research Project XIX: demographic findings: population structure and fertility. Annals of Tropical Medicine and Parasitology 76:381–391. Cline-Cole, R.A., J.A. Falola, H.A.C. Main, M.J. Mortimore, J.E. Nichol, and F.D. O'Reilly 1990 Woodfuel in Kano. Tokyo: United Nations University Press. Federal Department of Forestry 1977 [Nigeria] Vegetation and Land Use, Map on the Scale 1:250,000. Ibadan, Nigeria: Federal Department of Forestry. Field, N.J., and W.G. Collins 1986 Land use from aerial photographs in the Nigerian savanna. Pp. 435–440 in M.C.J. Damen, G. Sicco Smit, and H. Th. Verstappen. eds., Remote Sensing for Resources Development and Environmental Management. Proceedings of the Seventh International Symposium, Enschede, 25–29 August. Rotterdam: Balkema, Green, L. 1970 Population Models for National Planning. Ibadan: Nigerian Institute for Social and Economic Research. Grove, A.T. 1961 Population densities and agriculture in Northern Nigeria. Pp. 115–136 in K.M. Barbour and R.M. Prothero, eds., Essays on African Population. London: Routledge and Kegan Paul. Hill, J. 1986 Kano ADP Fertilizer Study Results. Draft manuscript. Kano State Agricultural and Rural Development Authority, Kano, Nigeria.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop Hill, P. 1977 Population, Prosperity and Poverty: Rural Kano, 1900 and 1970. Cambridge, England: Cambridge University Press. Jones, M.J., and A. Wild 1975 Soils of the West African Savanna: The Maintenance and Improvement of Their Fertility. Harpenden, England: Commonwealth Agricultural Bureau. Labaran, A. 1987 Land appropriation for capitalized farming in the Sokoto region: some preliminary findings. Pp. 46–57 in M. Mortimore, E.A. Olofin, R.A. Cline-Cole, and A. Abdulkadir, eds., Perspectives on Land Administration and Development in Nigeria. Kano, Nigeria: Department of Geography, Bayero University. Larson, W.E., and F.J. Pierce 1991 Conservation and Enhancement of Soil Quality. Paper presented at the International Workshop for Evaluation for Sustainable Land Management in the Developing World, Chiang Rai, Thailand. Lele, U., and S. Stone 1989 Population Pressure, the Environment and Agricultural Intensification. Variations on the Boserup Hypothesis. Madia Discussion Paper 4. Washington, D.C. : World Bank. Land Resources Development Centre 1979 The Land Resources of Central Nigeria. Land Resource Study 29. Tolworth: Land Resources Development Centre, Overseas Development Administration. McTainsh, G., and S. Stokes 1980 Environmental Data Bank for Central Northern Nigeria. Users Guide. Kano, Nigeria: Department of Geography, Bayero University. Mortimore, M. 1967 Land and population pressure in the Kano Close-Settled Zone, Northern Nigeria. The Advancement of Science 23:677–688. 1974 The demographic variable in regional planning in Kano State, Nigeria. Pp. 129–146 in B.S. Hoyle, ed., Spatial Aspects of Development. Chichester, England: John Wiley. 1989a Adapting to Drought. Farmers, Famines and Desertification in West Africa. Cambridge, England: Cambridge University Press. 1989b The Causes, Nature and Rate of Soil Degradation in the Northernmost States of Nigeria and an Assessment of the Role of Fertilizer in Counteracting the Processes of Degradation. Environment Department Working Paper 17, World Bank, Washington, D.C. 1993 The intensification of peri-urban agriculture: the Kano Close-Settled Zone, 1964–86. In B.L. Turner, Jr., R.W. Kates, and G. Hyden, eds., Population Growth and Agricultural Change in Africa. Gainesville: University Press of Florida. Mortimore, M.J., and J. Wilson 1965 Land and people in the Kano Close-Settled Zone. Occasional Paper 1, Department of Geography, Ahmadu Bello University, Zaria, Nigeria. Mortimore, M., E.U. Essiet, and S. Patrick 1990 The Nature, Rate and Effective Limits of Intensification in the Smallholder Farming System of the Kano Close-Settled Zone. Report to the Federal Agricultural Coordinating Unit, Ibadan, Nigeria. Morah, B.C., and O. Afolabi 1984 Nigeria Fertility Survey, 1981/82. State Level Report: Kano State. National Population Bureau and World Fertility Survey, Lagos.

OCR for page 42
Population and Land Use in Developing Countries: Report of a Workshop Pingali, P.L., Y. Bigot, and H.P. Binswanger 1987 Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa. Baltimore: Johns Hopkins University Press. Prothero, R.M. 1958 Northern Nigeria Density of Population by Districts, 1952 Census. Federal Survey Department, Lagos. 1962 Some observations on desiccation in North-Western Nigeria. Erdkunde 16:111–119. Pullan, R.A. 1974 Farmed parkland in West Africa. Savanna 3/2:119–152. Turner, B. 1977 The Fadama Lands of Central Northern Nigeria: Their Classification, Spatial Variation, Present and Potential Use. Unpublished Ph.D. dissertation, University of London. Watts, M., ed. 1987 State, Oil and Agriculture in Nigeria. Berkeley: Institute of International Studies, University of California. Wolman, M.G., and F.G.A. Fournier, eds. 1987 Land Transformation in Agriculture. Scientific Committee on Problems of the Environment 32. Chichester, England: John Wiley. World Bank 1984 Population Change and Economic Development. Washington, D.C.: World Bank. Young, A. 1993 Soil monitoring: a basic task for soil survey organisations. Soil Use and Management 7:126–130.