The cities of West Africa's Middle Niger, only recently brought to the world's attention, make us rethink the 'whys' and the 'wheres' of ancient urbanism. They present the archaeologist with a novelty; a non-nucleated, clustered city-plan with no centralized, state-focused power. This book explores the emergence of these cities in the first millennium B.C. and the evolution of their hinterlands from the perspective of the self-organized landscape. Cities appeared in a series of profound transforms to the human-land relations and this book illustrates how each transform marked a leap in complexity.
About the Author
Roderick J. McIntosh is Professor of Anthropology at Rice University and visiting Professor of Archaeology at the University of Pretoria, South Africa. His recent publications include The Peoples of Middle Niger: Island of Gold (1998), The Way the Wind Blows: Climate, History, and Human Action (2000) and Geomorphology and Human Palaeoecology of the Méma, Mali (2005).
Read an Excerpt
Cambridge University Press
052181300X - Ancient Middle Niger - Urbanism and the Self-Organizing Landscape - by Roderick J. McIntosh
Jenne-jeno "discovered" to the world
It can be said, without much fear of contradiction, that the ancient Middle Niger is the most recent of the world's major urban civilizations to be discovered. Sadly, this discovery only dawned upon the archaeological world in the late 1970s. Why sadly? Independently developed. Enormous in size and population. Sophisticated in crafts. Voluminous in production of their manufactures. And with dominion over the granary of West Africa, the Niger River's vast interior floodplain that, at its heyday, covered more than 170,000 square kilometers. (Compared to Pharaoh's 34,000 square kilometers and the maximum cultivable 51,000 square kilometers of Mesopotamia: Lehner 2000: 283; T. J. Wilkinson 2003: 76.) These Middle Niger cities comprised, by anyone's definition, a major and indigenous world civilization. Yet they were occluded from the archaeological eye until very recently (to put it delicately).
In fact, not until 1977, with the initial excavations at and regional survey around the first discovered of these cities, Jenne-jeno, was the world alerted to this highly original species of urbanism. Discovery: a word minted expressly to make archaeologists squirm! It is a word that archaeologists accept as connoting, for the larger public, the romantic in the archaeologist's job description; but it is a word they also cringe to hear directed at themselves (perhaps only rarely for false modesty!). Discovery? After all, these cities were hardly unknown before the first archaeologists walked their surfaces. The townspeople of the descendant city (Jenne) certainly knew of Jenne-jeno. Those same townspeople directed the archaeologists to the tall mound (a tell, or artificial hillock, a city site built up of centuries, millennia of trash and rain-melted mud buildings) marking the ancient urban settlement. "Discovered to the world," perhaps, is a better description of the archaeologists' role. But hardly discovered.
Nevertheless, the reason for the delayed appreciation of these Middle Niger cities by the archaeological world also has much to say about why their very originality has begun to shake up the conduct of pre-industrial
|Image not available in HTML version|
Fig. 1.1. Panoramic view of the Jenne-jeno mound, viewed from the north (east-west dimension, c. 800 m)
urban research in places as distant as Mesopotamia and northern China. That shake-up will be the last item to savor from this book's menu - dessert, as it were, for those of us who long have thought that Africa has received short shrift in the Grand Accounting of the world's prehistoric accomplishments (R. McIntosh 1999a)! (After all, we are all of us Africans, all Homo sapiens sapiens, whatever our mass, hue and flavor - shouldn't we all be more acknowledging of our homeland?) As the appetizer, however, let us turn to the entry (with annotations in brackets and italics) from my field journal for early 1977. This was the moment of "discovery" of Jenne-jeno as a small group of archaeologists (a Malian and three expatriates) trudged up the steep northeast slope of the Jenne-jeno mound for the first time (fig. 1.1).
3 February 1977: In the morning we went out, at last, to Djenné-djeno. [We had arrived in Mali's capital, Bamako, on 22 January, but official visits and delayed permissions, purchases, nursing a cranky car (forgive me!), and the two days needed to crawl the 400 kilometers of neglected national highway had delayed, and delayed, and endlessly delayed ... ] It's about three kilometers over the floodplain to the southeast of the city of Djenne [The spelling of the town and ancestral site will change over time, at the insistence of the government, as will our mode of transport from cranky car to even crankier donkey-cart - funding for Middle Niger research has never quite reached Mesoamerican, or even Mesopotamian, proportions!] [ fig 1.2].
The site is absolutely incredible - it defies description. It seems to be divided into two parts, the larger of which must be almost a half-kilometer in length. [I laugh now at my incredulity-driven underestimation - the "larger of which," the Jenne-jeno mound itself, is in reality over a kilometer long. The smaller of the pair, linked to Jenne-jeno by a causeway, is the separate satellite mound of Hambarketolo, itself a respectable 290 meters in length. In fact, the urban site of Jenne-jeno is divided into fully seventy physically discrete parts! And THAT is the core story of this book!]. It is literally a mound of potsherds. [It is this blanket of millions of sherds on the surface of all Middle Niger sites that most immediately strikes the tourist. A visitor, a Palestinian archaeologist who had once dug at Jericho with Kathleen Kenyon, said it was this overburden of broken pottery that astonished him even more than the site's enormous size.] We found at least twenty funerary mounds exposed, roof tuyeres, pottery bedsteads, Muslim burials. There seems to be definite regions of concentrations - a function of erosion or not, I don't know. All pretty staggering.
Staggering and, frankly, discouraging. In a short time, this site would utterly transform our thinking about the origins of cities in Africa south of the Sahara. However, what this entry in the field log of my first day's impression of Jenne-jeno does not mention is just how overwhelming, and, frankly, terrifying that first impression was for this green, excited (and impoverished!) graduate student. As my partner in "discovery," Susan McIntosh, puts it, "Months of preparation. All that sampling
|Image not available in HTML version|
Fig. 1.2. Donkey-cart, the only reliable transport over the tortured clays of the Middle Niger floodplain. Note the extremely flat alluvium, interrupted vertically only by tell archaeological sites
theory, all those pages upon pages of exquisite gridding and random number strategy design in the grant proposal - to see it all disappear in a puff of smoke the moment I topped the rise of the mound!"
Scared, yes, because here were graduate students visiting the site for the first time in the company of the newly appointed head of Mali's national bureau for heritage, Direction du Patrimoine, Dr. Alpha Oumar Konaré. How much more terrified would we have been had we, at the time, but known our colleague's political future! In 1991, Dr. Konaré would spearhead the overthrow of the country's twenty-eight-year dictatorship, then serve two terms in the 1990s as the first democratically elected President of Mali, before stepping down to become the Chairman of the African Union (successor body to the Organization of African Unity, where he has spearheaded the initiative for a continent-inclusive Etats Unis de l'Afrique, a United States of Africa). But that is a story for later in this book (chapter 5). Dr. Konaré had been trained in Poland in medieval, ecclesiastical archaeology and was anxious to witness firsthand some techniques more appropriate to the kind of sites for which he had responsibility in Mali. The pressure was on!
The immediate source of our anxiety was that we had to make some initial sense of a settlement that stretched out before us at a scale that defied our wildest imagination. Not only that - we had to make sense of Jenne-jeno quickly, so we could select the best places for our preliminary test units. Dr. Konaré only had a week to stay with us (most of that eaten up by travel). He could not wait around while we walked and walked the site in a daze of indecision! We had to tickle out patterns to the distribution of the mass of features and artifacts littering the surface. Superficially, there seemed a tornadic chaos of furnaces, houses built of clay bricks - round and square - burial urns, burials in urns, and skeletons planed off by sheet-wash erosion - all exposed to desecrating view etc., and of the thousands of smaller items: ceramic bed rests, roof tiles, furnace blow pipes, copper bits, iron slag, animal bones, etc. Of course, we cannot forget the tons upon literal tons of broken pieces of pottery - all at a site unresearched previously and set in a vast region unexplored, in any systematic manner, by archaeologists.
Between the two of us, Susan McIntosh and I had logged over a dozen seasons of archaeological experience at a variety of sites in the United States, in three European countries, and in the forests of Ghana. No previous experience, however, could have prepared us to grasp the sheer vastness of this ancient settlement. And here is the more critical point: Beyond sheer scale, there was nothing erected on the surface of this site to direct the eye, to focus the archaeologist's mind, on the vast cityscape arrayed before us. Details about where and why we selected the first four test units must await a later chapter (chapter 4, where the strategies by archaeologists for making sense of a deeply stratified tell sequence are explained). Here we must explore the more fundamental question of why Middle Niger cities were so ignored by archaeologists and historians, so neglected by the outside world, for so long. Let us return to this freighted concept of archaeological discovery.
Does this mean that the site of Jenne-jeno had been previously undiscovered? Hardly. As mentioned before, the site was well known to the locals. But even beyond that, it had even been walked over by otherwise quite competent colonial-era prehistorians. But its importance as a city and as a pre-Islamic center of trade, wealth, and population was obscured to them. To understand why this should have been, we need to understand the history of scholarship (classical Arabic and Western) concerning West African cities and society.
It is a strange thing, but ultimately not a mystery, that neither Jenne-jeno nor, indeed, any of the hundreds of Middle Niger cities deserve any mention in the first Arab accounts of trade, towns, and states of West Africa south of the Sahara (see S. McIntosh and R. McIntosh 1980, I: 41-45; R. McIntosh and S. McIntosh 1981: 2-8, 19-22). These sanctioned, authorized court geographies, being second- and third-hand compilations of travelers' accounts written to a template of received, canonical views of the world with an intellectual lineage going back to Ptolemy, began to appear in the eighth century AD (Levtzion 1973; Levtzion and Hopkins 1981; Levtzion et al. forthcoming). At that time, Jenne-jeno and the other Middle Niger cities were arguably at their apogee and had certainly been in existence for over a millennium - yet, silence. Trade across the desert in West African gold, that commodity wantonly lusted after by all states north of the Sahara, was already centuries old. Jenne-jeno and other Middle Niger cities were principal southern termini in that commerce - yet, silence. And the travelers and merchants from whom the court geographers heard of these exotic African lands and peoples would surely have themselves had heard (if not actually traveled there themselves) of the hundreds of cities along the Middle Niger - yet, silence still.
Silence in the written records until Jenne (the descendant town, 3 kilometers away), or "Geni," was discovered to the world across the Sahara in the fifteenth century by an Italian commercial spy. In 1447, the Genoan Antonio Malfante was poking around for intelligence about the sources of West African gold in the North African city of Tuat. Predictably, he was getting little out of the suspicious local merchants. But he did learn of fabled towns, of Thambet (Timbuktu) and of "Geni" - a city he calls a "civitate," with the clear implication, for a northern Italian of this epoch, of a mature city-state with a larger administered hinterland, or "contado" (R. McIntosh and S. McIntosh 1981: 4). Would that we knew just what his southern merchant informants in Tuat had told him to occasion use of this highly freighted term. "Civitate," with its implications of a self-administering city-state and hinterland, goes to the very core of our definition of a city.
If the modern town of Jenne remained so obscure to the world (even under the intense glare of interest from the Islamic and European worlds in this center of the trans-Saharan gold trade), what chance was there of any mention of the ancestral Jenne-jeno? Little. For that, the ancient city must wait two more centuries.
Around AD 1656, the imam and notable of Timbuktu, al-Sa˓di, wrote a personal history in Arabic, a ta'rikh, of the important towns, states, and historical events of the Western Sudan (al-Sa˓di 1900; by the Western Sudan we mean, roughly, northern West Africa, the southern fringe of the Sahara). In his Ta'rikh es-Sudan, al-Sa˓di recalls the oral traditions of his natal town of Jenne:
Jenne is one of the grandest markets in all the Muslim world. There, one will encounter merchants bringing salt from the mines of Teghazza, who meet merchants bringing their burden of gold from the mines of Begho ... Everyone conducting business in Jenne is blessed by God with great profit and, indeed, only God Himself (may His name be blessed!) keeps a true accounting of the fortunes that have been made there. It is because of this blessed city that caravans flow towards Timbuktu from all points on the horizon, from the east, from the west, from south and from north. (al-Sa˓di 1900: 22-23, trans. R. McIntosh)
According to al-Sa˓di, Jenne was founded by pagans in the middle of the eighth century AD. Its king and all its inhabitants converted to Islam around AD 1200 (1900: 23). But al-Sa˓di is very explicit that the "founding" spoken of was not of the present town, but of its ancestor, Jenne-jeno (that is, "ancient Jenne" in Jennéké, a Songhai language dialect). First site of the city is the mound of Zoboro (or Djoboro) (or "ancient Jenne" in the local Bozo language), located just south of the present town (1900: 23). Sparse enough reference, but an eighth-century date for the foundation of any town was so impossibly early, given received wisdom about when and why states and towns should have appeared in the Western Sudan, as to have created considerable skepticism amongst historians. That is, until proper excavation revealed the true dates for the settlement's founding.
This is the first surprise from al-Sa˓di: His founding date is, frankly, too early for a scholarly world convinced that states and cities could not have appeared amongst the "backward" people of black West Africa until there had been sufficient contact with the civilized peoples of Islamic North Africa and metropolitan Egypt and other Islamic nations (R. McIntosh and S. McIntosh 1988; R. McIntosh et al. 1989). But wait! - al-Sa˓di provides another strange tale about Jenne that will come to have enormous implications for understanding the originality of urbanism here. He is not terribly specific about when this anecdote took place. By the time of the gold trade to the far Akan forests (the Begho of the last quote, far to the south in West Africa) at the very latest, Jenne held authority over a large territory. The territory's size may have been fully one-tenth of the entire Middle Niger, a possible 17,000 kilometers square (al-Sa˓di 1900: 25; R. McIntosh 1979: 15). What really catches the archaeologist's attention is the statement that, throughout this territory, the land is fertile and was well populated; so densely populated, in fact, that 7,077 villages were packed close together. This allowed a curious form of communication:
If the head of Jenne, for example, has occasion to call on an inhabitant of one of the villages located near Lake Debo [approximately 160 kilometers away], his messenger will simply avail himself of one of the city gates. From there, he will cry out the message that he has been charged to transmit. From there, and from village to village, the people of the hinterland will repeat this message until it arrives, almost immediately, to the person sought. That person then rushes to the desired audience. There is no need further to illustrate how densely peopled is the territory of Jenne. (al-Sa˓di 1900: 24-25, trans. R. McIntosh)
Sadly, after these tantalizing suggestions of great antiquity and of enormous regional importance, the ancient city of Jenne-jeno is, once more, lost to history. There are some later elliptical mentions, as when in the fifteenth century an invading Songhai army garrisoned the site during its siege of Jenne lasting seven years, seven months, and seven days (R. McIntosh 1998: 285-92)! Amongst classical Arab or local Islamic scholars there appears to have been no further curiosity about its foundation date or the town's role in the earliest history of the Middle Niger.
In fact, the first colonial achaeologists shared with their Islamic predecessors a disbelief in the ability of West Africans to discover for themselves the fruits of complexity - cities, long-distance trade, states, and the like (R. McIntosh et al. 1989; R. McIntosh 2001). But with French colonialism came renewed interest - albeit not always of the most selfless, uplifting kind.
The archaeological mounds around Jenne were, in fact, objects of the first interest shown by prehistorians in the Middle Niger - unfortunately, an interest in the prehistoric that has made true, scientific archaeology literally a race against time - and against art. Art, specifically, statuettes: Terracotta statuettes are for the legitimate archaeologist here the mixed blessing that gold is for archaeologists elsewhere. On the one hand, they provide an important testimony to the activities and interests of the past population, and thus are necessary to record and publish. Sadly, on the other hand, however, the act of publishing can bring the valued commodity to the attention of the unsavory world of collectors and looters. Colonial figures stationed in Jenne found anthropomorphic terracotta statuettes eroding out of several of the mounds surrounding the city and further down river (Ligers 1957; Masson-Detourbet 1953; Monod 1943; Mauny 1949; "Sculpture soudanaise" 1947). The first recorded "archaeological" episode (of the Indiana Jones vintage!) was occasioned by schoolteacher Vieillard (1940), who sank robber pits into Jenne-jeno in 1938, without, as he complained in an unpublished letter, "any other result than finding some stone beads, pendant fragments, and bits of iron that were impossible to identify" (from Mauny 1961: 102). By the early 1970s, long before serious, legitimate excavation began at any of the tells of Jenne, battalions of peasant looters in the employ of international art thieves had started to turn scores of Middle Niger sites into ghastly, eviscerated cadavers. All the more reason to try to understand why the Jenne tells had not earlier attracted more legitimate scientific attention. Unquestionably, the archaeological obscurity of these ancient sites aided the clandestine activities of the looters and their patrons in the rarified world of the international art and antiquity trafficking.
To the best of anyone's knowledge, the next colonial personality to walk the surface of Jenne-jeno was a truly admirable prehistorian and polymath, a larger-than-life personality, Raymond Mauny. Unsung champion of the richness of the Western Sudanese past, Mauny was an administrator-turned-archaeologist-linguist-historian-historical architect-preserver of cultural heritage. Largely self-taught in many of those disciplines, he nevertheless provided, in his magisterial Tableau Géographique de l'Ouest Africain (A Geographical Overview of West Africa - 1961), a synthetic picture of the Western Sudanese "medieval" (post-Iron Age) past.
Every prehistorian's vision of the past is viewed through a particular lens. Mauny's lens was clouded by the prejudices and Eurocentric conceits of his time (more on these later in this chapter). But he was avidly curious, visiting and sometimes excavating into as many kinds of sites throughout francophone western Africa as he could. (Getting ahead of our story, somewhat, when the true age and urban significance of Jenne-jeno was first presented to him in the late 1970s, long after his field days were over, he was amongst the first enthusiastically to accept and to trumpet those results. This, despite the fact that the new interpretation stuck a dagger into the very heart of his customary explanatory paradigm for the appearance of states and cities. Before accepting the early Jenne-jeno dates, Mauny full expected urbanism along the Middle Niger [and, indeed, anywhere in sub-Saharan Africa] to be late and derived, blocked as it was by the great desert from the "centres of radiation" to the north and east [Mauny 1970: 76].)
It is curious, but highly instructive, to read Raymond Mauny's personal thoughts as he walked the surface of Jenne-jeno. Mauny recalls his frustration, and admits, openly, to being aimless and confused:
The archaeologist is utterly at a loss, lacking any useful diagnostic artifacts - likewise, the examination of surface ceramics, the rare statuette, beads, or tobacco pipes (indicating occupation after 1600) recalls just how little help we receive from local oral tradition or from the Arabic ta'rikhs. We are on fickle land here and we are left with the impression of being only at the very beginning of the pioneering stage of archaeology. (Mauny 1961: 95, trans. R. McIntosh)
It is difficult to find the precise English rendering of "fickle land" (the original is "... en terrain mouvant"). The translation could be the far more evocative "quicksand" - in the sense that Mauny felt swallowed up, drowning in the sheer immensity of the site without a conceptual lifesaver to make sense of what he was seeing (Mauny, pers. comm. 1976). The translation could even be the prescient "earthquake" - in the sense (and he alluded to this feeling in private conversations some twenty-five years later) that a site of this size, clearly urban, clearly important, simply lacked the urban signatures he was looking for. Lacking those urban signatures, Mauny and the colonial archaeologists to follow departed the site without further investigation or commentary, allowing it to languish in the sullen Sahelian sun, the surface eroding away millimeter by millimeter with every monsoonal rainy season, naked but for the dire attentions of the terracotta thieves.
What were those urban signatures? What had even so wise and experienced an archaeologist as Mauny expected to grace the surface of every archaic city - in West Africa, North Africa, or Mesopotamia? This question goes to the core problem, the principal conundrum that had to be solved before the originality and deep age of Middle Niger urbanism could be accepted: City without Citadel!
City without Citadel
This book is about a simple problem - a concise conundrum, really: City without Citadel. When is a city not urban? When there is no king to call it so. Kings need their citadels - if a city has no citadel, no manifestations of kingly presence and power...our conundrum, City without Citadel. Or do we need a logic adjustment? By one way of thinking about circumstances leading to cities, anywhere, the three words of our conundrum simply should not go together. The weight of well over a century of our archaeological experience blowing the dust of memory loss from prehistoric cities throughout the world tells us so. Traditional archaeological theory about the origins and the functioning of the ancient city tells us so. At least that is the message of customary, majority archaeological theory and expectation: In order to have a city, particularly in prehistoric times and, emphatically, those grounded in pre-industrial economies, one also has to have a demonstrable seat of power, with all that implies about social and political relations.
Citadel: seat of coercive power (fig. 1.3); signature of the power of the state. The fantasy in figure 1.3 is the prehistorian's received dream - citadels of monumental scale, of every imaginable cultural variety, but all remote, Olympian, lording the uncontested power of the local despot over the figuratively (and, here, literally) enslaved populace. Thus, our conundrum implies the requirement of Citadel as signature of despots with a monopoly on force and at the apex of a bureaucracy that functions
© Cambridge University Press
Table of Contents
1. Discovery; 2. Transformed landscapes; 3. Accommodation; 4. Excavation; 5. Surveying the hinterland; 6. Comparative urban landscapes.