jwsr-v7n2 265 Timbuktu: A Lesson in Underdevelopment* Riccardo Pelizzo Riccardo Pelizzo Department of Political Science Th e Johns Hopkins University 338 Mergenthaler Hall Baltimore, MD 21218 rpelizzo@earthlink.net http://www.jhu.edu/~polysci/ journal of world-systems research, vii, 2, fall 2001, 265–283 http://jwsr.ucr.edu issn 1076-156x © 2001 Riccardo Pelizzo INTRODUCTION The present paper is a study of development which diff ers from other such studies in two major respects. First, rather than analyzing what factors have promoted the development of a previously underdeveloped area, I will try to assess what factors may lead to the decline of a previously highly developed area. In doing so, I will try to show that development is fragile and precarious, because it depends on both domestic/institutional and international/economic factors and conditions. Second, this study diff ers from most other studies of development because instead of focusing on modern events, I will devote my attention to a rather small sub-region of Western Africa between 1526, when Leo Africanus1 visited Timbuktu and described it as one of the richest cities he had seen in his travels, and 1830, when Rene Caillié noted that Timbuktu was “nothing but a mass of ill- looking houses, built of earth” (Caillié 1830:49). Th ere are several reasons why I decided to utilize Timbuktu’s history as a case study: I focus on Timbuktu because its decay can be used to test competing explanatory models and, most importantly, because it can be used to show that Timbuktu’s decline was neither an accident nor the result of inevitable natural * Th e author would like to thank Harry Gould, Paulina Ochoa Espejo and Salvatore J. Babones, as well as the anonymous referee, who greatly improved my paper with their insightful comments. 1. Th e name of Leo Africanus was Al-Hassan Ibn-Mohammed Al-Wesaz Al-Fasi. He was then baptised as Giovanni Leoni, and he is known generally as Leo Africanus. His name was translated into English as John Leo. Th e purpose of the present paper is to inves- tigate Timbuktu’s economic decline in the three centuries elapsed between 1526, when Leo Africanus reached the Mysterious City, and 1830, when the fi rst European explorers arrived in Timbuktu. It is argued that Timbuktu’s decline was neither an accident nor the result of inevitable natural conditions. Timbuktu’s decay was the product of historical and social forces. Specifi cally, it is argued that Timbuktu lost power and prestige because its market decayed. However, it is also suggested that no single factor can account individually for this event. Th e crisis of Timbuktu’s market was provoked by the interaction of two factors: fi rst, the gen- eral decline of Mediterranean trade resulting from the emergence of the new Trans-Oceanic trade, and the crisis of the system’s component parts at the individual level, and their inability to function as a system; and second, the insti- tutional decay which followed the fall of the Songhai empire, which made Trans-Saharan trade particularly risky and, henceforth, eco- nomically ineffi cient. Building on this, the paper ends on a methodological note as it underlines the theoretical inability of monocausal expla- nations to capture the inherent complexity of social phenomena. abstract http://www.jhu.edu/~polysci/ http://jwsr.ucr.edu mailto:rpelizzo@earthlink.net Riccardo Pelizzo266 Timbuktu: A Lesson in Underdevelopment 267 conditions. Timbuktu’s decay was the product of clear historical (social) forces and as such it needs to be analyzed. Specifi cally, I will argue that Timbuktu decayed when it lost, with the decline of its commercial activities, the source of its wealth. In the present paper, it will be suggested that this outcome was produced by the interaction of two factors: the decline of the Mediterranean trade system on which Timbuktu’s trade was dependent, and the demise of the effi cient economic organization and institutions following the fall of the Songhai empire. Th e paper is divided into fi ve sections. Th e fi rst section provides a theoretical justifi cation for my research, i.e. why it is worth investigating the determinants of Timbuktu’s economic decay. Th e second section places Timbuktu’s economic decay in historical perspective by analyzing the memoirs of Arab and European travelers who have visited Timbuktu. Th is discussion allows me to localize the timing of Timbuktu’s decline. Building upon this discussion, the third sec- tion of the paper addresses the determinants of Timbuktu’s wealth. It will be argued that Timbuktu’s wealth was mainly generated by the intense activity of its market, where Western Sudanese goods were exchanged for the commodities brought from North Africa and also for the goods brought from the near-forest and the forest-land in the South. In the fourth section, I will argue that the eco- nomic decay of Timbuktu is associated with, and I believe caused by, the decay of its market, which was promoted by the interaction of both international/ economic and domestic/institutional factors. In the fi fth and fi nal section, I will assess what lessons can be learned from the experience of Timbuktu. THE MYSTERIOUS TIMBUKTU AND ITS MYSTERIOUS DECAY Timbuktu has often been referred to as to the mysterious city, as its location, name2 and origins3 have long been unknown to the European powers. Yet, none 2. In a letter written by Antoine Malfante and addressed to Giovanni Mariono at Genoa in 1447, Timbuktu is called Tambet or Th ambet. Th ambet is also the name used by Ca Da Mosto in his “Voyages of Cadamosto, who also adopts the name Tambucutu and Tanbutu. Gerald Roe Crone, in his introduction to the 1937 English edition of Th e Voyages of Cadamosto (Crone, 1937: xii) noted that in a Catalan Atlas made for Charles V by Abraham Cresques in 1375 Timbuktu was referred to as Tenbuch. A few years later, Diogo Gomes called it Tambucutu. When Pory translated Leo Africanus’s Th e History and Description of Africa and of the Notable Th ing Th erein Contained in 1600 (Africanus 1896), the name of the city was translated as Tombuto. Yet, Réné Caillié in his Travels trough Central Africa to Timbuctoo and Across the Great Desert to Morocco, (Caillie 1830) prefers the word Timbuctoo, and Heinrich Barth used alternatively the diction Timbúktu and Timbu’ktu in his Travels and Discoveries in North and Central Africa (Barth 1965 [1857] ). Th e best example of this confusio linguarum is represented by William Allen’ s Narrative of the Expedition to the Niger (London, Bentley, 1848) in which the author used several variant names. In fact, Allen used the diction Tomboktu in the map prepared for his book, the diction Timbuktuh in the Index of his book (p.1), the diction Tombutto in the discussion of Leo Africanus’s work (p.6), and both Timbaktú (p.8) and Timbaktuh (pp. 16-17) in the discussion of the works of Lang and Caillié. Interestingly enough, the name of Timbuktu was a mystery not only for the Western explorers and travelers, but it was a mystery also for some of the native populations. In the third footnote on page 6 of his book, Allen makes the following remark “it is curious that one of our interpreters, Mohamed Lamina, said he knew two large cities, Timbaktu and Tombuttu, and the latter was very much larger than the other”. Nowadays, the mysterious city is called Timbuktu in English, Tombouctou in French and Timbuctu in Italian. 3. Th e mystery surrounding the proper spelling of the name of the city would not be of particular interest, if it were not associated with the mystery concerning the origins of name and, in the end, of Timbuktu itself. What are the origins of the name Timbuktu? Even in this respect, our knowledge is far from being unshakably certain. In fact, if we reject, as Cissoko suggests, Miner’s argument that the meaning of Timbuktu“ is purpoted to be ‘the woman of the big navel’ ” (Miner 1965:4-5) we are then left with three major etymological and historical interpretations of Timbuktu’s name. Leo Africanus believed that the name of the city and the city itself were of Songhai origins as he argued that “the name Tombuto (was) imposed on the kingdom from the name of a town so called which (they say) king Mansa Suleiman founded in the yeere of the Hegeira 610” (1896:824). If this were true, the story told by Mohamed Lamina would not be as surprising as it sounded to Allen’ s ears. Th is said, according to Leo Africanus the name Timbuktu is made up of two components: tin (wall) and buktu and it means, henceforth, the wall of Buktu. Unfortunately, Leo Africanus does not resolve the most interesting aspect of the puzzle, i.e. who or rather what is this buktu. Th e belief that Timbuktu and its name had a Songhai origin was also shared by Heinrich Barth, who suggested that “the original form of the name was the Songhai form Túmbutu, from whence the Imóshagh made Tumbýtku, which was afterwards changed by the Arabs into Tombuktu” (1965[1857]: 284). According to Barth “the town was probably so called, because it was built originally in a hollow or cavity in the sand hills. Túmbuto means hole or womb in the Songhai language: if it were a Temáshight word, it would be written Tinbuktu. Th e name is generally interpreted by Europeans well of Buktu, but tin has nothing to do with well” (Barth 1965: 284-285 footnote). In addition to this interpretive line, that identifi es Timbuktu as a Songhai word, Cissoko (1996) recalls that some scholars have suggested that the name Timbuktu has a Berber origin and was imposed upon the city by its Tuareg founders. If this interpretation is believed, then the word Timbuktu is made up of two parts: tim feminine form of In, which means “place of ”and bouctou, which is a contraction of the Arab word nekba (small dune). Timbuktu would then mean “place covered by small dunes” (Cissoko 1996:19). A third interpretive line holds that Timbuktu was founded by the Tuareg as their summer camp. Th is interpretation is suggested by Es-Sadi in his Tarikh Es-Sudan, where he wrote that “ in the beginning it was there that travelers arriving by land and water met. Th ey made it the depot for their utensils and grain. Soon this place became a cross-roads of travelers who passed back and forth through it. Th ey entrusted their property to a slave called Timbuctoo, word that, in the language of those countries means the old” (Es-Sadi 1981:35-36). It is interesting to note, as Hodes noted in his 1898 translation of Tariks Es-Sudan that “this etymology would indicate that the Tuareg would not have designed this town with a name belonging to their own language” (Es-Sadi 1981:36). Modern historiography has suggested that the second and the third interpretations should be preferred to the fi rst one, because there seems to be archeological evidence of Timbuktu’s Tuareg origins ( July, 1970). Yet, this evidence does not pro- vide suffi cient evidence to decide whether the second interpretation is to be preferred to the third or whether our preferences should be ranked the other way around. Riccardo Pelizzo268 Timbuktu: A Lesson in Underdevelopment 269 of these mysteries is as mysterious, interesting and intriguing—for the social sci- entist—as the fall of Timbuktu, i.e. as the process that transformed Timbuktu from one of the wealthiest cities of the Middle Age into the dusty city encoun- tered by European explorers in the early 19th century. Why did Timbuktu decay? Th e literature has emphasized the role of three main factors. Th e fi rst factor is represented by the transformation of the cli- matic conditions that made Trans-Saharan trade more diffi cult and, hence, had a negative impact on Timbuktu’s commercial activities.4 Th e exhaustion of the extractive capacity of the goldmines in the region also had a negative impact on Timbuktu’s trade, as gold trade represented a considerable portion of Timbuktu’s commercial transaction.5 Th e exhaustion of the extractive capacity represents the second factor harming Timbuktu’s economy. Th e third contributing factor is the collapse of the third major empire in the region, i.e. the Songhai empire, which was defeated by the Moroccans in 1591. Important as these factors may have been, they do not provide, however, a satisfactory explanation of Timbuktu’s decay. Th e climatic explanation is not entirely satisfactory, because, in spite of climatic changes travelers and explorers were still crossing the Sahara with some ease, as shown by the successful expedi- tion of the Moroccans in 1591. Similarly, the crisis of Timbuktu is not adequately explained by the exhaustion of the extractive capacity in the region for two sepa- rate though related reasons: Timbuktu’s economy was mostly commercial and not industrial (or productive) and gold was only one of the various commodities exchanged on the Timbuktu’s market. In this respect, it is well to recall that salt was as an important as gold for Timbuktu’s trade, and very important were also Tortoise shell, ivory and ostrich (Davidson 1991).6 Finally the decay of Timbuktu cannot even be conceived as a necessary or inevitable consequence of the fall of the Songhai empire, because Timbuktu managed to survive successfully not only the fall of the Ghana empire, which created the conditions for the rise of the Mali empire, but also the fall of the Mali empire itself. Th is second event is of particu- lar importance, because in the fi rst years of the Songhai empire, Timbuktu suf- fered persecutions at the hands of the Songhai empire’s builder, Sunny Ali.7 And yet, neither the fall of two empires nor Sunny Ali’s persecutions represented a serious threat for Timbuktu. Timbuktu’s previous ability to survive instability and related diffi culties makes it rather unclear why the fall of the Songhai empire would “inevitably” lead to Timbuktu’s decay. In sum, what has been said so far suggests that the wealth of Timbuktu was certainly not fostered by any of the above mentioned factors nor by their combination, but it also suggests that their loss does not entirely account for Timbuktu’s decay. Th e purpose of the present paper is to integrate the previous explanations by suggesting that Timbuktu’s decline was, to a large extent, the consequence of the interaction of domestic/institutional and international/ economic factors. TR AVELS TO TIMBUKTU: THE STOR IES OF THE AR AB AND EUROPEAN EXPLORERS We know that after its foundation Timbuktu experienced major economic development, and it remained a major economic center for about three centu- ries, i.e. from the 13th up to the mid of the 16th century. We know this with some certainty because after Ibn Battuta wrote about the wealth of Timbuktu (that he had visited in 1355) during one of his travels, the wealth, the power and the richness of Timbuktu were mentioned also by Cadamosto and, later on, by Leo Africanus who reached the Mysterious City in 1526. In that year, Timbuktu was so developed that Leo Africanus could write “here are great stores of doctors, judges, priests, and other learned men, that are bountyfully maintained at the kings cost and charges. And hither are brought divers manuscripts or written bookes out of Barbarie, which are sold for more money than any other merchan- dize” (1896:825). As Leo Africanus noted, and several contemporary historians have repeated, Timbuktu was not just a very wealthy and highly cultured city. Timbuktu was also militarily strong. Leo Africanus is our source of informa- tion, even in this respect, with his observation that the king of Timbuktu “hath alwaies three thousands of horsemen, and a great number of footmen that shoot poysoned arrows, attending upon him” (1896:825). After this we have a black out of roughly three centuries, which ended in the 1820’s with the expedition of Major Lang, followed by Caillié’s voyage in 1828, by that of Barth in 1853–54 and that of Haywood whose memoires were pub- 4. Th at the climatic conditions in Western Sudan underwent a profound change has recently been suggested by James Curry (1999:160-161). Note that there is no unani- mous agreement on the point. For example Davidson (1991:72), argued that the climatic conditions in the region remained fairly stable over time and that climatic changes were rather negligible. 5. Th is point, which had been originally made by Oskar Lenz (1886:153), was more recently proposed by Mauny (1954). 6. July (1970:59) also emphasizes the importance of the spices, copper and slaves trade. 7. On this aspect, see at least July (1970:56, 59). Riccardo Pelizzo270 Timbuktu: A Lesson in Underdevelopment 271 lished in 1912. Th ese accounts tell us of some encounters between the European explorers and the Mysterious City, encounters characterized by satisfaction and surprise. Th e account given by Caillié deserves to be quoted at some length: “at length, we arrived safely at Timbuctoo, just as the sun was touching the horizon. I now saw this capital of the Sudan, to reach which had so long been the object of my wishes. On entering this mysterious city, which is an object of curiosity and research to the civilized nations of Europe, I experienced an indescribable satis- faction. I never before felt a similar emotion and my transport was extreme…I looked around and found that the sight before me, did not answer my expec- tations. I had formed a totally diff erent idea of the grandeur and wealth of Timbuctoo. Th e city presented, at fi rst view, nothing but a mass of ill-looking houses, built of earth” (1830:49). Th e day after Mister Caillié, after paying the due respects to his host, took a tour of the city. His impressions were not much better than the one he had received “at fi rst sight.” In his tour of the city, he noted, “I found it neither so large nor so populous as I had expected. Its commerce is not so considerable as fame reported (1830:50). At the time of Caillié’s visit, Timbuktu had “at most about ten and twelve thousands inhabitants” (1830:56). Timbuktu was not even the shadow of its past glory. In sum, although we still ignore the causes of Timbuktu’s decay, we know that it must be temporally local- ized between the visit of Leo Africanus and the visit of Caillié. Th ese three cen- turies are the temporal ort, the locus, the place in which Timbuktu’s decay occurs. Yet, the temporal localization of the crisis does not provide an explanation and the puzzle for the social scientist is, at this point, to discover what were the determinants of this collapse and to assess whether and to what extent scholarly hypotheses and theories help provide an explanation. THE R ISE OF TIMBUKTU In order to understand what factors are responsible for the decay of Timbuktu, it is necessary to understand what was the basis on which Timbuktu’s wealth, power and prestige hinged. In a fi rst approximation, it is possible to say that Timbuktu’s wealth and prestige resulted from its being an important trade center. Timbuktu was the market where the Western Sudanese goods could be traded for the goods brought from Northern Africa and it was at the same time the market where Western Sudanese commodities were traded with the goods coming from the near-forest or forest-land in the South. Timbuktu owned its status to its commercial activities.8 Th e wealth generated in Timbuktu, as it was previously recalled, created the conditions for the existence and development of a bourgeois culture and a cos- mopolitan population. Mansa9 Musa developed an important cultural center in Timbuktu, which attracted a very large number of scholars and men of religion, and was considered as important as some of the most renowned European uni- versities. Th is is consistent with Leo Africanus’s remarks on the large market of books and manuscripts in Timbuktu that were previously quoted. However, knowing that the wealth, the power and the prestige of Timbuktu resulted from its commercial activities, still does not tell us what factors favored the development of Timbuktu into a prominent trade center. No mono-causal explanation can be provided in this respect, since several factors were involved in making Timbuktu develop. Geographic factors certainly played a crucial role in development of Timbuktu’s market. Timbuktu was strategically located between the Southern boundary of the Sahara desert, which could be reached using the Trans-Saharan trails, and the northernmost part of the Niger river, which allowed merchants from the South to bring gold and their other commodities to the market. Second, because of its strategic location Timbuktu was not only the closest market for the Southern goods, but it was also the closest market for the salt extracted in Taghaza and Toudení. Finally, because of its strategic location Timbuktu could entertain dense commercial exchanges with the most developed cities in the region, Djenné and Gao. As important as location may be for the effi cient functioning of an economy (Krugman 1997:37–58), the location of Timbuktu was not and could not pos- sibly have simultaneously been a necessary and a suffi cient condition for the fl ourishing of its economic activities. Two other factors were required for the development of the commercial activities: fi rst, the demand for the goods and the commodities produced in Sudan and in the forest-land in the South; and second, the creation of what North and Th omas (1973) call an effi cient economic organization. In the fi rst respect, historical studies have shown that there was a large demand for the Western Sudanese goods, especially gold, ivory and tortoise shell (Davidson 1991:71; July 1970). Western Sudanese goods were in demand in Spain and Morocco in the West as well as in China in the East. Th e best analysis of the development of Western African trade was elaborated by Davidson, who argued that “the expansion of West African trade dates back to the ninth century. Started by Muslim demand in North Africa and Near east, it increased steadily 8. On the point see, Curtin et al. (1978:93) 9. Mansa is generally translated as King, see e.g. July (1970:56). However, a more appropriate translation would be Sultan. Riccardo Pelizzo272 Timbuktu: A Lesson in Underdevelopment 273 as the cities of Southern Europe embarked on their own expansion after the XII century. West African gold became a staple export to Europe and without it there would have been no general use of gold as medium of exchange in high medieval times” (Davidson 1991:76). Neo-institutional economists have suggested that the creation of an effi cient economic organization is conditio sine qua non of economic development. Th e creation of an effi cient economic organization, in turn, requires the creation of the proper economic incentives for economic actors. Th e effi cacy of these eco- nomic incentives is due to the fact that they “channel individual economic eff ort into activities that bring the private rate of return close to the social rate of return” (North and Th omas 1973:1). Property rights are among those institutions that make acceptable to risk private resources for social gains. In the Trans-Saharan trade, the enforcement of property rights reduced the costs of commerce and increased its extent. Th e only major departure between the Western Sudanese history and the analysis of North and Th omas (1973) is represented by who enforces the property rights. Th is task, which according to North and Th omas (1973) should be performed by the state, in the case of Trans-Saharan trade was performed by the various empires10 that ruled in the area and by the Islam. Several travelers, including, Ibn Battuta and Leo Africanus were greatly impressed by the relative safety of crossing the desert. Th e interaction of these three factors favored the economic development11 of Timbuktu and of its neigh- boring cities, including those that are now lost or forgotten.12 It is interesting to note that economic development in Timbuktu paved the way to the emergence of a commercial bourgeois class, of cultural institutions and the accumulation of both capital and social capital. Th ese transformations, in turn, furthered the process of development in the region. Yet, it is even more interesting to note that none of these socioeconomic transformations can be invoked to explain the decay of Timbuktu. Nonetheless, the wealthy and cul- tured Timbuktu decayed in spite of all its resources. Th e purpose of the next sec- tion is to discuss some of the factors, or rather, the factors that determined the fall of Timbuktu. THE FALL OF TIMBUKTU AND ITS CAUSES Why did Timbuktu decay? Timbuktu decayed because it lost the very source of its wealth and prestige: Timbuktu lost its status of leading trading center. What is important to determine is why Timbuktu lost its status of major international/economic market. In this respect, it is possible to note that the decline of Timbuktu’s market and, therefore, Timbuktu’s crisis, are intimately connected, and, I suggest, are caused by a the interaction of two factors. Th e fi rst factor is represented by the crisis of the Mediterranean trade system, on which Timbuktu’s market was dependent, while the second factor is represented by the demise of those insti- tutions that had made Trans-Saharan trade an effi cient enterprise during the Songhai empire. In the fi rst respect, we may recall that Timbuktu was part of a trading system which was centered on/around the Mediterranean sea and the Mediterranean trade (Fage 1988: 216–218). As July aptly put it, Timbuktu was part of a trade system and trade routes “which stretched as far as Venice, Genoa and Cairo” (1970:59) and also included Spain and Turkey, or, to be more precise, Istanbul. In other words, Timbuktu’s economy was on the semi-periphery13 of the Mediterranean economy. Th e implications of this connection are quite intui- 10. Th e two empires I am referring to are the Empire of Mali and, in its wake, the Songhai Empire. 11. If innovation, economies of scale, education, capital accumulation, etc. “are not causes of growth; they are growth” (North and Th omas, 1973:2), then one of the most evident aspects of the economic development of Timbuktu and the neighboring area is represented by remarkable increase in the number of Trans-Saharan trails and wells. In the fi rst respect, it is possible to note that in contrast to the past, when the Berbers were using mainly two Trans-Saharan trails (a westward one from Morocco to Mauritania; and a southwared one from the Fezzan to the Middle Niger and Lake Chad), many more trails were used by the growing Trans-Saharan trade. In the second respect, it is possible to recall that the increase in the number of wells was realized by dig- ging again the old wells and also by making new wells. On this and related aspects, see Davidson (1991:71). 12. In this respect, the case of Audogast is emblematic. Th is city, which had been a major city in Sudan at least from the tenth century, today is just, to quote Davidson (1991:76), nothing more than “a puzzle for archeologists”. Similarly, the city of Kumbi, which was the largest city of Western Sudan with a population of about 20,000 inhabitants today is just “a mark on the Kaarta region—200 hundred miles in the Southern fringe of the desert from Bamako” (Davidson 1991: 76), the capital of Mali. On this, see Basil Davidson, Africa in history. Th emes and Outlines, New York, Collier, 1991; see also Basil Davidson, Th e Lost Cities of Africa, New York, Little, Brown and Company, 1987. 13. According to Wallerstein (1974), world economies are “divided into core-states and periph- eral areas...there are also semieripheral areas, which are in between the core and the periphery on a series of dimensions, such as the complexity of economic activities, strength of the state machinery, cultural integrity, etc. Some of these areas had been core-areas of earlier versions of a given world- economy. Some had been peripheral areas that were later promoted, so to speak, as a result of the changing geopolitics of an expanding world-economy” (Wallerstein 1974:349). Th is said, and this is the most relevant aspect for the purposes of the present discussion, Wallerstein also notes that “ the semiperiphery, however, is not an artifi ce of statistical cutting points, nor is it a residual category. Th e semiperiphery is a necessary structural element in a world economy” (Wallerstein 1974:349). Riccardo Pelizzo274 Timbuktu: A Lesson in Underdevelopment 275 tive: Timbuktu’s fortunes and prosperity were associated with and subordinate to those of the system as a whole, i.e. to the core. Th e inevitability of this link, and its potentially destructive consequences, became apparent after the discovery of the Americas in 1492 and “the opening of the Southern sea route to the Indies round the Cape of Good Hope” (Glamann 1977:210). In fact, with the development of these new, trans-oceanic sea-routes European trade (could) “no longer be treated in isolation but (had) to be viewed in a more or less global context” (Glamann 1977:190). With the emergence of the trans-oceanic trade, there occurred what Braudel calls (Braudel 1977:66–67) “a shift of the center of gravity.” Th is transformation was particularly relevant with regard to the Mediterranean trade. In fact, the Mediterranean sea (and trade for that matter) lost its previously hegemonic status as trade started to follow alter- native routes. Although the impact of the Trans-Oceanic trade on the Mediterranean economy was not immediate, its eff ects were quite evident by the second decade of the seventeenth century (Braudel 1949). Following Braudel, economic histo- rians have, correctly, underlined that the crisis of the Mediterranean powers, most notably Spain and Italy, could not be entirely ascribed to the emergence of the Trans-Oceanic trade. Th is argument holds true, as it will shortly be demon- strated, also in the case of Timbuktu. However, it is also undoubtly true that nei- ther the Venetian trade nor that of Timbuktu benefi ted from the new patterns in international trade. Th is said, Timbuktu’s trade partners, and Timbuktu itself, were not simply hurt by the fact that Mediterranean trade had lost its previous hegemonic posi- tion in the world economy. Th ey were also hurt because the trading system, that resulted from their interaction and in which they had prospered, was rapidly fall- ing apart for two related reasons: fi rst, the crisis of each of the system’s compo- nent parts; and second, the system’s component parts’ inability to function as a system. With regard to the fi rst, it is possible to note that the crisis of both Spain and Italy were clear and were clearly illustrated by several socio-economic indica- tors. First of all, analysis of the demographic trends in Spain has recently sug- gested that Spanish population either stagnated or diminished in the course of the 17th century (Vives 1970:122–124). In contrast, the Italian population suff ered a marked decline in the same century. Th is decline is well documented in the literature. Table 1 reports demographic trends in eighteen Italian cities from the end of the 16th century to the end of the 17th century. Th e data show that in ten out of the eighteen cities under study the population declined, though the mag- nitude of this decline varies from city to city. In this respect, note that while the population dropped by more than 40 in Verona and Catania (respectively by 40.9 and 43), it declined only by 10.8 in Napoli and by 7.11 in Venice. On the contrary, the population remained fairly stable in Modena, where it grew from 18,001 inhabitants in 1581 to 18,025 in 1683, while it grew in Vicenza, Milano, Roma, Bologna, Torino, Reggio Emilia and Livorno. It is probably interesting to note that the population declined in all the maritime cities with the exception of Livorno. Th e Spanish and Italian data, which show, at best, demographic stagnation and, at worst, depopulation draw a picture of demographic decay in the two Southern European countries. Th is decay becomes particularly evident, and one should probably say dramatic, when Spanish and Italian demographic trends are compared and contrasted with those of Northern European capitals. As the end of the 16th century, Venice was the second largest city in Italy after Napoli and it had a larger population than London, Amsterdam and Lisbon. Only Paris, among the European capitals under consideration, was larger than Venice. At the end of the 17th century, the picture has dramatically changed.14 Th e Venetian population has declined by slightly more than 7, while the population City Year Population Year Population Venezia 1586 148647 1696 138067 Verona 1586 52109 1692 30685 Brescia 1586 42660 1658 40220 Vicenza 1586 21268 1663 25143 Udine 1586 14579 1672 13389 Napoli 1591 209470 1688 186769 Milano 1592 120000 1688 125829 Roma 1600 113760 1700 141784 Firenze 1609 72000 1701 66000 Bologna 1600 62844 1701 63346 Genova 1581 47708 1682 39457 Messina 1583 42004 1713 40393 Ferrara 1601 32860 1704 25500 Catania 1583 28465 1713 16222 Torino 1612 24410 1702 43866 Modena 1591 18001 1683 18025 Reggio Emilia 1591 11137 1697 14803 Livorno 1581 4985 1694 17709 Source: Daniele Beltrami, Storia della Popolazione di Venezia dalla Fine del Secolo xvi alla Caduta della Repubblica, Padova, cedam, 1954, p. 63 and 65. Table 1 – Population of Some Italian Cities at the End of the 16th and 17th Century 14. Although our data are considerably diff erent from those presented in Pelizzon (2000: 87-195), they show the same pattern displayed by Pelizzon’ s data. Riccardo Pelizzo276 Timbuktu: A Lesson in Underdevelopment 277 has increased in all of the other European capitals. Specifi cally, during the 17th century the population grew by 1014 in London, by 127 in Paris, by 362.5 in Amsterdam and by 53.8 in Lisbon. Th e implication of this change is that, at the end of the 17th century, only Lisbon had a smaller population than Venice. Venetian decline continued during the 18th century, so much so Venice was the smallest European capital at the end of that century. Data are reported in Table 2. Th is said, it is also worth recalling that the crisis of the previous Mediterranean powers—Spain and Italy—was not just a demographic crisis. It was, above all, a crisis in the industrial/productive activity. Th is crisis was quite evident both in the Spanish and in the Italian case. Vives (1970:139) illustrates the collapse of the Castillian industry by presenting and discussing the data con- cerning the “population decline in the old and industrial and mercantile cities of Castile” (1970:139). Th e decline is truly impressive: from 1594 to 1694 the indus- trial population in Burgos declined by 29.4, by 55.2 in Valladolid, by 68.2 in Palencia, by 51.2 in Salamanca, by 65 in Avila, by 70.7 in Segovia and by 54.2 in Toledo. In sum, the population decreased in all major Castillan eco- nomic centers. Th ese data are interpreted to refl ect the severe economic recession in the region. Data are reported in Table 3. It is, however, worth noting that the Italian crisis was as severe as the Spanish one. Cipolla (1970) presents very interesting data in this regard. Th e number of cloths produced in Como fell from an yearly average of 8000 per year to just about 400 cloths a year at the mid of the seventeenth century; in Milano the pro- duction of cloths dropped from an yearly average of 15000 cloths at the begin- ning of the 17th century to a yearly average of about 3000 cloths around 1640; “in Genoa the silk industry was in full decline by the last decade of the sixteenth cen- tury. A seventeenth century tradition holds that, whereas there had been 18000 silk looms in the territory of the Republic, by 1608 there were only 3000 and scarcely more than 2500 in 1675. Th e total of 18000 weavers for the middle of the sixteenth century must be considered an exaggeration. However, there can be no doubt that the silk industry did decline during the course of the century” (Cipolla 1970:199). Among the various data on the Genovese economy presented by Cipolla, the most relevant for the purposes of the present analysis is represented by the fact that the annual value of goods imported and exported by the port dropped from 9 million lire at the beginning of the century to 3 million lire at the end of the century. As dramatic as the recession might appear, we should probably recall that few places in Italy were hit as hard as Venice by the economic crisis. Cipolla (1970:196) showed that the production of cloths dropped by almost ten times in the course of the 17th century. Yet, the Venetian crisis did not simply aff ect the clothing industry, but it aff ected every aspect of the Venetian economy. Th is problem can be highlighted by looking at the number of contracts of Garzoni15 registered every year. In contrast to the last decade of the sixteenth century, when an average of 1309 contracts were registered, the seventeenth century had an yearly average of about 1174 registered contracts. In other words, there was a drop of more than 10. Data concerning the number of contracts of Garzoni are pre- sented in Table 4. In sum, the demographic indicators as well as the industrial indicators con- sistently show that the economy of two, previously crucial, component parts of Table 2 – Population in Some European Capitals Population (year) Population (year) Population (year) London 100000 (1568) 1114644 (1701) 2235344 (1801) Paris 216000 (1590) 492000 (1700) 548000 (1801) Amsterdam 40000 (1514) 185000 (1685) Lisbona 65000 (1550) 100000 (1600) 175000 (1775) Venezia 143637 (1586) 138067 (1696) 137240 (1797) Source: Daniele Beltrami, Storia della Popolazione di Venezia dalla Fine del Secolo xvi alla Caduta della Repubblica, Padova, cedam, 1965. Table 3 – Population Decline in the Old Industrial and Mercantile Cities in Castile 1530 1594 1646 1694 Burgos 1500 2665 600 1881 Valladolid 6750 8112 3000 3637 Palencia 1364 3063 800 972 Salamanca 2459 4953 2965 2416 Avila 1523 2826 1123 965 Segovia 2850 5548 1625 Toledo 5898 10933 5000 5000 Cuenca 3095 800 1641 Source: Jaime Vives, “The decline of Spain in the Seventeenth Century,” in Carlo Maria Cipolla (ed.), The Economic Decline of Empires, London, Methuen, 1970, p.139. 15. Th e contracts of the Garçons. Riccardo Pelizzo278 Timbuktu: A Lesson in Underdevelopment 279 and Dutch services and goods were off ered at lower prices” (Cipolla 1970:203). Cipolla ascribes the higher prices of Italian goods to a variety of factors: the qual- ity of the Italian goods was higher, the Italian productive system was obsolete, the pressure of taxation in Italy was very high and the cost of labor was very high (Cipolla 1970:204–209). Yet, there is another reason why Venetian goods and services lost their competitiveness: the Turkish custom maintained a 5 import tax on all Venetian goods, that had been established in 1505. In contrast, “the Austrians were allowed, already in 1600, to pay only 3, and the same privilege will be obtained by the French in 1673, by the English in 1675 and by the Dutch in 1680” (Zorzi 1991:391). Finally, the Venetian trade with the Near East was nega- tively aff ected by the fact that Italian ships were not as fast and technologically advanced as the English and the Dutch ships. High prices, high import taxes and an anachronistic shipping industry made it impossible for Venice and the other component parts of the previously hegemonic trade system in the Mediterranean to function as a system. Given Timbuktu’s economic dependence on the Mediterranean market and trade system, it is not terribly surprising that the combined eff ects of the emer- gence of the Tran-Oceanic trade, the decline of the previous commercial powers in the Mediterranean and their inability (and unwillingness) to function as a system had catastrophic consequences on Timbuktu’s economic life and wealth. Th is said, Timbuktu’s economic decay—as that of Italy and Spain—cannot be entirely ascribed to the action of international/economic economic factors. Domestic/institutional factors also played a major role in provoking Timbuktu’s crisis. Th e most important of these domestic/institutional/institutional/regional factors is represented by the fall of the Songhai empire, which had been attacked and defeated by the Moroccans in 1591. Th is event is of particular importance in the history (and especially for the fortunes) of Timbuktu because it created a vacuum of power in the region.16 Th e creation of a vacuum of power repre- sents the major diff erence between the fall of the Ghana and the Mali empires on the one hand and that of the Songhai empire on the other hand. Th e Ghana empire was directly followed by the Mali empire and this, in turn, was immedi- ately succeeded by the Songhai empire. In neither case, did the transition from one empire’s rule to the other’s leave a vacuum. On the contrary, the region was left in a state of anarchy, in which no power was able or willing to maintain the Mediterranean trading system were suff ering a major economic recession. However, it is also worth noting that the trade system on which Timbuktu’s economy depended was not falling apart only because it had been challenged by the emergence of the Trans-Oceanic trade system nor because some of the system’s component parts were facing—as the Spanish and the Italian case illus- trate—a major economic crisis. Th e Mediterranean trade system was also fall- ing apart, and did fall apart by the end of the 17th century, because its component parts proved to be unable to function as a system. Th is problem can be best exemplifi ed by the transformation in the economic relationship between Venice and Istanbul. Venetian merchants, who had been a remarkable presence in Constaninople (or Istanbul) since, at least, the end of the tenth century (Luzzato 1961:13) were no longer hegemonic within the Turkish market. Zorzi illustrated the problem very clearly when he observed that “the Eastern Mediterranean sea is increasingly less an Italian lake” (Zorzi 1991:366). Cipolla provides a very interesting example in this respect, by noting that “it appears that during the fi rst years of the sev- enteenth century Venice was able to export to Mediterranean ports something like 25000 cloths…a hundred years later, according to the Venetian ambassador at Contantinople, Venice was exporting not more than 100 cloths a year to that area, about fi fty to Constantinople, and the same number to Smyrna” (Cipolla 1970:203). Cipolla suggests that “the fundamental reason why Italian goods and services were supplanted by those off ered by others was always the same: English, French Table 4 – Number of Contracts of Garzoni Year Contracts 1592 1447 1646 842 1597 1078 1656 1057 1598 1402 1658 1044 1621 1120 1662 1673 1626 1240 1663 1150 1643 1080 1664 1504 1645 1022 1670 1180 Source: Daniele Beltrami, Storia della Popolazione di Venezia, Padova, cedam, 1954, p. 200. 16. For a discussion of this point, see Robin Hallett (1970:152); see also James Curry (1999:157-160). Riccardo Pelizzo280 Timbuktu: A Lesson in Underdevelopment 281 effi cient institutions or acceptable institutional performances. Th e institutional decay was particularly evident with regard to property rights. With no power left in the region to protect property rights and to prevent caravans and traders from being robbed in their Trans-Saharan journeys, both the risks and the costs of trade increased. Th ese increasing costs are responsible—in the end—for a fur- ther decline in the size of Trans-Saharan trade and for the irreversible crisis of Timbuktu.17 CONCLUSIONS Th is analysis of the rise and, more importantly, the fall of Timbuktu, with its many mysteries, seems to have several important implications. Th e purpose of this section is to discuss the most important lessons that can be learned from Timbuktu’ s history. Th e fi rst lesson concerns the nature of development: development is very fragile, it is diffi cult to achieve and very easy to lose. In a few years the rich, cultured, civic and developed Timbuktu degenerated into what July calls “an isolated town, a handful of unimpressive mud buildings housing a listless popu- lation which time had fi nally passed by” ( July 1970:60). Culture, wealth and large stocks of social capital were not suffi cient to prevent Timbuktu from decaying. Th e second lesson is that neither development nor decay are promoted (at the ontological level) or explained (at the theoretical level) by any single factor. Although some factors might individually have a larger eff ect, or a better explanatory power, than others, causation and/or explanation result only from the interaction of multiple factors. Th e case of Timbuktu is exemplary in this respect: Timbuktu lost power and prestige because its market decayed. However, no single factor can account individually for this event. In the present paper we have suggested that the crisis of Timbuktu’s market was provoked, among other things, by the interaction of two factors: fi rst, the general decline of Mediterranean trade in the world trade resulting from the emergence of the new Trans-Oceanic trade, and the crisis of the system’s component parts at the indi- vidual level and their inability to function as a system; second, the institutional decay which followed the fall of the Songhai empire, which made Trans-Saharan trade particularly risky and, henceforth, economically ineffi cient. Th is said, two points need to be stressed: fi rst, the survival of an hegemonic Mediterranean trade system would not have preserved by itself the wealth of Timbuktu in the absence of an adequate institutional framework. Similarly, a stronger institutional system—per se—would not have prevented Timbuktu from decaying but would have certainly limited the extent of the decay. In sum, the third and fi nal lesson to be learned from Timbuktu’s story is that social phenomena result from diff erent factors, and, of course, their com- plex interactions. Phenomena, especially social phenomena, become meaningful only when they are analyzed in what critical phenomenologists (or neo-phenom- enologists) call the infi nite web of relationship in which they exist. In this case, Timbuktu’s decay becomes meaningful only if it is analyzed in the light of the relationship between external and internal factors. To focus on one set of fac- tors, instead of focusing on both of them, might increase the elegance and the parsimony of our explanatory model, but it undermines its precision and reliabil- ity. But this is not the purpose of social research, whose objective is (and should be) the understanding of reality rather than its misrepresentation. Th is objective can be achieved only if we constantly remember that words of Guicciardini, who noted in his Ricordi Politici e Civili “it is a great error to speak of the aff airs of this world in absolute terms, without discrimination, and—so to speak—by rule; for they almost always involve distinctions and exceptions, because circumstances vary, and they can never be subject to one single measure. Th ese distinctions and exceptions are not to be found in books: this must be taught by mature discre- tion alone.” REFERENCES Africanus, Leo (1896)[1600]. Th e History and Description of Africa and of the things Th erein Contained. London: Hakluyt Society. Allen, William (1848). A Narrative of the Expedition to the NigerRriver. London: Bentley. Barth, Heinrich (1965) [1857]. Travels and Discoveries in North and Central Africa. London: Frank Cass. Beltrami, Daniele (1954). Storia della Popolazione di Venezia dalla Fine del Secolo XVI alla Caduta della Repubblica. Padua: CEDAM. Braudel, Fernand (1949). Le Mediterranee et le Monde a l’Epoque de Philippe II. Paris: Colin. Braudel, Fernand (1977). Afterthoughts on Material Civilization and Capitalism. Baltimore: The Johns Hopkins University Press. Cadamosto, Alvise (1937). "The Voyages of Cadamosto." Chapter in G.R.Crone (ed.), Th e Voyages of Cadamosto. London: Hakluyt Society. Caillie, Rene (1830). Travels through Central Africa to Timbuctoo and across the Great Desert to Morocco. London: Colburn and Bentley. Cessi, Roberto (1944). Storia della Repubblica di Venezia, Volume 1. Milan: Principato. 17. July (1970:59-60) observes that “ Timbuktu fl ourished as long as the civilization it represented was able to provide peace and political stability for its trade. With the col- lapse of the Songhai empirre, the fortunes of Timbuktu took a permanent turn which saw the once thriving city degenerate into an isolated dusty town” . Riccardo Pelizzo282 Timbuktu: A Lesson in Underdevelopment 283 Moore, Jason W. (2000). "Sugar and the Expansion of the Early Modern World Economy: Commodity Frontiers, Ecological Transformation and Industrialization," Review XXIII, 3: 409-435. North, Douglass and Robert Paul Thomas (1973). Th e Rise of the Western World. Cambridge: Cambridge University Press. Oguto, Matthias A. and Simon S. Kenyanchui (1991). An Introduction to African History. Nairobi: Nairobi University Press. Parry, J.H. (1967). "Transport and Trade Routes." Pp. 155-219 in E.E. Rich and C.H. Wilson (eds.), Th e Cambridge Economic History, Volume 4: Th e Economy of Expanding Europe in the Sixteenth and Seventeenth Centuries. Cambridge: Cambridge University Press. Pelizzon, Sheila (2000). "Grain Flour, 1590-1790." Review XXIII, 1: 87-195. Pertusi, Agostino (ed.) (1973). Venezia e il Levante fi no al Secolo XV. Florence: Leo Olschki Editore. Vives, Jaime Vivens (1970). "The decline of Spain in the Seventeenth Century." Pp 121-167 in C.M.Cipolla (ed.), Th e Economic Decline of Empires. London: Methuen and Co. Wallerstein, Immanuel (1974). Th e Modern World System I: Capitalist Agriculture and the Origins of European World-Economy in the 16th Century. New York: Academic Press. Zorzi, Alvise (1991). La Repubblica del Leone. Venice: Rusconi. Cipolla, Carlo Maria (1970). "The Economic Decline of Italy," Pp. 196-214 in C.M.Cipolla (ed.), Th e Economic Decline of Empires. London: Methuen and Co. Cipolla, Carlo Maria (1980). Before the Industrial Revolution: European Society and Economy, 1000-1700. New York: Norton and Company. Cipolla, Carlo Maria (ed.) (1970). Th e Economic Decline of Empires. London: Methuen and Co. Cissoko, Sekene Mody (1996). Toumbouctou et l’ Empire Songhai. Paris: L’ Harmattan. Crone, Geral Roe (ed.) (1937). Th e Voyages of Cadamosto. London: Hakluyt Society. Curry, James (1999). Africa from the Sixteenth to the Eighteenth Century. Paris: UNESCO. Curtin, Philip et al (1978). African History. Boston: Little, Brown and Company. Davison, Basil (1987). Th e Lost Cities of Africa. New York: Little, Brown and Company. Davidson, Basil (1991). Africa in History: Th emes and Outlines. New York: Collier. Davidson, Basil (1993). African Civilization Revisited. Trenton: Africa World Press. Dudney, Daniel (1999). "Bringing Nature Back In: Geopolitical Theory from the Greeks to the Global Era." Pp. 25-57 in Daniel Dudney and Richard Matthew (eds.), Contested Grounds: Security and Confl ict in the New Enviromental Politics. New York: State University of New York Press. Dudney, Daniel and Matthew, Richard (eds.) (1999). Contested Grounds: Security and Confl ict in the New Enviromental Politics. New York: State University of New York Press. Es-Sadi (1981) [1898-1900]. Tarikh Es-Sudan. French Translation by O.Houdes. Paris: Librarie d’ Amerique et d’Orient. Elliot J.H. (1970). "The decline of Spain." Pp. 168-195 in C.M.Cipolla (ed.), Th e Economic Decline of Empires. London: Methuen and Co. Fage, J.D. (1988). A history of Africa. London: Hutchinson. Glamann, Kristoff (1977). "The Changing Pattern of Trade." Pp. 185-289 in E.E. Rich and C.H.Wilson (eds.), Th e Cambridge Economic History of Europe, Volume V: Th e Economic Organization of early Modern Europe. London: Cambridge University Press. Hallett, Robin (1970). Africa to 1875. Ann Arbor: The University of Michigan Press. Haywood, Austin Hubert (1912). Th rough Timbuctoo and Across the Great Sahara. London: Seeley, Service and Co. Hunwich, John O. (1999). Timbuktu and the Songhai Empire. Boston: Brill. July, Robert W. (1970). A History of the African People. New York: Scribner’s Sons. Krugman, Paul (1997). Development, Geography and Economic Th eory. London: MIT Press. Lenz, Oskar (1886). Timbuctoo, Volume II. Paris: Hachette. Luzzato, Gino (1961). Storia Economica di Venezia dal XI al XVI Secolo. Veneice: Centro Internazionale delle Arti e del Costume. Mollat, Michel, Philippe Braunstein, and Jean Claude Hocquet (1973). "Refl exions sur l’ Expansion Venetienne en Mediterranee." Pp. 515-539 in Agostino Pertusi (ed.), Venezia e il Levante. Florence: Leo Olschki Editore. Riccardo Pelizzo