unpublished thesis

Monique Lakroum, "Chemin de Fer et Réseaux d'Affairs en Afrique Occidental: le Dakar-Niger 1883-1960" (University of Paris VII, 1987)

ANRT Doctorat d'État 87/PAO7/0079, copy available at Bibliothéque de l'Université de Paris VII
Notes © 1999 by Jim Jones, Ph.D.

Go to Archive Table of Contents Read Disclaimer

Table of Contents


Chapter One: Introduction

(p8) Larkoum warns the reader about certain problems with the statistics on freight carried by the Chemin de Fer Dakar- Niger: The French omitted statistics for certain months because the figures were so low. The railroad only carried large quantities during May-Sept. In addition, the statistics more often reflected the wishful thinking of officials in metropolitan France than the realities of Africa.

(p9) Types of things measured by Lakroum: amount of traffic, amount of freight, profit margins, freight rates, interest rates, amount of capital at risk

(p10) The fundamental difference between African and railroad-based business networks was the hierarchical nature of railroad-based business. African networks consisted of many transactions between poles of roughly equal size, while the railroad organized commerce into a pyramid topped by a few large commercial centers like Dakar and Bamako. This was the main reason why African trading networks, despite their inherent flexibility, were unable to adapt to the railroad.

PART ONE: When big business (les industriels) discovered Africa (1878-1923)

Chapter Two: direct administration or concession: the bets of construction [page 99]

(p117) This chart shows the construction costs for the Kayes- Niger from 1881-1910:

Period       cum. km           cost       new km     cost/km
1881-1884         54     13,680,500           54     253,343
1885-1890        128        400,000           74     110,004
1891-1898        176      6,417,000           48     116,643
1899             195      4,214,000           19            
1900             240      3,873,000           45            
1901             283      6,365,000           43            
1902             312      9,757,000           29            
1903             385      5,008,000           73            
1904             555      2,620,000          170      94,296
1905             555      1,214,000            0            
1906             555        136,000            0            
1907             555         28,000            0            
1908             555        429,000            0            
1909             555        273,000            0            
1910             555        135,000            0            
TOTAL            555     54,550,000            0      98,287

(p136) The members of the Chamber of Commerce of Kayes (founded in December 1892), tried to convince the French authorities that they needed to build a railroad along the entire length of the Senegal River in order to make regular commerce with the interior feasible.

(p137) Negotiations over railroad construction between the European merchants and the administration broke down because their interests were different. Kayes was prone to flooding (Galliéni called it "un coin de pestilence") so the administration moved its facilities up the hill to the plateau. Only the merchants remained near the river, where they could unload easily from ships.

(p149) Table showing the total amount of money spent in building the Chemin de Fer Thiès-Kayes between 1904 and 1923

(p150) Table showing the industrial cost of living index from 1913 to 1922

(p152) The task of converting a military railway into an economic instrument was slow, late and often incomplete.

(p152) From the earliest years of railway planning, it was widely believed that the construction of a railway had the effect of creating a new coastline. This idea took as it premise the belief that the railway would create its own traffic. The military officers in charge of the railway took this as justification for other policies such as resettlement of populations. However, they had a difficult time convincing European merchants that the railway would create traffic, as shown by the struggle to set railway tariffs.

(p153) The arguments were made to justify work already in progress, as well as extensions that were still in the planning stage.

(p153) Usually, the Dakar-Saint Louis was chosen to illustrate how a railway could turn an arid plain into fields of export crops. It showed how imperialism could complement economic development.

(p153) The administration wanted European merchants to participate in the conquest by installing comptoirs near forts as quickly as possible, as far as the Niger Bend.

(p154) Records of military officers showed that they envisioned their role as organizers of the space into which French commerce would flow. However, this idea was based on an illusion -- that the land was totally empty or "vierge."

(p154) Moreover, French merchants had nearly two centuries of prior experience conducting trade without administrative control. As a result, official reports consistently mention the hostility that they faced from European merchants.

(p154) In particular, European merchants refused to divulge the contents of their balance sheets. That made it impossible to determine the value of the commerce created by the railway, or predict the results of additional construction.

(p154) Without real data to go on, the railway planners used theoretical models based on the population density in the squares of a grid drawn along the railway line, allowing for trade between squares, and assuming that imports would be double exports.

(p155) Formula used to predict railway traffic: Number of Tons of Imported Freight X Population near railroad / Portion of the population that uses European goods

(p155) Description of an early model for the area served by a railroad. Let P=cost of porterage/weight and p=cost of rail freight/weight. Y=distance from the coast along the rail line and X= the perpendicular distance from the rail line. Then the equation becomes PY=PX + pY, which reduces to Y=PX/(P-p). This formula describes a parabola. The model assumes that traffic will follow the shortest route and that transporters will always seek the shortest route to the cost. [See: Bulletin de la Société Francaise des Ingenieurs Coloniaux, Capt. Crosson-Duplessis, "Étude sur les Zones de Trafic des Chemins de Fer Coloniaux de l'Afrique Occidentale" #39 (1er trimestre 1906), pp59-99]

(p161) Tariffs were set at the opening of the railroad, but modified several times. In 1896, different rates were set for inbound (0.50 franc per ton-kilometer) and outbound (0.07 franc per ton-kilometer) rates. In 1900, additional rates were defined for low value bulk items. All of these tariffs were reorganized in April 1902. [see Arrêté du 25 Avril 1902 in Journal Officiel de l'AOF (1902).] The 1902 tariff retained different inbound and outbound rates, but favored long distances and provided different rates for different classes of items. In 1910, passenger rates were reformed. [Arrêté du 8 Julliet 1910]

(p169) Railroad tariffs were modified in 1917 to favor the import of salt and construction materials. ["Arrêté du 20 Septembre 1917, créant les tarifs spéciaux n°9 et 9bis, pour le transport du sel marin; "Arrêté du 24 Décembre 1917, créant le tarif spéciaux n°10, pour le transport du sable, du gravier et des moellons" in Journal Officiel de l'AOF (1917).]

In 1923, a special railroad tarif was introduced for the transport of peanuts. It was a regressive tariff that favored the transport of peanuts over 50 and 80 kilometers, and in quantities of more than 5 tons. [GGAOF, Chemin de Fer Thiès-Kayes-Niger, tarifs des transports à grande et petite vitesses (Gorée, 1923), 86pp.]

(p170) The French planned for the railroad to serve as the logistical axis for agricultural colonization.

(p170) The railroad served to drain labor out of the interior towards the coast, beginning with the Chemin de Fer Thiès-Kayes construction.

(p170) By the beginning of WWI, a large number of laborers, mostly Bambara, had already migrated from the headwaters of the two rivers towards Sine Saloum. See "Proces Verbal de Chambre de Commerce de Kayes," (Kayes, 30 May 1911) in ANM 1 Q 105.

(p171) In 1912, a siding from Guinguinéo to Kaolack was built to facilitate peanut exports.

(p171) Extension of peanut culture along the railroad: In 1900, it began at Kaffrine (Senegal), by 1910 at Tambacounda (Senegal), and by 1925 at Kayes (Mali).

(p182) charts showing the distribution of capital in the largest French-owned enterprises in AOF, 1900, 1910 and 1920. The private railroads (Chemin de Fer Dakar-Saint Louis and Chemin de Fer de Dahomey) were first in all three years.

SECOND PART: Confrontation between two sorts of logic

Chapter Three: Restructuring the railway network [page 216]

(p218) This section tries to explain the poor integration of African and railroad-based networks.

(p218) African merchants: every African dreams of traveling to become a merchant. It defies the European laws of the marketplace, supply and demand, etc. Endnote #2 (which appears on p302) has a bibliography of the informal economic sector in West Africa.

(pp220-221) The African merchant did not use a European concept of market. His market was at the edge of population centers, or in the houses of acquaintances. The latter was invisible to all but himself. This structure of trading relationships was strong enough to resist assimilation by other groups, political or agrarian (probably means family lineage).

(p221) Questions concerning "tradtional trade": what is the relationships between exchanges and transcontinental trade? What routes did European goods follow to any given location? How did the form of transport affect the outcome?

Attempts to answer these questions have produced contradictory answers. Some say that the European commercial system destroyed the traditional system (Colin Newbury, Claude Meillassoux). Others say that it established a superstructure that provided continuity to all of the different trading structures in existence. (Coquery-Vidrovitch)

(p222) One thing is for sure. Merchant capitalism was a force in the 19th century that was capable of transforming society, just like religion and war. [Endnote mentions Yves Person and Samory: Une Revolution Dyula.]

(p223) The essential difference between the European and African trade networks was the profit motive. An unknown portion of the African trade was siphoned off to maintain the social network. The great empires of the Soudan were created by governments who learned, not how to squeeze the most taxes out of the villages, but how to maintain the conditions for long distance trade, and tax it.

However, long distance trade also meant theft, requisition, redistribution, booty and tribute. This entire system of commercial distribution contained internal contradictions. It could not exist unless zones of productions were separated and commercial relations controlled. This was done by controlling the transport system.

(p225) Precolonial trade networks depended as much on the cost of transport as on the value of the merchandise traded. The African merchant was both trader and transporter.

(pp224-226) example of Banamba, where Moors and Julas traded salt and cloth. Lakroum uses a complicated calculation to show that both made out better by trading in Banamba than in Nioro. She attributes this to the cost characteristics of the repsective methods of transport.

(p226) Typical series of trades used to generate a profit. Colas from the forest for desert salt. Salt for cloth produced in the forest. Cloth for gum gathered in the Sahel. Gum for European merchandise. Then start over.

This form of trade followed routes that were so dynamic, that it is impossible to draw it on a map with precision. We have to find another way to portray its essentials.

(pp220-230) Lakroum's work provides insight into the relationship between African merchants and the rest of the population. It suggests the ways that the statistics might fail to reveal what actually happened - trades beyond French awareness, trades made to obtain social advantage where the profit is not immediately evident.

(pp226-227) Lakroum pays tribute to Baillaud's Sur les routes du Soudan (Toulouse: Privat, 1902).

(p227) Baillaud used customs statistics, but recognized that when they showed an increase, it was probably due to increased French efficiency at issuing patentes (trade permits).

(p229) Baillaud's map of caravan and portage routes in the Soudan.

(p230) Quotation to the effect that news is the only thing that travels fast in Africa: "Une seule chose va vite en Afrique, ce sont les nouvelles, et, d'après elles, les caravanes ne détournent de leur route sans se préoccuper de leur chemin ou de leur point d'arrivée. Un fois sur les sentes du Sahara ou de la brousse, les marchandises changent de destination au fur et à mesure des événements." [See E. Baillaud, Sur les routes du Soudan (Toulouse: Privat, 1902), 135.]

Europeans and Africans both took advantage of the speed of news to open new routes, create new trading points, etc. The only fixed points in the trading system were the transhipment points (points de rupture) along rivers and at desert oases.

The constantly shifting locations and route of the African trade system are the subject of Lakroum's inquiry. She asks how the introduction of the railroad, which produced fixed trading structures, affected the fluid African structures.

(p231) Le Charge est l'unite de compte que le transport escompte. In other words, the fundamental unit of trade was the "load," a fact that was obscured by European concepts of trade.

(p231) Pieces of guinea cloth became accepted as units of value, treated like currency in Nioro in 1898-1899. The French tried to accept them in payment of taxes, but without being able to control the number of "pieces" in the market, the value was subject to wide fluctuations that threatned tax revenues.

(p232) The nature of the load determined many things about a caravan - its destination, the types of transactions and the margin of profit.

The nature of the transport method, by determining the amount that could be carried in a load, simultaneously determined its price. Europeans have trouble with this concept because they are accustomed to thinking of price and quantity as independent characteristics.

(p233) When the railroad introduced bulk shipments, it destroyed the traditional system of setting prices based on the size of the load.

(p234) Prices of guinea cloth in St. Louis and on the Niger in 1887, showing the difference (markup and transport costs) was roughly 300% by the time it reached the Niger Valley.

(p235) Sugar at St. Louis, Kayes and Bamako in 1887 showed a 13.63% markup at Kayes and a 36.36% markup at Bamako.

(p235) Salt. Desert salt did not compete with Senegalese sea salt (mined at Gandiole) until 1896. French military personnel assigned to the railroad believed that the trade in sea salt was essential to the railroad's economic viability.

The European product was no better, and actually more likely to be damaged in transit, but it benefited from advantages in transport and packaging.

(pp236-237) Detailed discussion of competition in the salt market. European salt was certainly priced to beat desert salt. The two types of salt did not compete directly in many places. Nevertheless, in 1896 the Moors increased the flow of salt to markets, but the Jula distribution channels absorbed much of the price advantage.

(p239) African trade system had two major regions - the north, pieced by Moor caravans that traveled to specific transhipment points like Timbuktu, and a network of unfixed trading routes in the Sahel.

For traders, the selling price was not as important as the availability of a return cargo.

Time had a completely different meaning in African commerce, or rather, less meaning than it did for European merchants. Long slow routes, warehousing, delays, etc were common.

(p240) Description of a trading voyage in 1898 from E. Baillaud:

Location Bought Sold
Ouagadougou 2,500 cola for 150 francs -
30 days march - -
Timbuktu 20 barres salt for 500 francs 2,500 colas for 500 francs

The merchant continued on in this fashion, selling cattle, slaves, etc, and built his fortune. Although most merchants started with very little capital, it was rare to successfully increase it at each transaction. Due to the lack of coins, many transactions remained more theoretical than real.

(p240) Credit played a large part of African transactions. Baillaud told of a merchant in Timbuktu who died leaving behind IOUs worth 200,000 francs and a single 5-franc piece.

(p242) Martonne's map of caravan routes and days of voyage in 1921

(p243) Days on the road was the measure of a journey, not kilometers. Europeans calculated averages based on the daily distance covered by a mule or ox, but the real times differed. To the north of the line Matam-Bamako-Segu, they tended to excede the average, but towards the south, the times were relatively shorter.

Lakroum attributed this, not to climate or geography, but to the way transport in each region was organized.

(p244) Profits were the result of local scarcity of products that was maintained by separating zones of production from zones of consumption. As a result, the speed of transport was not a factor in determining profit. The density of the trading network was more significant for the level of profit. In the region of Haut Sénégal-Niger before the railroad, travel was slow and short distances added lots of time to the total trip.

(p247) The railway introduced a fixed trade route that cut across the major precolonial trade routes, which were themselves fluid. At first glance, one would expect the railroad to siphon off trade from the precolonial routes at points where they crossed the railroad.

(p248) On way to measure the impact of the railroad is to measure the flow of trade along each segment of the trade routes, and then determine how much trade from each segment was diverted by the railroad. Lakroum cites two weaknesses in the method, but the translation is difficult.

(p275) There were plans made at the time the railroads were consolidated (1923-1927) to close the stations between Kayes and Bafoulabé, including Galougo. [See ANS Series 6 P]

(p294) Although the four French West African railroads were never connected together, the French persisted in joining them under a single federal administration in 1947.

(p295) Although the federal administration of the railway was located in Dakar, the actual power became concentrated in Thiès where the workshops were located.

(p295) Two systems of work discipline developed as a result of the growing professionalism of the AOF railways. One was the French system, described by cheminots as "avec tout sa rigeur, et sans aucune familiarité." The other system existed among African railway workers, which they described as "impliquait travail, dévouement, sacrifice . . ."

(p296) The railways and ports of West Africa were all grouped under a single financial administration in 1936. [see endnote#133: Comptes définitifs des recettes et dépenses du Bidget unique des transports (Governement Général de l'AOF, 1939-1939).]

(p296) The creation of the Chemin de Fer de l'AOF had a salutary effect on the operation of the Chemin de Feer Dakar-Niger, whose "expenses d'exploitation" peaked in 1939 and 1948 before dropping to a gradual decline in the 1950s.

(p297) Railway worker salaries were the largest part of railay operating expenses. They rose continuously from 1924-1939 and from 1945-1960, interrupted only by World War II.

(p297) The biggest problem was to convince older railway workers to accept new methods of promotion, transfer and production. The administration introduced the graduates of technical schools as a means for federalizing the work force by creating l'esprit réseau.

(p300) One of the first grievances of African railway workers was the existence of two cadres for workers. They went on strike in 1947 for the creation of a cadre commun. [this statement includes endnote #130, notes on which appear immediately below]

(p314) Endnote #138: At first, workers in the cadre local were regulated by various arrêtés of the governors and lieutenant governers. The first attempt to standardize procedures regarding their status came with an arrêté fédérale on 12 March 1925, which defined the cadre local and its operation. A second arrêté fédérale of 7 March 1936 organized the cadre commun supèrieur. See "Arrêté du 12 Mars 1925" in Journel Officiel de l'Afrique Occidentale Française (1925), 179-248; "Arrêté du 7 Mars 1936" in Journel Officiel de l'Afrique Occidentale Française (1936); "Arrêté promulgant en A.O.F. le décret du 19 mai 1939 portant organisation du statut du personnel ferroviaire" in Journel Officiel de l'Afrique Occidentale Française (8 August 1939); 1018-1041.

(p300) Although the unification of personnel on all the railways should have logically been followed by unification of track and material, there were strong forces in opposition.

(p300) The "décret-loi de 28 fevrier 1944" turned the railway from a public corporation into an entreprise of "caractère industriel et commerial." [Endnote #139: See M. Martin, Organisation des chemins de fer de l'ancienne A.).F. (6 janvier 1959), 1. Unfortunately, two searches throygh Lakroum's bibliography yield no further information--I think she left it out.]

(p301) The law was not implemented until late 1946, but as of 1 January 1947, all of the workers, rolling stock and installations of the four West African railways were placed under a sinlge federal administration directed by a "conseil d'administration" composed of 18 members representing railway workers, local authorites and local business.

(p302) This placed the railway administration in a difficult position. They had to pay for the "federalization" program while managing a railway that was in precarious financial shape, and adding the cost of an additional layer of administration.

(p303) Lakroum says that the "évolution des effectifs et des status rend compte des modalités adoptées." She demonstrates that as the administration reduced the size of the workforce between 1938 and 1956 (15605-14042 by 1953 in AOF), they also raised an increasing number to the cadre (11-33% for AOF; higher on the Chemin de Fer Dakar-Niger).

(p303) During the debate on funding for the Chemin de Fer de l'AOF in 1957, proponents mentioned not only transport and economics, but also the railways usefulness as an instrument of political power. "les Chemins de fer constiuent un outil exceptionnel pour l'autorité qui en dispose, il faut bien recoonaŚtre que cet outil quelle que soit son importance, n'a de valeur que dans la cadre d'une politique générale." [Endnote #146: Intervention de HAMACIRE N'DOURE, Ministre du Commerce et des Transports du Soudan, Procés-verbal du Comité consultatif du Bakar-Niger (Bamako: 2 September 1958), 9, in ANM Fonds Chambre de Commerce de Bamako, non classé.]

Chapter Four: Railway negotiations -- the impossible accord [page 316]

(p316) decisive elements in the conflict between the railroad and the African trading network 1. le controle des circuits et les moyens d'echange 2. la maitrise du debit des railroad 3. la fixation des valeurs marchandes

(p317) In her attempt to describe trading networks in West Africa, the author has focused on the nodes and intersections, rather than the length of segments or their location.

(pp317-318) There are fairly detailed records made by cercle administrators of the population and identity of merchants in their jurisidiction. However, these lists tend to emphasize the role of European merchants over that of Africans or Levantine, and to overstate the role of the large merchant houses.

(pp318-319)) Prior to WWI, the European merchant houses were organized around large firms based in major French ports like Bordeaux. Each "mother firm" extended merchandise on credit to local comptoirs operated by Europeans. They in turn extended credit to African merchants who took the goods into the interior for sale, and returned with peanuts. These African merchants organized local transport using porters and animals, and supplied even smaller local merchants with Europan trade goods.

(p319) Each comptoir operated independently, and maintained its own books. The author suggests that each comptoir served as the focus for a complex set of influences on the African population in the vicinity.

Notes on the location of comptoirs (trading posts)

(p320) In 1912, comptoirs were found at the location of trading routes, especially where cargo had to be offloaded, as at the junction of a river and a piste, for example. Total number of comptoirs in AOF in 1912: 541

(p322) In 1920, there were fewer posts (494) and the decline took place at interior points along the pistes and rivers.

(p323) WWI helped to simplify the European trade network. So did the Conakry and Dakar railroads. In particular, the number of comptoirs along the Chemin de Fer Conakry-Niger and the Chemin de Fer Dakar-Saint Louis declined.

(p324) In 1924, the completion of the Chemin de Fer Dakar- Niger drew business away from the Conakry railroad. Total number of comptoirs in AOF in 1924: 574. Kayes declined while the number of posts increased around Dakar, Thiès and Kaolack, where peanut production was concentrated.

(p325) In 1931, the effects of the depression were still barely perceptible and the number of comptoirs reached the all- time maximum of 891. They were concetnrated along the Chemin de Fer Dakar-Niger and the Niger River, as well as at the coastal end of the AN railroad and the BN railroad. The Chemin de Fer Conakry-Niger suffered further losses in the number of commercial comptoirs.

(p326) Most commercial trading post development took place along the Chemin de Feer Dakar-Niger near Kayes because of peanut cultivation, and the Niger River bend, also along the Chemin de Fer Abidjan-Niger.

(p327) In 1931, the increase in trading posts reached a climax. The tendency was to create posts along the routes of the pre-colonial network, at the intersection of the routes between the two rivers and the Sahara-forest. The center of gravity of AOF commerce moved gradually towards the interior.

(p328) The depression forced the closing of many posts. Total number of comptoirs in AOF in 1937: 571

(p329) The retreat was greatest from the Niger River bend and other interior locations, including the north of Ivory Coast ...

(p330) ... and stations along the Chemin de Fer Dakar- Saint Louis. In 1947, total number of comptoirs in AOF: 574

(p331) Further centralization in Dakar, St. Louis, Kayes, Bamako, Conakry and Abidjan

(p332) The retrenchment was not only spatial, but also organizational. In 1953, there was increased hierarchization. French companies maintain trading posts in urban centers while they subcontracted to task of diffussion and concentration to African and Lebanese traders. The French acted as wholesalers for African and Lebanese retailers. At the same time, urban consumption increased with respect to rural consumption, permitting the French companies to earn enough in the cities. WWII had a lot to do with urbanization of African people.

(p334) This table shows the number of company comptoirs operating in AOF, by year.

Year Number of company comptoirs
1947 574
1951 675
1953 738
1955 1,262
1958 867
1962 795

(p334) In the last years, new trading posts were constructed at places that were expected to become new junctions between the railroad and local routes, like Niamey, Ouagadougou, Zinder, and Parakou (Dahomey).

(p335) Posts were located along the two main axes: Chemin de Fer Dakar-Niger and Chemin de Fer Abidjan-Niger.

The railway: the largest of the "marchés ambulants"

(p337) Rather than eliminate precolonial commercial networks, the railway increased their capacity. African commercial routes were so flexible that merchants easily incorporated the railroad into their travels. Europeans misunderstood this and attributed it to an African love of travel and disregard for comfort. "L'indigène voyage beaucoup et le confort lui import peu. Il voyage simplement pour le plaisir de voyager et et il lui arrive d'acheter du transport pour un prix determiné." [see Ingenieur en Chef de Travaux Publics des Colonies, Chef de Services techniques de l'Office central des Chemins de Fer de la France d'Outre-mer (J. Pialoux), cited by B. Peyret, 109.]

At first, 3rd and 4th class passenger receipts were the key to financial success, rather than frieght receipts. On the Chemin de Fer Dakar-Saint Louis, receipts for "petite vitesse" freight (bulk, heavy freight like peanuts) did not exceed passenger receipts until 1892. On the Chemin de Fer Kayes-Niger, the situation was reversed and freight made up three quarters of all receipts until the First World War. After the war, the proportion of freight revenue declined to 56.24% in 1923. After the junction in 1923, passenger recipts became more important, but fast freight (JJ: grande vitesse--shipped on passenger and mixte trains; the opposite of bulk freight) still acounted for 29.56% of all receipts between 1923 and 1957.

(p338) By offering a ticket holder the right to bring 30kg of baggage, the railroad changed the way that African merchants conducted business. 30kg was not much for a European civil servant, who could easily pay for overweight baggage, but it was about equal to a normal porter load.

(p338) Once the railway removed the physical limit on how much a colporteur could carry, there was temptation on the part of Africans to carry two or three times the normal load of 30kg. The railway administration tried to control baggage without success, and in 1936, in response to actions by European commercial houses, extended the 4th class tariff of the Chemin de Fer Kayes-Niger to the entire system, and baggage restrictions were relaxed to allow ticket holders to bring baggage like goats, other animals and fowl.

(p338) Also in 1936, the railroad removed the "tax d'enregistrement" (on goods?) which was tough to collect.

(p338) As a consequence, the amount of baggage rose 130% from 1935-1938 until the Vichy government, under pressure from European merchants who were being undercut by Jula merchants who carried as much as 60kg of goods on the train, reinstated the pre-1936 regulations in 1940.

(p339) The railroad ended up by creating "trains de marché" that rolled from market to market, permitting julas to conduct transactions in each one. But the train did not alter precolonial commercial practices. Only the price of the ticket was fixed. ...

(p340) ... Even railway employees became accomplices ("bon bougre") to precolonial commerce.

(p340) Concerning theft and corruption on the railroad: the archival sources provide fragmentary evidence of a widespread phenomenom. It is not possible to know what quantity of goods traded under the table, but with difficulty, it is possible to prove that it existed and determine how it worked.

(p340) Cheminot mutual assistance was based on both the example of European cheminots and various forms of traditional African group action. The government tried to control these organizations by making them offical, then restricting them.

(pp341-342) Railroad workers were able to acquire goods from the cooperative before the end of the month. They could sell them below market cost to local merchants, who could then undercut other merchants farther in the interior. cheminots and their cooperative also obtained free transit on the railroad for them and their goods.

(p343) This functioned like an interest-free advance on cheminot salaries.

(p345) Small entrepreneurs developed their businesses along with those of the large firms. At first, they depended on the large firms for their supplies of European goods, but as transport improved, they were able to go further to find supplies and could begin to challenge the large firms directly. (JJ: she's talking about trucks )

(pp347-356) series of maps showing the spread of merchant enterprises in AOF from 1912 to 1959.

(p352) Main conclusion is that little enterprises followed the big ones, so that by World War II, commerce was organized around a number of trees whose roots were in the port cities.

(p354) In the 1953, Bamako was a relay station for the largest commercial network based in Dakar. The second largest network was based on Abidjan and included Bobo and Ouagadougou. Lesser networks were based on Conakry and Lome and Cotonou.

(p357) The creation of comptoirs along the railway insured wider diffusion of trade goods. But seasonal variations in trade, plus bottlenecks that occurred on the railway made it difficult to rationalize the trade.

(p357) For the railroad to maximize profits from the shipmenmts of bulk items like peanuts, they had to release the peanuts slow enough to prevent a market crash, yet be able to take advantage of the moment when prices reached their highest. It appears that the effect of the railroad was to even out the variations in the price of peantus by releasing them slowly to the coast.

(p357) Correspondence between Chambers of Commerce and the railway administration show that there was competition between the two groups. The basic issue was how to control the railway's output.

(p358) The peanut trading season barely lasted six months from November to May. It began at the end of the hivernage when granaries were full, and consisted of lengthy palabres. The rate of transaction accelerated between December and January -- prices fluctuated, men and goods circulated. By April, the railway was completed backed up while ships waited for cargoes at the ports of Dakar, Kaolak and Rufisque. By May, the activity became more feverish as they hustled to ship the last peanuts before the rains came. By June, the peanut trading season was completed for that year.

(p359) The responsibility for warehousing and transporting peanuts was not well defined. Merchants complained about peanuts lost in transit, which was limited by Tarif Spécial Petite Vitesse n°4 (arachide) to 3%. The railway required merchants to package the peanuts and refused to accept responsibility for the deterioration of peanuts in transit.

(p359) In 1930, peanut shippers organized a Ligue pour la Defense against the railway and demanded that the metropolitan laws concerning transporters' responsibilities be extended to the AOF. [Loi Rabier, 17 March 1905.] The initiative started with the Chamber of Commerce in Saint Louis.

(pp359-360) The railway attributed some of the problem to the slow loading and unloading of peanut shipments, which was the shipper's responsibility.

(p360) Both the Chemin de Feer Dakar-Niger and Chemin de Fer Dakar-Saint Louis railways requested merchants to extend the hours of unloading to include Sundays and holidays, but they resisted, claiming that the slow pace was due to a shortage of workers at the railway stations, and that the railways were responsible for that.

(p360) In 1932, a reform plan offered by the Chamber of Commerce of Kayes, Dakar and Rufisque, called for shippers to handle all loading and unloading. Under this plan, the railway's output would have been entirely under the control of the shippers. Even without this kind of control, peanut shippers were accustomed to using railway transport as a way to infleunce peanut prices -- for example, by incompletely filling railcars at the beginning of the season (to keep local prices low) and by overloading them at the end of the season in order to ship the maximum before rain damaged the remainder.

(p361) Until 1924, shippers and recipients were responsible for loading and unloading bulk freight. Merchants worked out an informal system for apportioning railcars to move bulk freight during times of high volume and railcar scarcity. Lakroum called the system "systeme du tour." [See Paul Defferre, "Rapport sur le Dakar-Saint Louis et le Thiès Niger" (circa 1925), in ANFOM Archives de l'Union Coloniale, carton 407.]

(p362) After 1924, the railroad took over control of freight loading and dispatching in order to maximize wagon loads. According to the Arrêt de la Cour d'Appel de l'AOF (25 May 1925), peanuts were to be loaded in the order that arrived at the station, not according to their ultimate destination.

(p362) The introduction of new types of loading equipment after WWII further damaged the position of the merchants.

Evolution of loading techniques
("L'Evolution des Techniques de Mise en charge")

(p363) Peanuts were loaded in sacks at the point of harvest and transported by donkey (each carrying two sacks of roughly 60kg a piece) to the shipping point. It was easy to estimate the amount of peanuts shipped simply by counting the number of donkeys tart passed the entry to the echelle.

(p363) The change in the way goods were handled. Peanuts were packed in sacks and weighed at the point of production. Then they were carried to the "pointe de traite" and emptied onto open-air platforms at the comptoirs. They were piled high, as much as 1500 tons, by hand using an improvised stairway of peanuts still in their sacks.

(p364) They were rebagged and reweighed for shipment by rail to the docks. Then they were loaded onto railcars by hand under close supervision by European employees of the commercial houses.

(p364) The factors were paid for the weight of peanuts that arrived at the docks, with an allowance of 2% for losses in transit and drying.

(p364) Decauville rails were used alongside the railway sidings in the largest ports to facilitate the handling of the peanuts.

(p365) At the docks, the peanuts were emptied again for further drying and inspection. Finally, they were bagged one more time for loading on ships or transport to the oil factory or deshelling plant.

(p366) Finally, the peanuts were emptied out of their sacks directly into ships' holds.

(p366) Dakar was the first port to get bulk peanut loading equipment. During the 1920s, bulk-loading techniques spread to the other ports.

(p367) All of the bagging and rebagging cost the railroad time and hence money, so they promoted bulk shipments. The peanut exporters resisted.

(p368) The depression in the 1930s forced the government to take action. They passed laws concerning the processing of peanuts prior to shipping, while the railroad made a few small changes to speed shipping. After 1932, they offered special rates for shelled peanuts, even though this made the product more perishable and threatened to reduce railway revenues by reducing the number of tons carried by the railroad. Only Dakar had sufficient dock workers to ensure prompt handling at the quays. Railroad rolling stock had to be used more efficiently to avoid delays, but this required unloading at either Rufisque or Kaolack, which saved 2-3 days off the trip to Dakar.

First positive results occurred only after WWII when special bulk peanut rates were combined with regressive railroad tariffs for freight shipped over long distance.

(p369) Table showing the evolution of peanut shipping rates 1949-1951, with allowances for distance.

(p369) The use of differential tariffs encouraged small industry in the Soudan devoted to preparing peanuts for shipment. Senegalese businessmen protested that preferential tariffs had the effect of increasing the price of peanuts in shells at the coast.

(p370) By 1954, 3/4 of peanuts from the Soudan still traveled in shells, but the 1/4 shelled peanuts traveled the farthest.

(p371) In 1954, the railroad bought 54 self-loading peanut wagons and converted other cars to bulk transport of peanuts.

Trucks ("Le camion, l'arme des negotiants")

(p373) Auto imports to AOF began after WWI and an initial plateau was reached in 1926 with 1214 French made vehicles. Ford was the first to understand the value of a sales network that carried spare parts and technical assistance too.

(p373) "... l'automobile, considerée avant 1914, articulièement à la colonie comme un objet de luxe est regardée maintenant en général comme un instruement de travail." [interview de Chardy, Ancien directeur du Thiès-Niger, président de la Commission de coordination des transports ferroviaires et routiers, "Procés-verbal de la séance du 12 Juillet 1933 de la Commission institute par décision n°987 du GAOF en date du 26 Avril 1933, pour étudier et assurer la coordination des transports ferroviaires et routiers" in Archives de la Chambre de Commerce de Dakar, carton 509.

(p374) The growth in railroad freight tonnage from 1935-1940 and 1947-1958 was due to an increase in long distance freight thanks to the degressive freight rates. It masks the decline in short distance freight which was taken over by trucks.

Most of this happened in the coastal areas of Senegal where trucks that were intended to take peanuts to the railroad began to carry peanuts directly to the coast.

(p375) This table shows the number of trucks in circulation in Senegal and French Soudan from 1932 to 1954:

Year Dakar Senegal Senegal Total Soudan Total
1932 - - 1373- -
1949 752 764 1516 611 2127
1950 734 784 1518 628 2146
1951 891 938 1829 552 2381
1952 691 - -475 -
1953 707 700 1407 514 1921
1954 700 789 1489 597 2086

(p375) In 1952, many older trucks from before WWII were scrapped.

(p376) The increasing number of trucks and the modernization of the truck fleet meant that trucks began to carry a larger percentage of peanut freight.

(p376) European commercial houses protested these changes. They formed a "Ligue de defense contre les chemins de fer"

(p377) Trucks were well-suited for the early part of peanut trade when merchants bought peanuts in small quantities from many producers, prior to assembling them into large shipments. They replaced caravans of donkeys and mules. In Senegal, they allowed merchants to transport their peanuts to the coast, get paid and return in the same day, compared to five days for the railroad. Not only did large merchant houses use them, but so did African merchants.

In 1953 in the Senegalese peanut zone, the cost of road transport rose from 9 francs to 12 francs per ton-kilometer. Railroad rates were 7.75 francs in Senegal and 7.35 francs in Soudan.

(p378) The director of the railroad observed that these regressive tariffs had the effect of passing an increasing part of the railroad's fixed costs on to people in the interior because they were forced to use the railroad for short and medium distance shipments. Coastal people avoided this by using trucks, and the big companies avoided it because they shipped their goods over long distances.

(p379) In 1953, the railroad announced a program to integrate itself with road traffic using containers, refrigerated railcars, tariffs, etc.

Trucks were used as early as 1923 to collect freight and bring it to the railroad, but was not continued in subsequent years.

(p380) Local merchants resisted, especially African merchants. So did some African cheminots. They argued that the decision to bolster the railroad's performance benefited large companies and the railroad at the expense of small entrepreneurs.

PART THREE: Story of a failure (1878-1961)

Chapter Five: Story of a brief period of growth [page 425]

(p425) For a long time, railways, industrialization and growth were associated with each other, but no one asked how they complemented each other. The first questions developed during studies of later industrialization, such as R. W. Fogel on the US, Alexander Gerschenkron, etc.

(p425) P. Bairoch gave a radical analysis by concluding that the diffusion of steam technology in ships and railways was a cause of underdevelopment. [see P. Bairoch, Révolution industrielle et sous-développement (Paris: La Hayes, 1963, 1974).] As long distance transport costs dropped, European goods were able to compete over a wider area, retarding the development of local industry.

(p426) Local economies were limited to producing raw materials, and their exchanges with the developed countries were consequently unequal. This idea was first expressed in 1963, and extended by many people, especially Samir Amin for West Africa.

(p452) At first, all evidence suggests that there was steady if modest growth from 1890-1913. The drop in prices suggested that productivity increased, while the volume of trade increased.

(p453) GRAPH showing changes in leading economic indicators. Railway transport costs dropped, as did the price of peanuts in Dakar. The value and quantity of peanuts exports increased, as did the volume of imports and the number of ton-kilometers of railway use.

(p455) The surge in imports coincided with the completion of the first section of the railway. Thus, the railway was not only a tool for draining produce towards the coast, it also guaranteed the victory of European import goods over local products.

(p455) For a brief period after WWI, prices and the costs of transport became disassociated from the costs of production and exchange.

(p506) Who used the railroad? Petty merchants, migratory peanut farmers, seasonal workers headed for the city. All of these created a seasonal demand for transport.

(p513) The railroad's architectural contribution was a style of construction with raised floor and veranda called "les batiments Paindevoine"

(p514) metal frame, masonry on first floor, brick and metal plate on second floor. Examples: gares de Koulikoro, Kita and Bamako. The gare de Kayes was an exception. Its monumental style imitated a mosque. This style was called "mauresque."

(p518) Over time, population became concentrated in cities at each end of the railroad. After 1950, many small urban centers developed in the peanut districts.

Chapter Six: The railway: creative destroyer? [page 538]

(p541) The construction of a railroad in a region without any form of infrastructure posed two problems: how to get material there, and how to find a workforce. Political and military considerations guided the French response.

(p542) French method combined some of the best techniques used in Europe with a high degree of improvisation. Lakroum says that railroad construction was more of an "art" than an "industry." The result was a unique form of construction that most resembled the form used in the USA: "telescopic construction" that proceeded from a single point towards a vacant interior.

Due to the absence of other means of transport, the distance between the work site and the railhead remained roughly constant at 25-30 kilometers. Material was transported along each working "section" with a Voie Decauville; when the rails were finished, the Decauville was dissassembled and moved to a new location beyond the new railhead.

(p543) Another problem was the limited period available for work each year. The dry season was only six months long (December - June) and work during the rainy season only "filled the hospitals." The failure to coordinate the arrival of construction materials with the dry season was largely responsible for the failure of the first three campaigns (1880- 1883) to achieve results.

(p544) There is a quotation relating how Chemin de Fer Kayes-Niger construction failed in the first years (1880- 1883) because the Senegal River failed to rise and materials could not be delivered in time to Kayes.

(p546) Bridges on the Chemin de Fer Kayes-Niger were trusses constructed of prefabricated metal parts using the "system Eiffel."

(pp547-548) There is a description of the bridge at Galougo, the first large structure undertaken on the railroad. It included a quotation from Gallieni's 1891 book, Deux Campagnes au Soudan Français 1886-1888.

(p549) Foreign workers were killed by disease. Only African construction workers could function in the African climate.

(p553) Quotation states that African workers were capable of all tasks required for railroad construction. In other words, African workers were competent.

(p554) After 1907, the French found it easy to recruit skilled laborers from Senegal. Masons came from urban centers while wood and iron workers came from the Senegal River Valley. The two groups had very different approaches to their work. The urbanized masons worked for salaries in a disciplined manner, while the river people followed a more traditional artisanal approach. They maintained their autonomy by performing piecework. [see Comptes d'Administration du Chemin de Fer Thiès-Kayes, Rapports Trimestriels" (1914-1921)]

(p554) The French had the greatest difficulty recruiting "terrasiers" (earth movers). Local tools were inefficient and the local farmers had no experience with dirt excavation, since they were accustomed to slash and burn farming. Africans readily adapted to picks and shovels, but relied on headloading baskets to move excavated earth. [P. Pelissier, Les Paysans du Senegal: les civilisations agraires du Cayor à la Casamance (Saint-Yriex, 1966), 942, cited in Lakroum, "Chemins de Fer," 564. Note that Lakroum's bibliography says that Pelissier's book has a total of 942 pages, so the page number in the footnote is probably wrong.]

(p555) Earth moving held up construction more than bridge building or anything else. The French chose to increase earth moving productivity by increasing the size of the work force through massive recruitment, rather than by changing techniques. They used three types of forced labor: local prestation labor, prisoners and 2ieme portion.

(p555) The last big work sites were in 1919-1923 between Tambacounda and Kayes. In 1919-1921, only 16 kilometers were completed due to large scale desertion, including 500 of 1300 workers at the end of 1921. The French responded with massive recruitment of both forced and salaried workers in 1922.

(p556) Railroad salaries were good compared to salaries in the private and public sector in 1921. African construction workers received 3 francs per day in 1920 and 4 francs per day in 1921. However peanut agriculture, which was accorded higher priority by the government, competed with the railroad for labor.

(p557) Mechanization of railroad construction sites began in 1927. [Z. Semi-Bi, "La politique coloniale des Travaux Publics en Cote d'Ivoire, 1900-1940" (Thèse de doctorat, Université de Paris VII, 1973), 165-7.] They constructed large buildings for workshops using metal pieces and larger cranes.

The military administered new construction, first under the Service des Travaux Neufs du Chemin de Fer Thiès-Niger and later under the Sapeurs Indigènes.

(p558) Lakroum devoted several pages to the chantier de Sine as an example of mechanization of construction. In December 1932, they used a steam shovel built by Jules Weitz of Lyon.

(p559) Decauvilles were used to remove excavated earth. Each wagon held 1000 liters (1 cubic meter).

(p560) The construction site had cement mixers manufactured by Weitz and Campistron.

(p560) In 1932, the Diamou bridge was rebuilt using poured concrete for the abutments.

(p561) Concrete was used most often at construction sites near urban centers, like the region between Dakar and Thiès.

(p562) Other mechanical tools involved the use of air compressors to power pneumatic tools.

(p562) The machines were not always used effectively. Instead, they were too dispersed along the line.

(p563) Reconstruction of 500km in Mali started at Dinguira in 1930, finished in 1957. Directed by Lt. LaClavere of the colonial artillery. Lakroum gives details of the section between Dinguira and Galougo (pp563-571) which was rebuilt in 1932. Reconstruction was necessary because rails were bent and twisted from derailments, ballast was washed away and ties were broken.

(p564) Surveying was done by the private firm of Entreprise Jarre. Workers were housed in camps, with the rails dividing the Europeans from the Africans.

(p565) October 1932, yellow fever strikes workers on reconstruction. Otherwise, there were few obstacles, as the railroad supplied plenty of railcars to haul dirt and material. However, there were no doctors assigned to the camps.

(p571) Reconstruction was complete as far as Fangala by 1939. Work slowed during WWII due to a shortage of material, and light 20-kg rail had to be relaid on new ballast at one point (although Lakroum doesn't say where).

(p571) After WWII, reconstruction work was carried out by private companies like Sie. des Dragages (Kita-Sebekoro), Sie. Nationale des Travaux Publics (Sebekoro-Kassaro), and Entreprise Ayme-Juillan (Kassaro-Negala).

(p572) In the period from 1947 to 1957, Chemin de Feer Dakar-Niger reconstruction absorbed 31% of all FIDES money spent in AOF. Of that amount, 71.57% went to Chemin de Fer Kayes-Niger reconstruction and the rest went to doubling the track between Rufisque and Thiès.

(p578) There is a quotation concerning black labor on the Chemin de Fer Dakar-Saint Louis in 1900. All locomotives were driven by blacks with black firemen. With few exceptions, they were excellent and careful drivers. [Chemin de fer de Dakar à Saint Louis, "Réponse au questionnaire, rapport de la commission spéciale nommé à Berlin" (1900), 322]

(p579) The original locomotives (Types Kipps, Marchi and Huvenoit) required so many jury-rigged repairs that each developed into a completely unique machine. In the 1980s, children of the first generation of African mechanics all remembered the names of the first locomotives even though they had never seen them. [from 1984 interviews with Chemin de Fer Dakar-Niger cheminots]

(pp578-580) French built locomotives required adaptations to work in local conditions, including modifications to permit the use of wood and peanut shells as fuel, tougher shock absorbers and suspension, etc.

(p580) After 1905, locomotives on the Chemin de Fer Kayes- Niger were assigned to specific stations and the line was operated in sections divided at Toukoto and Bamako, so from the point of view of workers, each belonged to a micro-system. Rivalry developed between Kayes and Bamako.

(p581) This type of organization was more flexible and facilitated the use of special trains and work trains, not regular scheduled service.

(p594) The first lines to convert completely to diesel were the Tambacounda-Bamako and the Thiès-Saint Louis- Linguére, both by 1953.

(p605) Despite the passage of the Code de Travail and the law for the 8-hour work day (passed on 23 April 1919), there were many violations. The typical work day was ten to twelve hours long.

(p606) The length of a day's work varied throughout the years.

Period Hours/week Hours/year Index value
1953-1958 40 2,080 1.00
1919-1952 50 2,600 1.25
Before 1919 72 3,744 1.80

(p606) After 1912, railroad operations were based on the use of a core of regular salaried workers and a floating mass of transient laborers. This system was expedient and continued until WWII.

(p609) Table showing indices of worker productivity.

(p611) After each technical innovation, worker productivity dropped, then gradually rose to a "normal" plateau, establihsed in 1924, the first year of consolidated operations.

(p612) In 1954, the Chambre de Commerce opposed the agreement between the railroad and the union that dieselisation would not result in wholesale layoffs. The railroad announced a hiring freeze in 1954. After 1956, dieselisation was complete, but productivity remained low, so layoffs began.

(p612) In 1957, the Africanisation of the cadres began.

(p614) Railroad workers developed a group consciousnes determined by their association with the railroad, their particular job on the railroad, and the section of the railroad for which they woked. In other words, railroad workers identified themselves as carpenters on the Kayes section, steamfitters on the Thiès section, etc.

(p615) Lakroum notes that social divisions perceived by European ethnologists were not always perceived by the Africans themselves.

(p615) The first railroad workers came from the river regions where they benefitted from experience gained on steam boats.

(p615) At first, Africans took railroad jobs according to their role in traditional society. Forgerons and other artisans became skilled workers, nobles became foremen and slaves became laborers. A survey of the social origins of workers at the Thiès workshops in 1952 showed that this did not change much. However, over time job qualification replaced traditional social class as the determinant of social status. Technical job skills could be passed to offspring more readily than social status.

(p616) This page includes the results of a survey of metal and woodworkers in Thiès in 1952. It shows the diverse ethnic origins of the workers. 834 of 1334 responded (62.5%)

Ethnic Group Number Percentage of work force Note
Wolof 301 36.1% -
Sérére 71 8.5% -
Sérére 372 44.6% from the region of Thiès
Peul, Toucouleur 189 22.6% -
Sarakolé 39 4.7% -
Khassonké 40 4.8% -
Bambara 115 13.8% -
Bambara 383 45.9% region de fleuve et Soudan
Other territories 79 9.5% -

(p616) Lakroum asserts that the mixture of ethnic groups modified social relations. The job hierarchy was superimposed on the tribal hierarchy.

(p617) More figures that show the division of railroad workers by job and by pre-colonial social status. They indicate that social status was repeated in the distribution of positions with the railroad. In other words, nobles were more likely to hold higher positions than former slaves.

Origines Sociales et fonctions dans les Ateliers de Thiès en 1952 [See G. Savonnet, "Evolution Demographique de la Ville de Thiès" in Notes Africaines (October 1952), 122-124]. This table contains the raw numbers from the end note on page 639. It shows the number (in parentheses) and percentage of each social group, out of 82 workers surveyed.

SOCIAL STANDING Craftsmen Workers Laborers
Nobles (5) 33.34% (7) 46.44 (3) 20
Blacksmiths (5) 18.5 (14) 51.9 (8) 29.6
Griots (1) 6.2 (9) 56.2 (6) 37.6
Slaves (1) 5 (7) 35 (12) 60
Others (0) 0 (2) 50 (2) 50%

Noble power was based on pre-colonial strength and status, strengthened by the colonial government's policy of governing through local elites. The artisans's owed their strength to their pre-colonial skills plus experience gained with steam ships. The artisans were the only ones who were capable of performing and transferring these skills, so they became more important over time. This became evident in a social hierarchy based on "metier" or job.

(p618) Despite improved technical skills and material, worker productivity did not improve during the period 1957-61. Lakroum asks if the railroad workers were to blame for this relative reduction in productivity; could they transform a society that was unchanged by technological innovation? Lakroum asserts that cheminots were a small, closed group. Railroad workers lived in a closed world that Lakroum called "micro- société." Over time, the distinctions between technicians (métiers) and bureaucrats (fonctions) became confused.

"Distribution des metiers et des fonctions dans les ateliers de Thiès en 1952" (From Savonnet. The raw numbers appear in the endnote on page 639.]

Ethnic Group Manuel Labor Technicians Office workers Supervisors
(Health & Police)
Wolof, Sérére (194) 38% (44) 46.3% (98) 62.8% (36) 50%
Soudanias&fleuve (261) 51% (44) 46.3% (49) 31.4% (29) 40.3%
Other (56) 11% (7) 7.4% (9) 5.8% (7) 9.7%
TOTAL (511) (95) (156) (72)

(p619) From the above table, it is clear that "Originaires du Sénégal" held a disproportionately large number of administrative (tertiary) positions, while technicians were evenly distributed between coastal and inland groups.

"Niveau de qualification des ouvriers des ateliers de Thiès en fonction de leur origine en 1952" [From G. Savonnet. The raw numbers appear in the end note on page 640.]

Ethnic Group Craftsmen Workers Laborers
Wolof, Sérére (26) 29.5% (248) 45.2% (98) 49.8%
Soudanias et fleuve (44) 50% (257) 46.8% (82) 41.6%
Other (18) 20.5% (44) 8% (17) 8.6%
TOTAL 88 549 197

From the above table, it is clear that nearly 3/4 of skilled manual laborers (maitrise) came from regions other than that of Thiès. This was due in large part to efforts by the French government to create training schools at Toukoto and Bamako, and to employ the graduates of those schools. Lakroum mentions the Loi Lamine Guèye (1951) which offered financial incentives to AOF workers who went to areas away from their homelands. Since the Chemin de Fer Dakar-Niger was always in financial difficulties, it employed only the most qualified personnel from other territories, as shown in column 1.

(p619) The intersection of traditional social structure and railroad employment led to a new basis for improvement of social status derived from technical competence.

(p620) Outside of work, traditional social status still took precedence over professional categories. However, a social hierarchy based on one's job slowly began to replace the social hierarchy based on ethnic origin in the 1950s.

(p620) Lakroum interviewed a number of cheminots (6, at Toukoto and Bamako. There names are found in note 108 on page 640.) from "le troisième géneration" in 1984. Most entered railroad service in the 1940s. They were affected by the changes incurred through the introduction of diesels, but they also inherited a sense of professionalism from the second generation (post WWI) of railroad workers. They tended to define themselves in terms of the technical training they had received rather than their hereditary position in African society.

(p621) Railroad workers interviewed in 1984 spoke of their history in terms of a few major events: the stages of construction of the Chemin de Fer Kayes-Niger, the general strike of 1947, the separation in 1960 and their expulsion from Senegal. They all told the story in the same terms, suggesting the existence of a collective memory of the history of the railroad. They preferred to let one man serve as a spokesman for the entire group.

They only showed their individuality when it came to describing their particular profession. However, they all followed the same form in their descriptions, describing masters and tasks. Some described their memories in terms of exceptional workers that they had known; and others in terms of exceptional jobs that they had performed, such as the fabrication of the first piston in AOF.

Another common element was their attitude towards tools. There was idea of the tool having its own life and name, such as "Dieu donne" for the large steam hammer at the Thiès workshop (locomotives all had names). This is summarized as "the tool responds, the masters initiate and the artisans consecrate." This idea was part of pre-colonial attitudes of the "fondeurs, forgerons and bijoutiers" who incorporated it in the legends they used to transmit their skills to apprentices.

(p622) To conclude the material on p621, cheminots retained traditional artisanal attitudes towards their work, as demonstrated in their descriptions during the 1984 interviews.

(p622) Over time, technical skills lost their social value. In 1955, 30% of the 42,503 people in Thiès were cheminots (railroad Workers). Thanks to their workshops, cooperatives and industries, Thiès was quasi-independent, depending on Dakar only for electricity, which arrived by 30,000 volt line. They were organized in four independant sections (Traction, Exploitation, etc) and sub-sections created in 1947 to repair the new diesels.

Composition des Ateliers de Reparations Ferroviaires de Thiès en 1952

Ateliers African employees European employees
Forge & Foundry 104 3
Matérial roulant, section bois 150 0
Matérial roulant, section fer 250 12
Traction (Vapeur) 600 16
Traction (Diesel) ~80 ~20
Voie et Batiments 100 3
Electricité 50 3
TOTAL 1334 57

(p623) At the same time that the new sections were created for diesel locomotives, the number of Europeans was reduced. However, for a brief period, technicians from the SNCF were sent to AOF to teach diesel skills. Nearly 3/4 of the Europeans in the above table were young, recruited after the war.

(p623) According to the 1984 interviews, they were resented by the older railroad workers, who still associated both diesels and French technicians with the effort to break strikes by African workers, even though they realized that the technicians and diesels arrived before the 1947 strike began. The Europeans remained isolated in the Cite Ballabey where they became objects of African suspicion.

Before the war, there was limited, yet regular contact between African and European cheminots, using a limited number of familiar expressions and gestures. This was altered following the war by the influx of new Europeans and the demands of dieselisation. Africans interviewed in 1984 said that the new Europeans were not of the same calibre as the pre-war Europeans, and that their salaries became substandard (salaires de misère).

Effect of the introduction of diesel engines on labor:

  1. The work became faster since diesels required much less preparation than steam engines.
  2. Diesels required less physical effort and technical know-how to control.
  3. Diesels had the most disruptive impact in the workshops. They threatened several of the professions including carpenters and boilermakers, since rail cars could no longer be made of wood and diesels did not require boilers.

(p624) The new diesels required special maintainance and adjustments that were performed by Europeans. This directly threatened many of the skilled African workers, and even though an attempt was made to retrain some of them at the school opened in Toukoto in 1952, some Africans had to be reassigned to less skilled jobs. As evidence, Lakroum stated that in 1894, Africans held 20.4% of qualified positions on the Chemin de Fer Dakar- Saint Louis, but by 1951, they held only 11.3% in the Thiès workshops, the principle technical center of the rairoad.

Overall, after the railroad consolidation of 1924, human skill became less important to the railroad because the French began to use to new techniques to solve the problems of productivity and economy.

(p625) A comparison of salaries over the period 1919 to 1939 shows that railroad engineers (chefs de train) earned more than mechanics after 1926, writers earned more than tracklayers after 1925 and European station masters earned more than skilled workers after 1920.

(p626) By the 1920s, office workers' salaries rose to the same level as that of skilled workers, and in subsequent years became greater. All of this is evidence of the gradual devaluation of mechanical skills in favor of white-collar skills. This was an adaptation of attitudes that were prevalent at all levels of colonial administration. Function became more important than skill and power became more important than competence.

(p627) After the Loi Cadre in the late 1950s, local autonomy increased and the railroad, as a federal institution, became an impediment to growing decentralization. The uncertainty over the future of centralization in AOF affected the railroad as well.

In September 1958, Alioune Sissoko, a member of the Soudanese government, demanded that each territory be allowed to place a representative on the railroad's Conseil d'Administration.

The Soudanese Minister of Commerce and Transportation, Hamacire N'Doure, suggested that additional railroad repair facilities be constructed in Bamako or Kayes.

A spokesman for the cheminots hesitantly accepted the possibility of divided operations, but demanded that the financial administration remain centralized with the words, "nous sommes pour l'unité africaine, nous ne suivrons pas, nous ne sommes partisans de la Balkanization."

(p628) In 1958, Hamacire N'Doure, Soudanese Minister of Commerce and Transportation, suggested the transfer of some repair operations from Thiès to Bamako or Kayes.

(p629) Spokesmen for the cheminot union accepted the possibility of a divided operations administration, but insisted on their support for a centralized financial administration. [

The "Arrêté du 31 Mars 1959" created separate railroad administrations. Each branch of the Chemin de Fer de l'AOF was subject to local authority, in imitation of the Conakry-Niger, which came under Guinean control in 1958.

Epilogue [page 629]

(p630) Since Chemin de Fer Dakar-Niger operations were based on joint administration, there was no simple way to restructure its administration. The declaration of the Federation of Mali in January 1959 solved the problem temporarily, and the "Ordonnance du 29 Juin 1959" created the "Regie des Chemins de Fer du Mali." [ Journal officiel du Mali (18 July 1959), 154.]

Senegalese cheminots and authorities feared the power implicit in the presence of a large contingent of organized Soudanese cheminots at Thiès. The Soudanese cheminots demanded control over hiring and promotions at the Thiès workshops.

The break came in 1960 when Modibo Keita tried to seize total power in the Mali Federation. The archival documents are not available to study the financial and technical aspects of the breakup.

(pp631-632) This is a document written by former railroad workers in 1984, describing their expulsion from Senegal at the breakup of the Mali Federation. It describes the sudden expulsion, terrible conditions on the trains, the help obtained from Malians in Kayes and Bamako and the eventual reassignemnt of Soudanese cheminots to jobs with SONAREM, CMTR, administration, finance, navigation, etc. It also mentions that griots sing songs about the exodus.

(p643) The history of the railway lay somewhere between the realm of what actually happened and what was planned to happen. Much more is known about the plans and dreams of the French. They produced an enormous pile of documents and studies that hide the ways that local realities provided resistance. In the end, resistance won because the utility and efficiency of the innovations imported by the French were offset by other limiting factors.

The railroad offered certain advantages - speed, regularity and increased capacity - but imposed certain limitations as well, namely the rigidity of the tariff system, a fixed itinerary and the necessarily large size of the loads that it carried.

The railroad was an extension of French intent, not a response to local needs. But is that enough to explain the railway's failure? Analysis of the failure requires a study over a long period of time.

(p644) The process of turning the railway from a military line to a profitable commercial operation aggravated numerous forms of conflict between importers and local traders, the terms of trade and the rules governing commerical profits.

(p644) The railroad was built as a result of European imperialism in the late 1870s as they sought new regions to employ industrial techniques developed in Europe. They were inspired by the success of similar projects in new countries like the USA and Russia. Justification was phrased in Saint-Simonian language that included the desire to link continents and unite the world. At the time Africa was still largely unknown, so it offered the engineers and planners of Europe a new place where they could experiment with their machines and theories.

The Dakar-Niger was a byproduct of the Trans-Saharan railroad plan and military expansionism. Although these two forces managed to find common support in the parliament, they were not united on the ground in Africa. Military officers chose to penetrate the African interior in the time-honored fashion of follwoing navigable water. Thus, the railway became a simple device for united stretches of navigable rivers.

The construction of railways to connect navigable rivers created a railway that ignored the most important fact about West African trade: for three centuries prior to the railroad, internal trade was conducted by networks of human porters and traders.

(p645) Pre-colonial trade responded to conditions of transport rather than the value of goods. Value was added by the displacement of goods and their packaging for travel, so commerce and transport were equivalent activities. Value was determined by the route taken, not by the prices set in permanent markets. Profits resulted from the creation of temporary contact between zones of production.

The new commercial system that accompanied the railway disrupted a fragile and delicate system. Lakroum points out that innovation alone did not disrupt the African system - automobiles would fit in easily 50 years later - but the requirements of rail transport were incompatible with the local system of commerce. But this was not evident at first because the railroad served mainly to aid military conquest of the interior.

However, the plans for the economic exploitation of the AOF included direct control over markets and prices. European merchants on the coast decided to establish permanent trading posts in the interior, and bypass the existing network of African and Lebanese traders.

(pp645-646) Why did merchants who were already successful on the coast attempt to bypass the African system in order to enter territories that were not even completely pacified at the time? The first reason was the complete ignorance of the Europeans for the system they were trying to bypass. The completion of the original two sections of the railway made little change in commercial operations. As long as they were not linked, commercial innovations were limited to their respective regions. At the same time, French private firms withdrew from the management of colonial railways because they were difficult to make profitable and they became subject to increasing administrative restrictions.

(p646) A second reason was the creation in 1905 of the Gouvernement Général de l'AOF which had a much closer relationship with local merchants than did the Ministére des Colonnies. Not only did it complete the middle section of the railroad, but it inaugurated a program of agricultural colonization along the railroad right-of-way. It certainly revived the peanut trade, but by providing a single link from interior to coast, the railway threatened the local cost differentials that created the profits that motivated local commerce.

After 1925, the conflict between French and local traders intensified. Merchant opposiiton was no doubt justified, since the railway's construction was flawed and it had to be almost entirely rebuilt. French commerical interests, coordinated by the local Chambres of Commerce and acting through the Union Coloniale during the 1930s, lodged numerous complaints about the railway -- tariffs, losses during transport, delays in peanut shipments -- that showed how fragile the commercial network was.

(p647) To understand the logic of the opposition between European merchants and the railway administration, one must decipher the rules of transport, the basis of value and the principles of exchange in the colonial economy. After that, one must understand the relationships between the many individuals who were involved. In the process, it is important not to become trapped in the use of preexisting categories.

Lakroum's method is "empirical and inductive." It consisted of identifying all of the different elements and steps in this ongoing quarrel. However, Lakroum admitted that even after identifying all of the pieces, it was impossible to put them all together. The pieces do not fit together. They coexisted, but developments in African trading activity obscured the development of other new activities that developed thanks to the reduction in transport costs created by the railway.

The model of a dual economy -- modern and traditional -- is not appropriate. If one sysem survived, it did so on the ruins of the other one. The only way to express this interaction is through the use of the idea of "networks" which shows the confrontation that existed at every level between two different organizational systems.

The French system channeled trade from a diffuse and dense network into a hierarchical system.

(p648) In lieu of a network of many small centers, the French system selected some to serve as centers in a hierarchical system. The railway was a centralizing force that tried to create a single integrated market out of an economic system that was comprised of many small markets.

This was not competition between two modes or production nor progress and conservatism, but rather competition between two forms of economic growth. The European industrial development model concentrated functions in fewer locations to achieve efficiency. The African model repeated many operations and transactions in different places over a large area. The economies of scale and efficiency that resulted from the railway supported the creation of new middlemen in the African system, so the reduction in costs never materialized. Eventually, even railway operations began to follow the same pattern by absorbing price reductions to support more cheminots and other intermediaries. As a result, there was no improvement in worker productivity.

The supremacy of commerce, the principles of exchange and the growth of the system did not produce significant social or economic changes. The construction of railways failed to stimulate the creation of the values of industrial development.

Statistical Annexes [page 712]

(p713) Traffic on the Chemin de Fer Kayes-Niger 1897- 1922 (passengers and freight)

(p714) Traffic on the Chemin de Fer Dakar-Saint Louis 1883-1932 (passengers and freight)

(p715) Traffic on the Chemin de Fer Kayes-Niger 1923- 1958 (passengers/km, ton/km, locomotives)

(pp716-717) Traffic Chemin de Fer Dakar-Saint Louis and Chemin de Fer Kayes-Niger 1883-1958 (pass/km, ton/km, ratio pass/freight

(p718) Chemin de Fer Kayes-Niger 1897-1922 (tons of freight in each direction, total)

(p719) Chemin de Fer Kayes-Niger 1913-1922 (freight carried to Bamako, by category, based on reports in ANM 1 K 80, fonds recents)

Inbound freight consisted of "vivres, houille, tissus, petrole/huile, armes explosifs, sel marin, metaux/quincailerie, materiaux de construction et divers." The biggest item was "sel marin," with "materiaux de construction" a distant second, "vivres" third, and "metaux/quincailerie" fourth.

(pp720-721) Chemin de Fer Kayes-Niger 1913-1922 (freight carried to Dakar, by category, based on reports in ANM 1 K 80, fonds recents)

Outbound freight consisted of "bois, riz, mil, arachide, karite, coton, construct., kolas, epice, peaux, kapok, gomme, sisal, graines, bois du pays, caoutchouc, et divers." The biggest quantities were for items like "bois de chauffe, arachide, construc, mil, riz, karite, bois du pays."

(p746) Chemin de Fer Kayes-Niger 1897-1922 (receipts for pass. & freight, expenses)

(p748) Chemin de Fer Dakar-Niger 1923-1958 (receipts, expenses, salairies in 1914 francs)

(p762) Chemin de Fer Dakar-Niger 1923-1958 (receipts for pass. & freight)

(p763) Chemin de Fer Dakar-Niger 1923-1958 (salaries, material, other expenses, reserves and investments)

(pp780-781) Chemin de Fer Dakar-Saint Louis, Chemin de Fer Kayes-Niger, Chemin de Fer Thiès-Niger, Chemin de Fer Dakar-Niger 1883-1958 (locomotive fleet)

(p785) Chemin de Fer Dakar-Saint Louis 1885-1913, (receipts, expenses, cost of transport in 1914 francs)

(p786) Chemin de Fer Dakar-Niger 1920-1958 (receipts, expenses, cost of transport in 1914 francs)

(pp787-788) Chemin de Fer Dakar-Saint Louis, Chemin de Fer Kayes-Niger, and Chemin de Fer Dakar-Niger 1885-1956 (cost of transport in constant francs)

(pp810-865) "Récensement des Comptoirs et des Entreprises Commerciales" Annuaire des Entreprises Coloniales
Click here for a spreadsheet showing the number of comptoirs at selected cities, and here for a spreadsheet showing the number of commercial enterprises at the same selected cities.