Thayer Watkins
Silicon Valley
& Tornado Alley

The Office du Niger and
the Scheme to Irrigate the Sahara Desert

A city planner in Chicago at the turn of the century, Daniel Burnham, advised,

Make no little plans. They have no magic to stir men's blood and probably themselves will not be realized. Make big plans; aim high in hope and work, remembering that a noble, logical diagram once recorded will never die, but long after we are gone will be a living thing asserting itself with growing intensity.1

Burnham's words are illustrated in the idea of irrigating the Sahel region south of the Sahara Desert in the vicinity of Timbuctoo. A French general at the turn of the century first proposed that this area, called Macina, be irrigated to grow cotton. Nothing came of the proposal at that time but later a French engineer devoted his life to the project and in the 1920's finally convinced some influential politicians to undertake the scheme. The project undertaken was, alas, a failure, but for social or ethnological reasons rather than engineering shortcomings.

The hydrological basis for irrigating the Sahara is not as farfetched as it might seem. A million years ago the present day Niger River was two rivers. The Lower Niger, as now, flowed into the Atlantic, but what is now the Upper Niger flowed northeast into the Sahara creating a lake comparable to Lake Chad. Later, when the Upper Niger broke through into the basin of the Lower Niger the lake dried up, but even then, during times of floods, water would flow north into the desert. Farmers used these flood waters to grow crops from the days of the New Stone Age down to about the ninth century A.D. Conquering Islamic armies destroyed the irrigation works and the inhabitants fled to the rain forests to the west. The area became virtually uninhabited but natives of the region still have legends of the time of agricultural prosperity.2

A study commission in 1919 concluded that the area at Macina where the prehistoric river had filled in the flood plain would be well suited to growing cotton by irrigation. De Wilde suggests that the commission hoped to have France duplicate the British success of the Gezira Scheme in Sudan.3 The Gezira Scheme in Sudan produced long staple cotton which requires a very arid climate to ward off plant diseases.

A pilot project near Bamako irrigated 7500 acres in the 1920's and provided a livelihood for 6000 people. Rice was the subsistence crop and cotton the cash crop.

In 1931 the French government approved a project for irrigating 2.5 million acres.4 Furon says the initial plan called for irrigating 4.5 to 5 million acres for 1. 5 million people producing, among other things, 330,000 tons of cotton per year.5 It was to be an island of prosperity. The Office du Niger was created in 1932 as an autonomous public enterprise. The Office du Niger had responsibility not only for building the irrigation system but also for recruiting and settling farm families in the region. The people to occupy the land would have to come from the more heavily populated regions of Upper Volta (now Burkina Faso) and French Soudan. Over time the Office du Niger became more and more self-sufficient in the sense of providing infrastructure, education, and credit to the settlement families, although it always needed public funds to meet costs. As de Wilde says, "Rather than letting all the administrative services of the French Soudan contribute, each according to its own competence, the Office du Niger became progressively more or less sovereign within the limits of its territory. " 6 This is a familiar phenomenon in bureaucracies; it is called empire building.

The construction of the irrigation system was carried out by the Office du Niger itself. The basic elements were a barrage across the river and canals to carry the irrigation water. The literature is a bit inconsistent on the details of the system. Furon speaks of a dam being completed in 1941 at Diamarabougou, but de Wilde makes no mention of Diamarabougou and indicates that only a barrage was built.7,8 The technical difference between a dam and a barrage is important. A barrage only diverts water from the river and does not create a storage reservoir as does a dam. Without water storage the system could not provide irrigation at the end of the dry season to soften the ground before plowing. It would be extremely difficult to create a storage reservoir in this region where the slope of the ground is only two or three inches per mile. This is where the Niger River has its Inland Delta,"a twenty-mile wide moving lake." 9

The unusual topography of the region did lead to quite a number of technical flaws. Some land which was leveled by the Office du Niger proved to be too high for irrigation and other land was too low for effective drainage. Some drainage canals had to be converted into water supply canals and vice versa. Insufficient drainage and maintenance in some areas led to an invasion of wild rice, a weed which hurt the production of domestic rice. 10

Although there were technical flaws, de Wilde says, after an extensive study of the project for the World Bank, "It is above all the shortage of manpower that has prevented the scheme from becoming the expected success."11 Most of the infrastructure was fixed costs for the project which was to irrigate at least 2. 5 million acres. Instead of these fixed costs being spread over a project providing the livelihood for 1.5 million people there were only 37,000 people and 112,000 acres in the Office du Niger when the administration was turned over to the newly formed government of Mali in 1962. Two years later the population had fallen to 33,000 but there were still 2,500 permanent employees of the Office du Niger12

At its inception the Office du Niger thought it would only have to offer free irrigated farms and people would flock to the project. But the settlers had to come from great distances which meant they had to foresake their familial, political and cultural affiliations and place themselves under the uncertain jurisdiction of the Office du Niger. he Office du Niger offered inducements such as two oxen and a plow, which for farmers used to hoe agriculture was revolutionary, but it was still not enough. 13

Even though the relatively high farm incomes in the project area were not enough to attract the numbers of people required to make the project a success these incomes were only achieved through the increasing indebtedness of the Office du Niger. Even its current operations had to be subsidized. The public capital invested in the Office du Niger by 1960 was $175 million.14 During the colonial period the Office du Niger absorbed 48 percent of the public investment in the region in the form of loans as well as grants. At various times the Office du Niger experimented with different approaches in an attempt to make the project work. It tried direct farming of the land using mechanized methods to cope with the labor shortage.16 On these efforts de Wilde concludes:

Although there are no recent or reliable data on the cost of mechanization, it has apparently been an expensive expedient. In the Office du Niger the costs have been high in relation to the value of output, partly because of the long distance of the scheme from both markets and the sources of supply.17

The failure of all the Office du Niger's best efforts should have signaled the fact that the project was a mistake--a false dream. The success of the Gezira Scheme in Sudan could not be duplicated because the Office du Niger territory, as dry as it was, still had too much rain at the wrong time of the year for growing long-staple cotton. Politicians in France hoped to create an island of prosperity, a "second Egypt" linked to France by a trans-Sahara railroad, which would serve as a strategic reserve in time of war. But this strategic isolation made it uneconomical for any type of agriculture. The continued support of the Office du Niger in the face of failure and mounting debt can only be explained as an attempt to salvage national prestige.

The people of West Africa who refused to migrate in spite of the inducements of the Office du Niger were probably economically justified in their decision. The settlers in the project did not get title to the land, although the Office du Niger gave them inheritable rights to cultivation of it. Soon after the country became independent the Mali government nationalized the land.18 Even without nationalization the farmers would have been hard hit by the drought that left the project without irrigation in the early 1970's.

National Geographic did a report on the Niger River after the end of the drought in 1977. In this report there is not one word about the Office du Niger.19 It is as though the project and all the badly needed development funds it used disappeared into the desert sands without a trace.

Although the Office du Niger was generally considered a failure it was not completely gone. In the early 1980s one third of the area that had been developed had been abandoned and only six percent of the targeted development area had been developed. Cotton production had been abandoned and paddy rice production yields were only 0.6 of a ton per acre. The settlers were unhappy with the conditions.

In 1982 the first rehabilitation project for the Office du Niger was initiated with assistance from the Government of the Netherlands. In the course of twelve years the yield of rice rose from 0.6 of a ton per acre to 2 tons per acre. The settler population more than doubled but area under cultivation did not increase. The productivity gain came in part from the adoption of high-yield rice varieties and a rehabilitation of the water delivery and drainage system. By 1994 only 40 percent of the area was served by the rehabilitated water control systems.

Another source of improvement in the project was the privatization of the processing and marketing functions. In the original scheme farmers were required to have their production processed by the mills of the Office du Niger organization. In the 1980s private businesses were allowed to compete for the farmers business and by 1994 only about 4 percent of the processing and marketing of the project production is handled by the Office du Niger organization itself.20


1. Quoted in Peter Blake, Form Follows Fiasco: Why Modern Architecture Hasn't Worked, (Boston, 1977), p. 111.

2. Raymond Furon, The Problem of Water: A World Study (New York, 1967), p. 15.

3. John Charles de Wilde, Experiences With Agricultural Development in Tropical Africa (Baltimore, 1967), VolumeII, p. 262.

4. Ibid., p. 245.

5. Furon, P. 154.

6. de Wilde, p. 248.

7. Ibid., p. 245.

8. Furon, p. 155.

9. Georg Gerster, "River of Sorrow, River of Hope, " National Geographic, August, 1977, p. 154.

10. de Wilde, p. 288.

11. Ibid.

12. Ibid. , p. 246.

13. Ibid., p. 254. The oxen and plow were sold at subsidized prices on credit rather than given outright. If the loans were never paid off then Office du Niger got the expense of a gift without the full inducement of a gift.

14. Ibid. , p. 247.

15. William Jones, Planning and Economic Policy: Socialist Mali and Her Neighbors (Washington, D.C., 1976), p. 25.

16. de Wilde, p. 246.

17. Ibid., p. 248.

18. Ibid., p. 254.

19. Georg Gerster, pp. 152-189.

20. World Bank Group, Findings: African Region, No. 61 (April 1996).

HOME PAGE OF applet-magic
HOME PAGE OF Thayer Watkins