The Horse and Slave Trade Between The Western Sahara and Senegambia

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/248715418
The Horse and Slave Trade Between the Western Sahara and Senegambia
Article in The Journal of African History · July 1993

DOI: 10.1017/S0021853700033338
CITATIONS READS
12 211
1 author:
James Webb
Colby College
49 PUBLICATIONS 325 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
The Guts of the Matter: A Global History of Human Waste and Infectious Intestinal Disease View project
All content following this page was uploaded by James Webb on 24 November 2018.
The user has requested enhancement of the downloaded file.

The Horse and Slave Trade between the Western Sahara and Senegambia
Author(s): James L. A. Webb Jr
Source: The Journal of African History, Vol. 34, No. 2 (1993), pp. 221-246
Published by: Cambridge University Press
Stable URL: http://www.jstor.org/stable/182427
Accessed: 26-10-2016 23:37 UTC
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms
Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to The
Journal of African History
This content downloaded from 137.146.217.180 on Wed, 26 Oct 2016 23:37:57 UTC
All use subject to http://about.jstor.org/terms
Yournal of African History, 34 (1993), pp. 221-246 22I
Copyright C 1993 Cambridge University Press
THE HORSE AND SLAVE TRADE BETWEEN THE

WESTERN SAHARA AND SENEGAMBIA
BY JAMES L. A. WEBB, JR
Colby College, Waterville
AT least from the fourteenth through the sixteenth century, the trades in
war-horses from North Africa and from the southern frontier of the western
Sahara played critical roles in the political economies of the Malian and Jolof
empires, and from the sixteenth to the mid-nineteenth century in the
political economies of smaller, cavalry-based states across the western
savanna.1 Following the onset of an increasingly arid period beginning in the
first half of the seventeenth century2 the zone of desert horse-breeding was
pushed south and, through cross-breeding with the small disease-resistant
indigenous horses of the savanna, new breeds were created. Although the
savanna remained an epidemiologically hostile environment for the larger
and more desirable horses bred in North Africa and along the desert fringe,
Black African states continued to import horses into the late nineteenth
century.
The fundamental problem with the importation of North African (or
Saharan) horses was that they lacked immunity to the disease environment
of the savanna and suffered extremely high mortality.3 The principal killer,
although certainly not the only one, was trypanosomiasis.4 One solution was
1 The author would like to thank George Brooks, Robin Law, Francois Manchuelle,
David Robinson, Jean-Loup Amselle and the participants in his EHESS seminar, and the
panelists and audience at the Panel on the Political and Military History of the Senegal
River Valley at the I99I African Studies Association annual meeting for their comments
on earlier versions of this paper.
2 For a periodization of climatic change in western Africa which postulates a wetter
period beginning in 15oo and lasting until I630 and a drier period from I630 to i86o, see
George E. Brooks, 'A provisional historical schema for Western Africa based on seven
climate periods', Cah. Ft. Afr., xxvi (I986), 43-62.
M Many authors have assumed, probably correctly, that the principal threat to the
imported horses was trypanosomiasis. There were, however, other equine diseases of the
savanna including tetanus, piroplasmosis, typho-malaria, glanders and lymphangitis that
contributed to the decimation of this horse stock. Surprisingly, even for the indigenous,
self-reproducing horse populations of the upper Senegal-Niger basin, in the early
twentieth century the average lifespan was a mere eight years. For a discussion of the
various maladies which affected horses in the late nineteenth and early twentieth
centuries, see C. Pierre, L'elevage dans l'Afrique Occidentale Franfaise (Paris, I906),
235-55, and Jacques Meniaud, Le Haut-Se'ne'gal-Niger, tome II: Ge'ographie economique
(Paris, 1912), 122-33.
4 The early evidence suggesting trypanosomiasis is as follows. Cadamosto reported that
horses imported into the Wolof country between the Senegal and Gambia rivers could not
survive 'because of the great heat' and because they were afflicted with a disease which
made them grow 'so fat that ... they are unable to make water and so burst,' according to
Robin Law a description which suggests symptoms caused by Trypanosoma brucei. Robin
Law, The Horse in West African History (Oxford, I980), 78. Inferential evidence may be
222 JAMES L. A. WEBB
to interbreed: over time, the Barbary horse was thereby able to pass on, in
some measure, the virtues of size and speed while the WVest African pony
contributed a measure of increased resistance to trypanosomiasis.5 The
resultant mixes were not particularly attractive aesthetically but were
serviceable and often praised for their hardiness and spirit.' Another solution
was to continue to import, even in the face of devastating mortality. The
logic here lay in the superior equine qualities of the Saharan and North
African breeds. Their greater height at withers translated directly into
greater speed. Their greater weight meant that they were capable of carrying
greater burdens over longer distances. Their larger overall size projected
more fearsome power. All were important for the smaller cavalry-based
states of the seventeenth, eighteenth and nineteenth centuries. For the slave-
raiding of entire villages, mounted warriors typically surrounded a settle-
ment, then burned it, and during the attack ran down on horseback those
who sought to escape. Captives were then tied together in coffles and
attached to the tails of the warriors' horses. For small-scale raiding into
agricultural fields, warriors needed to strike quickly, to stuff smaller children
into sacks and tie them on the horses' backs and, if exigencies permitted, to
abduct the larger children and the adults as well, and then to flee quickly in
order to escape the wrath of the raided community. And, although horses did
not figure as a direct assault force in large-scale conflict between states, the
cavalry, held in reserve, did carry out raids and pillage when an enemy force
dispersed in flight.
The imported horses were generally exchanged for slaves, and the link
between imported horses and exported slaves was direct. Horses were used
in state warfare which produced large numbers of prisoners of war who were
then sold into the Atlantic, Saharan and North African markets, as well as for
predatory pillaging which likewise produced slaves for export. In the
Atlantic, Saharan and North African trades there was an internal dynamic
element. Horses did not live long in the savanna, and slave mortality within
the desert and on the trans-Saharan crossing was high. In the Atlantic sector,
particularly during the eighteenth and early nineteenth centuries, slaves
were sold, in part, for guns and gunpowder, which facilitated more warfare
and raiding. Thus, the horse and slave trades fed upon one another and
reinforced the cycle of political violence.
Many historians have assumed that the horse and slave trade into
Senegambia was a principal feature only of the early centuries of the history
of desert-savanna trade. Robin Law, in his pioneering study The Horse in
West African History, suggested that the era of the dominance of the horse
adduced from climatic change. Over time increasing aridity caused the southward
extension of the desert edge. Drier climate meant less underbrush and thus far less
favorable conditions for the propagation of the tsetse fly.
5 Some early writers drew distinctions between Arab and Barbary breeds. Students of
the biology of the horse draw no such distinctions, arguing that the Arab and Barbary are
one and the same. In this paper, the terms Arab and Barbary are used interchangeably.
The direct evidence on horse mortality in West Africa before the twentieth century for
other than pure Barbary horses is fragmentary and vague. Law, Horse, 76-82.
6 See, for example, Frederic Carrere and Paul Holle, De la Senegambiefranfaise (Paris,
i855), 74.
THE HORSE AND SLAVE TRADE OF SENEGAMBIA 223
and slave trade from the Sahara ended in the sixteenth century.7 Philip
Curtin, in his landmark Economic Change in Precolonial Africa, documented
an apparent collapse in horse prices and inferred that from the late
seventeenth century the horse trade from North Africa or from the Sahara no
longer played a major role in the political economy of Senegambia.8 Other
historians of Senegambia have maintained that the Atlantic slave trade was
the dominant influence on the political economy of the coastal Wolof states
at least from the second half of the seventeenth century and have paid scant
attention to the horse and slave trade.9
This article will present evidence from a broad range of historical sources
to demonstrate the continued importance of the horse and slave trade
between the western Sahara and Senegambia throughout the eighteenth
century and into the second half of the nineteenth century. It will argue that
the horse and slave trade was fundamental to the success of the smaller
cavalry states of the savanna and that the volume of slaves exported into the
western Sahara from the coastal Wolof states of Waalo and Kajoor (and
probably from the other Black African states of northern Senegambia) was
quite possibly larger than the volume of slaves exported into the Atlantic
slave trade.
THE CAVALRY REVOLUTION IN SENEGAMBIA
The indigenous Senegambian pony was small, between 0-95 meter and I
meter in height to the withers. It was, in fact, too small for many purposes
to which the full-sized mounts could be put, including carrying warriors on
its back. By contrast, the Arabian and Barbary horses introduced from the
Sahara and from the Maghrib- and by the Portuguese in the early centuries
of European contact - were much larger, ranging in size from I 40 to I-45
meters and from I-42 to I-48 meters respectively.'0 Interestingly, although
Robin Law argued that horse-breeding was established in the Senegambia by the end
of the sixteenth century, but he based this argument solely on the fact that the Portuguese
stopped importing horses at this time. Law, Horse, 29-30, 49-53.
8 Philip D. Curtin, Economic Change in Precolonial Africa: Senegambia in the Era of the
Slave Trade (2 vols.) (Madison, Iy75), i, 222. Curtin noted that high prices were
occasionally paid for exceptional animals and likened this to European and American
patterns of horse-ownership.
9 Among the most influential views has been that of Boubacar Barry, expressed in his
study of Waalo before French conquest and in his more recent synthesis of Senegambian
history. According to Barry, the trade in horses and in slaves from the Sahara and North
Africa was in some measure in competition with the Atlantic trade for slaves until the last
quarter of the seventeenth century, when it was eclipsed by the failure of the jihad of Nasir
al-Din. For Barry, the failed jihad represented the victory of the desert warriors over the
desert clerics. It disrupted the flow of trans-Saharan trade and allowed slaves produced
in warfare and in pillage to be funnelled toward entrepots on the Atlantic coast, rather
than north to the Maghrib. Boubacar Barry, Le Royaume du Waalo (Paris, 1972), 157-8,
and La Se'negambie du XVe au XIXe siecle: traite ne'griere, Islam, et conque'te coloniale
(Paris, I988), passim. See also Jean Suret-Canale, 'The Western Atlantic coast i6oo-
i8oo', in J. F. A. Ajayi and Michael Crowder (eds.), History of West Africa, (2 vols.)
(New York, 1971-4), i, 387-440; Charles Becker and Victor Martin, 'Kayor et Baol:
royaumes senegalais et traite des esclaves au XVIIIe siecle, Revue Franfaise d'Histoire
d'Outre-Mer, LXII (1975), 270-300.
10 Georges Doutressoule, L'dlevage en Afrique Occidentale Franfaise (Paris, 1947),
238-9.
the indigenous pony was established long before the introduction of the
Arabian and Barbary horses, the Wolof words for horse are derived from
Arabic, indicating that in the Senegambian corridor the imported horses
were indeed a breed apart. By contrast, in the Malian savanna, the terms for
horse have local language roots,1" perhaps suggesting a longer-established
pattern of horse imports, perhaps before the establishment of Arabo-Berber
influence in the Adrar. As further evidence of a distinct equine history in
northern Senegambia, one might cite the Senegambian adoption of the Arab
saddle, rather than the Bambara, Masina or Mossi saddles of western Sudan,
suggesting that formative influences came from the north rather than from
the east and that these influences arrived after the establishment of Arabo-
Berber hegemony in the horse trade. This linguistic and cultural evidence
would also argue for a trade in horses, perhaps bred in the Adrar, from the
Sahara into Mali that pre-dated the cavalry revolution in Senegambia.
At least since the fourteenth century, Barbary horses had been associated
with Sudanic imperial authority. These imperial horses appear to have been
imported both from North Africa and from breeding grounds at the desert
edge.'2 We know little about the supply of horses from North Africa in this
early period, apart from two brief comments: one by the Portuguese
Valentim Fernandes, writing in I506-7, to the effect that Sanhaja traders
sold horses from Safi (in Morocco) to Jolof in exchange for slaves;'3 and
another by the geographer Leo Africanus, writing in the mid-sixteenth
century, concerning a community in the Moroccan Sahara that was involved
in selling horses purchased from the kingdom of Fez into Black Africa.'4
Information about the supply of Saharan-bred horses is not much better. In
the wetter period I 500-I630, the southern edge of the western Sahara was far
to the north of its present location, probably approximately at the latitude of
the Adrar plateau. Although the origins of the horses imported into the
savanna cannot be established with certainty, linguistic evidence suggests
that Adrar was an important breeding ground. The name Shinqit means
'springs of the horses' in Azayr,'5 and this would argue for horse-breeding
there, before the consolidation of Bidan (Hassaniya Arabic: White) ethnicity
and before the establishment of Bidan hegemony. Similarly, Wolof linguistic
evidence points to the importance of the Adrar, and specifically to the
caravan town of Wadan, in the horse trade: the Wolof term for mare is fas
" Law, Horse, 6-7.

12 Humphrey J. Fisher, "'He swalloweth the ground wvith fierceness and rage": the
horse in the Central Sudan, I. Its introduction', J. Afr. Hist., xiii (1972), 369-88, and
'.. Its use', J. Afr. Hist., XIV (I973), 355-79; Law, Horse, passim.
Valentim Fernandes, Description de la cote d'Afrique de Ceuta au Senegal
(I506-1507), eds P. de Cenival and Theodore Monod (Paris, 1938), 71, cited by Jean
Boulegue, Le Grand Jolof (XIIIe-XVIe siecle) (Paris, I987), 88.
14 Leo Africanus, Africa (London, i66o), Book VI: 'Of Numidia', reproduced in
Francis Moore, Travels into the Interior of Africa (London, I738), 'Translations from
Writers', 64. The English translation is at considerable variance with the more
authoritative French translation. Compare with Jean-Leon L'Africain, Description de
l'Afrique, trans. by A. Epaulard and annotated by A. Epaulard, Th. Monod, H. Lhote
and R. Mauny (2. vols.) (Paris, I956), ii, 433.
1 See the text by Sidi Abdallah b. al-Hajj Ibrahim in H. T. Norris, 'The history of
Shinqit, according to the Idaw 'Ali tradition', Bulletin de l'I.F. A.N., s6rie B, xxiv (i 962),
399.
wadan, or 'horse from Wadan'.'" In addition, Tukulor tradition indicates

that during the fifteenth century there were large numbers of horses in the
Hawd, and this is supported by Fernandes' account."7
To the west of Mali, the trade in horses had also been important to the
political cohesion of the Jolof Empire, perhaps since its inception. The
evidence for the patterns of horse supply in this early period is not extensive,
but it does indicate that the trade in horses for slaves was not centralized.
Networks of desert traders not only operated at the heart of empire but
fanned out to the courts of petty Black princes. In the mid-fifteenth century,
for example, Cadamosto noticed Berber zwaya of the Tashumsha who had
insinuated themselves into the courts of the Wolof aristocracies in the
western provinces of Jolof, where they enjoyed prestige through handling the
horse and slave trade."8 Some of these horses imported to the provinces of
Jolof were in turn drawn toward the center of empire through the collection
of tribute. The African provincial elites seem to have supported themselves
by slave-raiding in villages within their spheres of influence and by warfare
against neighboring groups and to have expressed their allegiance to the
Buur ba Jolof by sending horses and slaves and other goods to the imperial
court. Indeed, the fact that rulers required their subject populations to pay
tribute in horses underlines the fact that there were multiple routes of supply
from the desert into the savanna.19
In the second half of the fifteenth century, Jolof began to import large
numbers of horses. Fortunately it is possible to date this change with some
precision. In the mid-fifteenth century, Cadamosto noted that there were
'very few' horses in Jolof;20 of these, only a few were bred locally in Jolof,
the rest being imported.21 But by the early sixteenth century, according to
Fernandes and Duarte Pacheco Pereira, the Buur ba Jolof had a cavalry of
either 8,ooo or io,ooo horse, although these figures may refer to the larger
Wolof region, including Waalo, territories lost to desert forces in the Gibla
of the south-western Sahara, Kajoor, Bawol and Jolof.22
The massive build-up of Jolof cavalry by the early sixteenth century was
momentous. It transformed the political geography of the western savanna;
other polities struggling against the dominance of the center were obliged to
follow suit. This revolution of the horse-cavalry states apparently brought
16 Mission Catholique de Saint Joseph de Ngazobil, Guide de conversation fra

wolof (9o7; repr. Paris, I987), 63.
17 Fernandes, Description, 7I.
18 Gerald Roe Crone (ed. and trans.), The Voyages of Cadamosto and Other Doc
on Western Africa in the Second Half of the Fifteenth Century (London, I937), 78.
19 Ivana Elbl, 'The horse in fifteenth century Senegambia', Int. Y. Afr. Hist. Studies,
xxiv (i99i), 98-9.
20 Crone, Voyages of Cadamosto, 30, 33, cited in Law

21 Elbl, 'Horse', 93.
22 Fernandes, Description, 7, and Duarte Pacheco Pereira, Esmeraldo de situ orbis, trans.
Raymond Mlauny (Bissau, I956), 5I, cited by Law, Horse, II, 52-3. Evidence for the
scarcity of horses in the desert itself in the fourteenth century can be found in Ibn
Khaldun's Histoire des Berberes (Paris, I968/69), 33 I-2. In discussing the Berbers of the
western Sahara, Ibn Khaldun remarks, 'Only a few horses are to be seen amongst them'.
A passage from this work by Ibn Khaldun appears in H. T. Norris, The Arab Conquest
of the Western Sahara (London, I 986), 28-9. Jean Boulegue has suggested that the figures
may refer to the larger Wolof region. See his Le Grand Yolof, 48, 72-3.
common cultural elements to the broader region. The Black Africans of

northern Senegambia and western Mali, as wvell as the Whites of the western
Sahara, adopted the Arab stirrup.23 This allowed the warrior to manipulate
weaponry such as pikes and spears with greater force, using leg and upper
torso strength more efficiently than a bareback rider. With the adoption of
the stirrup the horse came to be more fully exploited as an animal of war and
predation. It allowed for the 6lite of sedentary states to exercise the same kind
of dominance over agricultural communities in the savanna and at the forest-
edge that horse- or camel-riding nomads could exercise at the desert edge.
It is at least possible that a substantial number of the horses in late
fifteenth- and early sixteenth-century Jolof were bred locally. But judging
from later observations of cavalry forces in Waalo in the seventeenth century,
when increasingly arid conditions more favorable to horse-breeding obtained
and still most horses were imported, it is highly likely that the majority of the
horses in Jolof were likewise imported. These animals had no immunity to
savanna diseases and must have suffered the staggering mortality of later
similar imports. Thus, it seems that the Portuguese contribution to the
build-up of the Jolof cavalry state could not have been decisive and may in
fact have been of minor significance. The majority of the imported horses
must have been either led overland from the distant Maghrib by desert
traders or sold off the desert frontier.24
About the last decade of the fifteenth century, a Fuulbe group from
western Mali interposed itself between Jolof and its source of desert
remounts. Led by Koli Tengella, the Fuulbe were successful in warfare to
the north of the Senegal River in the region of the Hawd. According to the
Tukulor chronicler Sire Abbas Soh, the Fuulbe were able to seize more than
40,000 horses of the best quality and to control a massive cavalry.25 Whatever
the number of horses involved, the control over cavalry seems to have been
critical. By about 15 10 the Fuulbe had succeeded in dominating the middle
valley of the Senegal river. Shortly thereafter, probably between 1530 and
1550, Jolof's control over her provinces dissolved completely. In the
resulting power vacuum, Fuuta, the Fuulbe state on the middle Senegal, was
able to expand its own sphere of influence. It dominated the agricultural
groups to the north of the river, and Fuuta began to collect tribute from
23 Pierre, L'e,levage dans l'A.O.F., 56-7. Pierre indicated that there were three types
stirrups found in the A.O.F.: the Arab stirrup, the Hausa stirrup and a stirrup used by
the horsemen of Masina which resembled the French military issue. On the introduction
of the stirrup to West Africa, see Law, Horse, 9I.
24 The argument that the large horse-cavalries noted by early sixteenth-century
Portuguese observers must have been supplied by desert traders first appeared in James
L. A. Webb, Jr., 'Shifting sands: an economic history of the Mauritanian Sahara,
I500-i850' (Ph.D. thesis, The Johns Hopkins University, i984), 88-9i. During the late
fifteenth century, Portuguese traders established themselves along the Saharan and
Senegambian coasts. These traders were closer to the Atlantic provinces than to the
metropolitan center of empire, and over the course of the century the local elite traded
slaves and local produce to the Portuguese in return for the horses that the Portuguese
stabled aboard their ships, and this trade may have helped strengthen the military position
of the coastal provinces in relation to the core of the empire. Jean Boulegue, Le Grand
Jolof, I 55-73. Recently, Ivana Elbl has argued that this horse trade could not have been
decisive in the shift of political power taking place in fifteenth-century Senegambia. Elbl,
'Horse', 99-I03.
25 Sire Abbas Soh, Chroniques du Fouta se,negalais (Paris, I9I3), 26.
\Vaalo down-river and from Kajoor to the south-west. As Jolof lost control
in the middle valley, the coastal provinces broke away conclusively.26 The
new horse-cavalry states emerged from the political disintegration of empire.
Those near the desert edge, in particular, needed to control large cavalries to
deter aggression from hostile neighbors. Toward this end, they imported
large numbers of Barbary horses and began to develop the new crossbreeds.
CROSSBREEDING, HORSE MORTALITY AND SAHELIAN CAVALRY
Over centuries, the pony breeds of Kajoor and Bawol, known as the MPar
and the MBayar respectively, were improved through crossbreeding prin-
cipally with the Barbary horse of the Sahara and the Maghrib. This
crossbreeding produced the smaller, less aesthetically pleasing but still
serviceable horses of Senegambia. But this was a long process indeed. By the
mid-twentieth century the MPar, for example, had been successfully bred up
to a height of 125 to 1 35 meters. But although this Kajoor horse was praised
for its exceptional endurance and hardiness, by French judgment the MPar
was still too small for any effective military use. And there was no way to
improve the breeds further using local stock. A French government mission
at the end of the nineteenth century sent to improve the MBayar breed of
Bawol, for example, found that it was impossible to do so with local stallions.
Stud horses had to be brought in, as they had been during the long centuries
of the horse and slave trade, from North Africa.27
By the late nineteenth century, to the north along the Senegal was another
spin-off of the desert horse trade known as the cheval du fleuve or river horse.
Colonial specialists considered it to be a degenerated sub-species of the
desert breeds. So did the desert herders, who referred to the horses as haratin
(freed slaves).28 The river horse was larger than the MPar or the MBayar and
enjoyed prestige in the river regions. But, like its desert progenitors, it too
was susceptible to trypanosomiasis, and this confined the expansion of the
river horse to the river region, where presumably it was pastured in the Ferlo
and in the desert steppe during the unhealthy rainy season. Colonial efforts
to cross the river horse with the MBayar came to nought.29
The pattern of horse-breeding that was found in the savanna lands of the
French Soudan was markedly different from that along the coastal corridor
of the Wolof states of Senegambia. In the French Soudan, the importation
and breeding of the Saharan or Maghribine Barbary horses were much more
successful. Even so, the horse stocks there tended to deteriorate in the
epidemiologically harsher savanna environment. Colonial veterinarians
charted the progressive degeneration of the pure Barbary horse of the Hawd
through the horse herds of the Sahel and those of the western Sudan, eastern
Senegambia and the middle and upper Senegal rivers. By contrast, in
western Senegambia crossbreeding produced more stable new breeds that
26 Boulegue, Le Grand yolof, 155-73, esp. I56-62.

27 Doutressoule, L'elevage en A.O.F., 253, 26o.
28 Paul Dubie, L'e'levage en Mauritanie (Memoire no. 56I, Centre de Hautes Et
d'Administration Musulmane, January 1937), 17. In Dubie's system of categorization the
horse of Fuuta was the MVbayar. Pierre identified eight principal 'families' of the river
horses. Pierre, L'e'levage, 30.
29 A. Cligny, 'Faune du Senegal et de la Casamance', in Dr. Lasnet, A. Cligny, Aug.
Chevalier and Pierre Rambaud, Une mission au Se6ne'gal (Paris, 1900), 278-80.
were able to reproduce themselves. The difference probably lay in the

genetic differences of the small, distinct pony herds of the Senegambian
Wolof states, apparently the only such horse stock to the west of Timbuktu.30
The supply patterns of the desert horse trades have left few traces. Late
seventeenth-century sources leave the impression that horses were raised to
the north of Waalo (perhaps in the northern Trarza or Inchiri), although
desert oral traditions give no indication that horses were ever bred in the
Trarza or Inchiri. By the eighteenth century or so, most western Saharan
horses exported to Black African markets were probably raised in the Tagant
and Hawd regions by White groups who themselves exercised a formidable
military capacity, and this pattern held into the nineteenth century and
beyond. By the mid-nineteenth century at least some of the horses of the
Trarza emirate were imported from the Tagant.3' The largest concentrations
of desert horses, however, seem to have been in the Hawd. There, the largest
breeders, particularly the Mashduf, are said to have raised six breeds and had
herds of hundreds each.32 By the twentieth century, French sources indicated
that the Hawd (cercle du Sahel) was a center of horse-breeding and that there
were additional sources of supply in the western Adrar, the Tagant and the
Awker.33
Even though conditions in the savanna became considerably more arid
beginning in the early seventeenth century, the mortality of the imported
pure Barbary horses does not seem to have improved. Evidence from the
period of early French colonial rule is persuasive. During the conquest of the
French Soudan and until the very end of the nineteenth century, the French
30 Doutressoule, L'e'levage en A.O.F., 240. The categorization of the pony herds of

Senegambia as discrete from those of western Mali is somewhat contested. Elbl follows
Hellmut Epstein's argument in his work The Origin of the Domestic Animals of Africa (2
vols.) (New York, I97I), that all of the small horses in the western savanna came from the
same stock. Epstein in his accomplished survey was, however, simply synthesizing more
detailed studies, and for the Senegambia and French Soudan the sources cited are the
works of Doutressoule and Pierre, who held that the horse stocks of the Senegambian
corridor and the Malian savanna were distinct.
31 Archives Nationales de France, Section Outre Mer [hereafter ANFSOM], Senegal
et dependances II, dossier 2: Notes sur le fleuve et la Colonie du Senegal recueillies en
I817 et i8i8, M. Laplace.
32 Both desert and savanna peoples attached great prestige to the ownership of horses.
In the desert, this was most evident in the sale of prized horses. It was common at these
sales to record the horses' pedigrees, figured matrilineally, in the bill of sale, and the better
horses were so expensive that negotiable shares of ownership rather than the horse itself
were sold. Partial ownership might entitle the purchaser to ride the horse into battle or
in a raid, and to partial ownership of the foal. For exceptional horses, a share of one-
quarter or one-eighth could sell for as much as one hundred milch camels. Interviews
with Muhammad Ahmad wuld Mshiykh, 22 February I98I, at Nouakchott, 3-4, and with
Mokhtar wuld Hamidun, i9 December I98o, at Nouakchott, 4; Law, Horse, 46. At these
prices, the best horses were clearly a luxury and an emblem of nobility. But in principle,
even at exorbitant prices, the skilful warrior could recoup his investment in one or more
successful raids.
3 Pierre, L'eevage, 35; Doutressoule, L'elevage en A.O.F., 243. By the twentieth
century, the Moors of the Hawd were selling colts to the Black African communities in
the upper river region. Farther east, the pattern was reversed. Sedentary communities
sold horses to the Twareg, Tukulor and Fuulbe, who raised few horses of their own.
G. Doutressoule, L'elevage au Soudan Franfais (Paris, 1948), 220.
colonial cavalry rode only on Arab horses imported from Algeria. This
presented a great problem to the French, because the horses died out rapidly.
TFhe figures for horse mortality in the i 88os in the French Soudan are
indicative of the severity of the problem (see Table i). Compounding the
problem was the fact that the disease environment for animals was changing,
in part as a result of the importation of new breeds. Epizootics of 'horse-
sickness', referred to in the veterinary literature as a bilious typhoid infection
and apparently quite distinct from the familiar trypanosomiasis (often
referred to as a typho-malaria), were first noticed by the French in the i88os.
It affected local horses as well as the imported Barbary and French breeds.34
In I898-I900 the colonial service decided to purchase horses from the desert
traders. They bought I2 to 15 horses per year brought down from the Adrar
or from Morocco. But this initiative failed as well. A solution to the problem
in the French Soudan was to purchase local horses there. These were large
enough for the purposes of the military cavalry, but they were apparently still
quite vulnerable to disease. The French estimated that they had to replace a
third of their horses annually.35
The high mortality of the desert horses clearly posed enormous problems
for the horse-cavalry states of the western savanna. The early eighteenth-
century traveller Francis Moore made an interesting observation about the
internal dynamics of the Saharan horse trade in the Gambia that bears on this
issue. He stated that the sale of desert horses below the Wolof kingdoms was
brokered by the Wolof themselves and that he had seen no mares in his
travels in the region:
The Generality of Horses in this River, are brought from the borders of Barbary;
but as the Grand Jolloiffs are nearest them, they buy them up, and reap an
Advantage, by selling them to the Mundingoes and Mahometans. They never sell
Mares, so that in all the Time I was in Gambia I saw no more than one Mare, and
she was brought ... from Cape de Verde Islands.36
Several historians have taken Moore's comments to mean that the horse
traders were able to keep horse prices artificially high by refusing to sell
mares. The importation of stallions, in this view, would have been critical in
maintaining dependence upon the areas of horse supply.37 The model,
though not explicitly stated, seems to be that this monopoly was eventually
broken by the importation of mares. The Wolof linguistic evidence is
Marcel Leger and L. Teppaz, 'Le "Horse-Sickness" au Senegal et au Soudan

FranSais', Bulletin du Comite' des Etudes Historiques et Scientifiques de l'A.O.F., V (I 922),
219-40. 3 Pierre, L'evage, ' o, 84.
36 Moore, Travels, 63. Horses from the Cape Verde islands were exported to
Senegambia from the late fifteenth century into the first half of the eighteenth century.
See George E. Brooks, Landlords and Strangers: A History of Western Africa, I000-1630
(Boulder, 1992), ch. iO. But there is little or no evidence of the volume of horses exported
from the islands, and the trade was apparently a minor one. See T. Bentley Duncan,
Atlantic Islands: Madeira, the Azores, and the Cape Verdes in Seventeenth-Century
Commerce and Navigation (Chicago, 1972).
3 Curtin, Economic Change, i, 222; George E. Brooks, Western Africa to c. I86o A.D.:
a Provisional Historical Schema Based on Climate Periods (Bloomington, I985), 82-3. In
fact, Moore's account does not indicate whether desert suppliers provided almost
exclusively stallions for the southern trades, only that the Wolof supplied stallions to their
clients to the south.
Table i. Annual horse mortality of the colonial cavalry in the French

Soudan
Year I 882-3 I 883-4 I 884-5 I 885-6 I 886-7 I 887-8

Percent 92 87 95 86 73 75
Source: C. Pierre, LYlevage dans l'Afrique Occidentale Franfaise (Paris, i906),

83. The high mortality of Arab horses was generally attributed to an unhealthy
climate.
equivocal. As mentioned above, the Wolof term for mare translates as 'horse
from Wadan', while the Wolof for white Arab horse is rendered 'Arab male
horse'.38 The fact that interbreeding was both possible and apparently
widely practiced, however, casts a rather different light on the dominance of
stallions among imports into early Senegambia. A stallion put to stud could
produce more crossbreeds than a mare, and for just this reason stallions were
able to command high prices. The importation of stallions into zones of
endemic trypanosomiasis would have made economic sense. Most imports
would not survive for a full year, and thus the possibilities of breeding pure
Barbary horses were practically non-existent. The viability of the crossbred
fetuses in utero would have been greatly enhanced if they were carried in non-
Barbary mares. In addition, stallions were larger and more powerful than
mares and thus more desirable for military purposes.39
Seventeenth-century observers drew sharp distinctions between locally
bred horses and imports from the desert edge and from the Maghrib. The
different horse breeds were understood to have discrete capabilities, and
consequently they brought vastly different prices. According to Louis
Chambonneau, in the mid I670s in Waalo there were both imported and
locally bred horses available for sale, whose prices ranged from a low of one
slave up to ten or 15 slaves, indicating the gap between savanna breeds and
the best of the imported horses. Other late seventeenth-century observations
confirm the high prices that desert mounts would bring. Jean Barbot noted
that horses were commonly bought from desert traders and that some of
them cost ten to I2 slaves apiece. La Courbe, writing of his journey up the
Senegal river in i 685, stated matter-of-factly that 'a good horse' would bring
25 slaves; and J.-B. Labat, drawing from the work of Andre Bruie and others,
noted that desert people 'have horses of great beauty, [and] they are valued
at fifteen slaves apiece'. These were not exceptional horses collected by avid
horse-fanciers. These were the war-horses sought by the Black African
princes and heads of state. It was these horses which enabled the Brak of
Waalo 'to make frequent excursions into the dominions of his neighbours, to
get cattle, slaves or provisions'.40
38 Mission Catholique, Guide de conversation, 63.

39 In the mid-twentieth century, Doutressoule noted that among sedentary peoples in
the A.O.F., stallions were reserved principally for the chiefs. L'levage en A.O.F., 65.
40 Chambonneau, in Carson I. A. Ritchie, 'Deux textes sur le Senegal (I673-1677)',
Bull. de l'I.F.A.N., serie B, xxx (I968), 332-3; Jean Barbot, A Description of the Coasts
of North and South Guinea (London, 1732), 38, 57-8; Pierre Cultru (ed.), Premier voiage
du Sieur de la Courbe fait a la coste d'Afrique en I685 (Paris, 1913), 126; J.-B. Labat,
Nouvelle relation de l'Afrique occidentale (5 vols.) (Paris, 1728), iii, 68, 235.
At least by the I670s, the military use of horses along the desert frontier
of the western Sahara was well-established. A fragment of Trarza oral data
relates that at the time of the jihad of Nasir al-Din 'every man had a
horse'.41 Similarly, evidence from the south-eastern frontier of the western
Sahara makes plain the fact that there were large desert-side cavalries raiding
the Black African settlements at the end of the seventeenth century. The
advance party sent by the traveller Cornelius Hodges to Tarra in the upper
Senegal river valley in I689 credited the Moors with mounting a cavalry of
40,ooo horse and camels to attack the town.42 And as La Courbe makes
clear, the horse and slave trade was not on its last legs in the late seventeenth
century:
These desert people wear their hair long and dressed in the back, clothe themselves
like the Blacks, always go bare-headed and are armed with long pikes and with
spears; they are great horse-traders and raise a lot of horses which they barter with
the Blacks for slaves who they then will sell later on far away."
The same seventeenth-century sources provide information concerning
the limited extent to which crossbreeding had crowded out the market for
desert imports. By the late seventeenth century, the Brak of Waalo was
credited with having the largest cavalry in the Senegambian region, at least
ten times as large as that of the Damel of Kajoor. This cavalry stood at
between 3,ooo and 6,ooo horse and, according to Chambonneau, most were
imports. Similarly in I697, when Andre Briue saw a display of the Saatigi of
Fuuta's cavalry, both imported and locally bred horses were in evidence. The
large stable of the Saatigi, however, was stocked with pure Barbary horses.44
To the south of the Senegal river states, horse mortality in the late
seventeenth century may have been sufficiently high to prevent cross-
breeding. Prices rose as traders ventured south into the zones of intense
tsetse fly infestation. Barbary horses were much less expensive in Waalo than
in Jolof, for example, and the Jolof consequently had few if any to serve for
war.45
European observations of the cavalries of the various Black African states
are informative and bear upon the question of relative parity in military force
among these states (see Table 2). The figures suggested by European
observers may not be precise, but they are accurate enough to detect a
41 Interview with Mokhtar wuld Hamidun, i9 December I980, at Nouakchot

Harry T. Norris, summarizing the account of MIuhammad al-Yadali, also writes that
during the war of Shurbubba both zwaya and hassani rode on horseback. Norris, Arab
Conquest, 36.
42 Thora G. Stone, 'The journey of Cornelius Hodges in Senegambia, 1689--90',
English Historical Review, XXXIX (1924), 93.
41 Cultru, Premier voiage du Sieur de la Courbe, 146: 'Ceux cy [the desert people]
portent les cheveux longs et tressez par derriere, s'habillent a la mainiere des negres, vont
toujours teste nue et sont armez de longues piques et de sagayes; ils sont grand
maquignons et nourissent beaucoup de chevaux qu'ils troquent avec les negres contre les
captifs quils vont apres vendre bien loing dans les terres.'
4 Labat, Nouvelle relation, iii, 234-5. Gaspard Theodore Mollien noted that the royal
stable of the Damel of Kajoor was stocked with pure Arab horses at the time of his travels
in i 8 i 8. Mollien, L'Afrique occidentale en i 8 i 8 (repr. Paris, I 967), 40-I.
45 Barbot, Description of the Coasts, 39. Somewhat surprisingly, the substitute for the
horse in use in Jolof was the camel, also imported from the desert. Unfortunately,
Europeans rarely commented upon this trade in camels.
cn
4-1
4-1
C, 4-1
cc
u
'4 cn cn
14, 4-1 cn C13 4.1
E
> > Z > cn
4-1
C, C,
cn ct
4-1 4-1 ct cn
'+-. V z
cr ct Ct
cn ct 4--l
a, ct v cn
Ct a 4-1
4--l C: cn1o
j- ct cn ct > C-t ct 4--l
4., cn 0 .4 .- j-- r. 0
:z > - 0 14,
cl ct cn ct cn
ct u
'n ct
u - - 4-1 7
ct v cn cn ct cn E
u -C cn 0 ct
E cn
ct -Ccn
ct 0 cn
ct -z-
ct
E u cn cn
rj 0 u u cn
v cn u
ct
ct u
ctE
4--l cn -s Z v 0 ct
x, .
4-1 x ct
ct ct E cn 0 cn
cn7:
cn0 4--l
m , cn 0
cn V cn
u C
cn cr, I
14. x 0
4--l cn Ct 4..'
CZ
U 4..' ct
ct 0 0 C E
t--4 C 4.1 '41 u
E 0 0 E 7
IZ41)
E
-C
4--, X 4--l cn
-Izt t cn
cn ct a ct >,
E 4--l
I ct
4--, E ct
ct
ct cn
cn ct
E
4--l
4-1
:-z
cn
cn
ct
ct
ct
0 4-1
0 ct
ct 4--l
4--l ct
ct ct cn > i. ct
j- 21 0 ct
0> M>
4--l 4--l cn 0 4-1 Cn ct u
cn cn cn 4-1
(1Z U v u ct 4..' 'I- :
V (1Z cn ct v
CZ 7z cn 0
-z-
r-I0 -
0 v
v m
0 ct cnct
-7-1 0 ct
Z 7 u 00
tt 0 0 -0 C V>
Cn 4-1 ct
03 ;:t z
V ct 4-1 > v 0 cn
cn 4-1 > cn cn ct
cn o 4-1 ct u E cn
Ct 4-1 cn 4-1
> 0 cn
4-1 4-1 03 00 V CZ
V x Ct
u
V 4-1
--
C C,3 v
ct
00 0
u
4-1 t4 - > 0 04-1 0
ct
o3 v v v o3 o3 C;
-C
4) ct -..
ct -= ct
.4-1 4-1 I
C3 v v - X 4-' 4-' 4-1 4-
ct x cn
14. -C 0- "O cn r. = C,3 cn
4-1 Ct cn
ct
0
C
zC,3
r. .- C,3 t4 v
v ct ct
c3 cn
-C
v 4.1 C3 4-1 Cn r. Z; , Z: " -C = -C
cn .- -r
ct v
cn ct -'C" , - . ", v v - ct
r. . ct
r-- 4-
v ,l . r. cn
>. ct 44 -C -C
cn U - ct >, E IE u u
4-1 c3
a, ct m ct
v V>
4-1 -C cl"Or.
:: vct -C
C3
cV >
E 44
ct
t4 cn
Z > C,3 cn CZ
72 0 z r. v ct v v
E c'n u m x U c3 V u C3 ct E E v R
ct ct
ct N 0 0 c3
t4l 4-1 ct tt C) - --O c
C3
ct
C3
tt ct
C3
C3 ct
ct
: - r.
ct
E
0 0
0 0
C
4-1 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 C 0
000 0
0 0 0
0 0 0
0 0
0
4-1
4-1 hc
C, C,
C3 c3 U "C
03
4-1 03 v o3 : r. N
4-1 E cn : 0
0 ct 7 Z
cn 0
rn 4-1 cn
o
03a)
cnct
cncn R
-c : V 4-1 OC)
ct u C) 4-1
ci 7 > rn
4-1 c3 4-1
03 4-1 4-1 03 ct
C) > C) 4- ct
ct ct 4-1
(1) u
0 > 03v
0 4-1
U
4-1
4-1
03 u
v q.) > C,3
4-1
U) -c q.)
4- -0 C,3
U 4-1Uc3
v c3 cn X
bk
r. ct cn ::3 Z Z cn
c3
c3 >
ctct
>
bk U) ct 4-1 cn q.)
C13 . - - v x t o 0 4-1
4-1 7-' u C)
4-1 c3 C, c3
4-1 c3 4-1 c3
C3
v 7c, Z- C3
06
-
c3
0
v
cn
6
0
r-
4- c3 a) a) u c3
4-1 c3
o OC)
4-1 ct c3 03
u U) C3
CZ 4-1 4-1
U Z r -c 4-1
::3 c3 4-1
4-1
c3 a) a) a)
> u a) 4-1 4-1 c3 a)
c3 v C3 a) C3 C) Z
>, -c >,
U
C3 a)
c3 41 C)
c3 a)
Q c3 ;..
4-1 z V C73
- -
u c,3 4-1 C)
cl 0 c3
ct c3 C13 4-1
a) 0 Lr)
-rna) 00 4-1
c3 C3 0. N
a) C3 a) cn m a)
11
c3 C,3 v+-
c3 4-1
a) C,
rn
4-1
u c3 > v " a) C3 c3;
a) v 4-1
v c3 c3 a) z, "O 0
4-1 c3
0 0 v u > a) CZ c3
00
4-1 0
4-1
-c ctU
o3
o bk 0 C,
tc Z - V 14. -
LI; v C,3
C3 C) 4- 03R cn - a)
bt > E :: O 4-1 00
> C, C) C3
a) c3 a)
c3 C,
c3 4- O.
a) c3 a)
c3 4-1 u C,3 E v
C3 j v v rn u
C3 u a) a) C) W 4-; a) u c3 C)
u a) c3 > C13 a)
C3 4-1 C3 a) c3 Lf)
u >> >
4-1 cn ct CZ 0
7 c3 u m a)
a) " z C3 >, 4 C) C)
-c CZ 4- EZI 03 UO 0 C3 v 0 -I 4-1 _C -.t
4-1 W W 0 0 " u
O -V 4-1
U
73 u
n
4-1 oc
C3C,3 4-1
>
c3 c3 C3
4-1 03 4- a) cn c3 CZ
CZ u cn cn
tc
48 cn 4-1 C3
cn 4- 4-1
4-1 c,3 4-1 c3 C) C,3 c3 >
c3 a) U c3 w 4-1
4-1 4-1
c3
Lf) C3 a)
c3
Lf- C,3
a-4
c3 u C'3
z O. 4.1
U
- C, 4-1
00 C,3 C3 -
-to bC
(Z) IZL) Z; c,3
"t
qj- Z, Lf) C3 W v
w z
03
R R (73 R R R c, u 4-1 '(Q'.)
co 0303
v a) (Q,.)'a)
a,4 cn: cn
En ; =
c3
4-1 (z)
4-1
,3
c3 cn 00 cn
C3 z :J
O LrI-
v C,
4,) 4-1 ct
;, CZ cn
0 0 0 0 c3 cn
0 0 0 0 U01Q, cncn
O O O O 0 Lf) 4-1
4- 4-1 C,3 03
Q.1) cn 4-1
C) (1) o3 C3
03 4-1
- z, -> cn
(1) IV - a.4
03 v
V 41 C3 >
cn o3 cn
(:t C3 4-1
C', o o 4-1 03 U
C, p C3 o3 o3
C, r- OC) - OC) cn
00 OC) C, C, - 11- ce cn x 'RC
r- OC) OC) 03
4-1 o3
a-4
significant change in the regional distribution of military power. In the

eighteenth century, Kajoor came to rival and then to supersede the cavalry
state of Waalo as the dominant horse power in the Wolof and Sereer states.
Waalo's cavalry force declined from 3,ooo-6,ooo in the i68os to 2,000 in the
176os. By contrast, the cavalry of the Damel of Kajoor increased from
approximately 300 in the i68os to 2,000-3,000 by the 178os.46
The principal demand for North African and Saharan horses came from
the rulers of the Black African states and their courts. The case of Kajoor is
the best documented. The Damel of Kajoor, for example, had his own stable
of exclusively Barbary horses. These would have been used by the slave
soldiers of the Damel to carry out raids against villages within Kajoor and on
the borders of neighboring states. Nobles and village chiefs apparently also
owned desert horses which conveyed prestige and affiliation with the political
economy of rapine. Thus, by I 68 I, for example, the Damel of Kajoor did not
have a great number of suitable horses under his immediate control, but he
was able to raise 300 horse throughout his realm. By the middle of the
eighteenth century, many more horses were available to the Damel for large-
scale military conflict. The mid-century observer Doumet (1769) thought
that the Damel might be able to raise i,000 horse for state warfare. By the
end of the century, observers credited the Damel with being able to raise
2,000, 3,000 or even io,ooo horse. Under increasingly arid conditions
Kajoor, the Wolof state to the south of Waalo, was able to mount a cavalry
force to rival that of Waalo in the previous century. There can be little doubt
that the nature of inter-state warfare in the region was changing as a result.
COMPARISON OF THE ATLANTIC AND TRANS-SAHARAN SLAVE
TRADES
Simple calculations are useful in giving some rough idea of the magnitude of
the slave-raiding that must have supported the northern Senegambian
cavalries. For the late seventeenth century, if one takes the lower figure of
3,ooo horses in the Waalo cavalry and estimates that only one-half came from
desert traders, and assumes that the vast majority of horses imported from
the desert were new crossbreeds from the desert edge with at least some
resistance to trypanosomiasis (and thus would have had to have been
replaced at a minimum of one third per annum - the replacement figure for
46 ' "Brak" est des plus puissants du pays, il peut avoir 4 a 5000 hommes capables de
porter les armes dont pres de la moitie en cavalerie.' Fonds fran,ais 9557, Bibliotheque
Nationale, Paris, M6moire sur les mines de Bambouc, 30 November I762, cited by Barry,
Waalo, 202. This was, of course, near the end of Waalo's regional military presence.
Additional confirmation of the rise of regional military capability of the state of Kajoor
may be found in a British document from the early nineteenth century. In i8iI
Lieutenant Colonel Maxwell, the Lieutenant Governor of Senegal, related to a com-
mission of inquiry that the present Damel was not as powerful as his predecessors but
that the Damel could still field 5,000 or 6,ooo men in an emergency. One of the Damel's
predecessors was said to have brought 5,ooo cavalry to bear upon a rebellion in Kajoor,
presumably in the eighteenth century, although no date was mentioned. Public Record
Office, London (PRO), CO 267/29 Answers to the Questions proposed to Lt Colonel
Maxwell, Lieutenant Governor of Senegal and Goree by his Majesty's Commissioner for
Investigating the Forts and Settlements in Africa, I January I8II.
locally bred horses in the late nineteenth-century French Soudan), Waalo

would have had to import from the desert 500 horses per annum. Even at the
minimum price of one slave per horse reported by Chambonneau in the
I670s, this would have required exports of 500 slaves per year. An unknown
number of these mounts, however, would have been pure Barbary horses and
would have sold at roughly I 5 slaves apiece. The volume of annual exports
was thus probably considerably in excess of 500 slaves per year, which
should be taken as a minimum figure; the actual volume was more likely a low
multiple of this figure. By way of comparison with the Atlantic slave trade,
approximately sixty slaves per year were sold from Saint-Louis du Senegal
during the i68os.47 These estimates of slave exports suggest that in the early
years of the French presence at Saint-Louis the number of slaves exported
from Senegambia across the Atlantic was a small percentage of the slaves
which passed from Senegambia into the desert. Exports from Waalo alone to
the Sahara and North Africa may have equalled or surpassed the export of
slaves from the entire Senegambia, including the Senegal and Gambia river
basins, into the Atlantic sector.
Along the Senegambian coastal corridor below Waalo, the Wolof states
likewise seem to have been dependent upon the importation of horses
acquired from desert traders. In i686, for example, when the Teefi of Bawol
defeated the Damel of Kajoor's army, he took many prisoners. He needed to
equip a military force. Guns and gunpowder were available through the
Atlantic trade, and horses from the Sahara. The Teefi sold only eighty of the
prisoners of war into the Atlantic trade. He sent the rest north to be
exchanged for horses, to mount his cavalry.48
Across the western Sahel, during the eighteenth century, the importation
of war-horses from the Sahara and the Maghrib continued to be directly
implicated in the political violence of enslavement. But in some respects the
dynamics of the horse and slave trade did change fundamentally over the
eighteenth century, as horse-breeding became a more successful concern in
some regions of Senegambia. This was owing, in part, to the increasing
aridity, which over time eliminated the habitat of the tsetse fly in the
northern zones and pushed the zone of trypanosomiasis infection to the
south. This, in combination with the breeding of new savanna horses that
enjoyed greater immunity to trypanosomiasis than did the North African and
the desert steppe horse, made horses common in northern Senegambia. But
it is also clear that the breeding of horses in the savanna did not eliminate the
" ' Relation du Sieur Chambonneau, commis de la compagnie de Senegal, du voyage

par luy fait en remontant le Niger (juillet i688),' Bulletin de Geographie Historique et
Descriptive, II (I898), 309. The trading relationship between the desert and savanna
peoples was symbolized by the ritual exchange of a horse for a slave that took place
between the Trarza emir and the Brak of Waalo. Cultru, Premier voiage du Sieur de la
Courbe, I 54-5.
48 'A French ship, that happened to be then in the road of Porto Dali, on board which
was Caseneuve [John Casseneuve, a former shipmate of Barbot], who gave this account,
bought 8o slaves of the prisoners of war. The rest of the prisoners the usurper sent
towards the country of the Moors, to be exchanged for horses, to mount his cavalry.'
Barbot, Description of the Coasts, 425. For further evidence on this conflict between Baol
and Kajoor (not mentioned in oral tradition), see Victor Martin and Charles Becker, 'Les
Teeui du Baol: Essai de chronologie', Bulletin de l'I.F.A.N., serie B, xxxviii (1976), 479.
X w z - = U =~~~~~~~~~~ v 0
Ca (L) *- O $ X ? O*< X t 3
E . v R X S W , s 4 E ~~~~~~~~. ? - w
k~~~~~~~c 4- X U C, =n
~~~~~~> cn C) ua -
i c3 c z X X; X; c
? D ~ : t Wt $; ~ : H t
~~~~~~t C;.c; X VX X X h._ Q .

S <4 t;i Q c cX, c ? c c S Q S Q > > > 0 7
0 7
0
0
v 0
> -r. 0 u
r- 14-.
00--4 ;.* .O 00
O
O
V Ln
O 0
0 0
0
0 0
0
0 0-4
Le)
4L 0--4
4.-, ct
4-j >
4-; 0
U. O
4-1
>
4-1 T.
> 4-J
ct
>
t4--l
1421
Ln
> cn C)
00
> "! Lf)
0 -C: -C:
ct
>
>
> >
ct
> > 0 0 0
cl O
Lf) 0
v 0
Lf)
0 0
C,
1
0
0
Ln
> >
ct
0
0 4-1 cl cl N
0 0 0 O rn cn
0 Lr) 0 Lf) Lf) 0
cn Ln
OC) ON 0
OC) OC)
0--4
OC)
Lf)
OC)
Lf) ON 4-J
OC) OC) OC) OC) OC) ct
o o o 0.4 0--4 0-4
(U
t
x OC
ct ,c
CZ CZ
7:1
+4
ct
-C C/) ct 4-J
ct u
7z u
ct
(.t
7z ct 4-4
C) 0
ct 0
4-J
CZ
ct
7z 4-1 OC
4-1
4- cl
ct
CV C
cn C: N cn
03 q) Ct Q
7 u
4-4 7:;
ct Ln t5o q,) CZ
u
t ; -Z
u -- q,) -Q
>1
u Lr)
CZ
4-J
u
-*
C)
4--l M
(ZL)
') (ZL)
7z 00
4-4 4-1 4-1
;Z C,3 m
cN
4-1 U oo w
Lr) 4-J
qo
cl 4-J
4-J
CZ 4-J
4-4
z .
cn 00 -4
7
00 CZ C)
I > 4-J
4-J
4-J Z3 Lf)
:z
U
C3 0c)
cn - ; 4- 4-1 0
z v OC)
ci ci C") co !! '- :
)--q
oc >, Ln
4-J
C's ct u
03 ;.O >1 u14' U
03
0 C;3 0 0 0
0 4.4 0
0
0 0 t4-4 ct C)
Lf)
0 CZ
N0
>
v
;.O
m
V
aq
qs
&)
-4
oc
fn Ln
I - v z CZ ;I
C's
0 Z L) C-Z
O\ biD 4.- ct
*4
-Cn
;-4 Z Z M
- u 0 :z
- c1l) u
411 7
v :z 1-:1, ;t 4-
7 u
00
4-0
4-0 4-J
4-1 Cj Ln
7z
4VJ u 4-J
V t4-*
u 75 CZ
-15 lao 0
ct 0 Z, t
7z Ln 4-J
V cc
Ct CZ U U ct
V
-lz
5
4-1
7Z
:Z U 7 m
CZ
41
ct U*
V Q ct
cn w ct t U I - : ct
Ct cn oc
Q cn -- .
4-1 ct cn
U7z
;..4 V
CZ
, 4- U U U ;"*
U 75 4-,
4-J
cn
;..( co CZ
ct
1-1 75
75 cn
4-
O
cn
<U cn
U
ct ct CZ
> U 4-1 ct
ct
cn U CZ 00 ct
cn cn
)-..4 V : v C. Z -: - U r. CZ
'z
ct - rv
...O cn
U) : ct
4-- .&
ct
C700
;-, m0
4 x
ct C
03 CZ
C m 1U- Ct
v C- r,
4-j 0 7z
cn
cn
ct cn ct
U ct ct ct
ct cn
E
4-1 U
>
75
4-4
> 4--J
0
4-J cn
U 0 -- cn
75 V 7. > 0 4--J
U U CZ
Ct
Ct
4-1
CZ 75 > cn
ct ct ct
cn Z
fl lza
C-" CZ
U >
cn cn CZ Q
v ct Z q 0 -, 4-J
4-1 0 q,) ct
> CZ
ct 1--, --
ctZ ct
C: . - cn 40,
V un > ct
4-1 ct
<0 W 4-J W
1-4 4--J 4-1

41
C13
Ct
4-1
-.fz Z
ct
C13 4-1
CZ
v U ct
J a)
;-,
U Z -15 U ct U 0 q6)
1Z.0 N - t+-4 U V .- C4.) U 7:$
V cn V 00
4-J CZCZ
4-J ::s ct U CZ C ct 4-J

lU -0
CZ
> - t v
U : 0
00 E
ct
U Ct
CZ cn qa ll cn
CZ cn Cn U 4-J ct
CZ qo ct Cr,
ct
w
U 4-j ct
cn Z
r. CZ 'V ct
4-1 4-1 '- U qo ct ct
4.: ryl'.." U C)
Z ;3;4
4-i 4-.J
ct cn
00 Ct t,
CZ ct
CZ CZ 0.4 U 4 +.,cn
Z, P., ct o CZ
C,3 .,- -1 -0 :
4-J ct >
O.,
ct
E
Ct
1, (:)
w .-
t+
t.) C. Z -C.) OC,) O 't x
demand for desert mounts. In 1736, for example, David Langlois de la Bord
noted a continuing demand for desert horses in Waalo:
Few slaves were taken along the river this year, Brak made five or six raids which
produced nothing [for the French], the larger part passed into the hands of the
Moors in exchange for their horses..."
And in Kajoor the royal cavalry horses employed for pillaging, which needed
to be faster than those available to hapless villagers, also continued to be
imported from the desert traders. There is little evidence of the numbers of
horses involved in these raids, other than the observations of Doumet, who
estimated in 1769 that there were 300-400 cavalry used for pillaging within
the realm of the Damel. Most of the mounts would have been the more
ordinary sort available from the desert traders at a price of roughly three
slaves per horse, or perhaps the largest of the new crossbred horses with a
considerable resistance to trypanosomiasis.50
In Kajoor, the Damel was the largest single purchaser of the best Barbary
horses, many of which were probably brought down along a coastal route
from Marrakish.51 These Barbary horses in the mid-eighteenth century sold
4 Archives Nationales de France (hereafter, ANF), C6 i I, Rapport de I4 Juin I736,

David Langlois de la Bord (?) [name only marginally legible]: 'II s'en fait peu de Captifs
en Riviere cette annee, Brak a fait 5 a 6 efforts qui n'ont rien produit, la plus part a passe
entre les mains des Maures pour leur chevaux; nous voulons tenter cette haute saison le
Commerce des Chevaux avec St. Yago et pour peu qu'il apporte de profit, nous le
continuerons pour le partager aupres des Roys negres avec les Maures.' The French
company, however, aborted this initiative, fearing that it was simply a means to increase
opportunities for private trade. French traders, however, were bringing at least a few
horses from the Cap Verde islands into southern Senegambia destined for the Damel and
Teefi at this time. See ANF, C6 io, letter of April I73I, Comptoir d'Albreda.
50 By way of negative evidence, Doumet makes no mention of any horse-breeding by
the Damel. His description of the Damel's military force for state warfare indicates that
a military draft in Kajoor would produce a diverse group of poorly armed conscripts on
foot and that the Damel, by contrast, would be looked after by his cavalry, with himself
at the head (although the cavalry stayed at the rear of aggressive actions). Doumet,
in Charles Becker and Victor Martin, 'Memoire inedit de Doumet (I769): le Kayor et
les pays voisins au cours de la seconde moitie du XVIle siecle', Bull. de l'I.F.A.N.,
serie B, XXXVI (I974), 39.
51 This route was used annually by the Idaw al-Hajj (who were known to the
Europeans as the Darmankour), the zwaya grouping which dominated the trade in gum
arabic. According to an account of the I78os, the Idaw al Hajj moved north to the Atlas
mountains from August to December or January. Gum arabic moved north on these
caravans, which probably indicates that it was a ballast trade; the slaves who made the
trans-Saharanjournev did so on foot. See ANFSOM D.F.C. Senegal 82, no. 75, Memoire
sur la traite de la gomme au Senegal, August I783, M. Eyries. The author of this memoire
reported that the marabouts (Darmankour) went to the mountains of Morocco each year
after the gum trade. Another report of yearly expeditions to Marrakish by the
Darmankour as well as their detailed coastal route was published by M. Baron le Roger,
in 'Resultat des questions adressees au nomme Mboula, marabout maure, de Tischit, et
a un negre de Walet, qui l'accompagnait,' Recueil des voyages et de memoires de la Societe
de Geographie de Paris, II (I825), 6o-i. An anonymous author of a French memoire dated
I 803 indicates that a desert route from Senegal to Morocco was open and in frequent use.
ANFSOM, D.F.C. Senegal et dependances, carton 83, no. I05, Des Peuples qui habitent
les Cotes du Senegal et les bords de ce Fleuve, Des Royaumes sur la c6te de Goree, Cayor,
Baol, Sin et Salum. Further evidence of a coastal route from Wad Nun south to Senegal
may be adduced from Auguste Beaumier, 'Le cholera au Maroc: sa marche au Sahara
at the 'ordinary price' of 15 to thirty slaves and generally must have died
within a year of purchase. Some of the slaves used for this purpose would
have come from the annual tribute that the Damel drew from his kingdom,
which Doumet estimated at 400 slaves and an unspecified (though con-
siderable) number of cattle.52 Of these slaves, perhaps I00-200 were sold into
the Atlantic sector.53 Many of the rest were likely destined for sale into the
desert: the sale of 200-300 slaves would probably have yielded only between
ten and twenty horses fit for the Damel. Other slaves for export were
gathered in pillaging expeditions carried out either on the initiative of the
Damel or on the initiative of other nobles.54 The nobles also needed to own
high-prestige desert horses, and on occasion they undertook pillage for the
explicit purpose of gathering slaves to sell into the Saharan trade.55
Any estimate of the annual volume of slaves which moved north to
Saharan and Maghribine markets must remain somewhat speculative. Much
detail about the sale of horses which were central to the prestige of the
cavalry-based states escaped the notice of (or was shielded from) European
observers. But it is possible that at the time of Doumet's observations (1769)
the number of slaves drawn solely from tribute revenues who were then sold
for horses and exported to the desert sector from Kajoor equalled or
exceeded the number of slaves sold into the Atlantic slave trade. And it is
important to emphasize that this would be only a partial estimate of the slaves
exported north to the Sahara and North Africa. It would exclude slaves
gathered either through pillage by the Damel and his nobles or through raids
by desert forces. And it is certain that the volumes of these additional two
streams of slave exports were sizeable.
Europeans commented frequently upon the pillages committed by the
Damel and his soldiers. Desert raids into Kajoor occasioned less com-
mentary, but it is nonetheless clear that these raids could produce significant
jusqu'au Senegal, en i868', Bulletin de la SociOte' de G#ographie, 6eme serie, iii (i872),
303-4; for evidence of a route from Wad Nun down the Atlantic coast and across the
desert frontier, see the account of the Moroccan merchant Sidi Hamet in James Riley, An
Authentic Narrative of the Loss of the American Brig Commerce (New York, i847), I60-3.
52 Doumet, 'Memoire', 44.
5 James F. Searing has estimated that in the period I760-90 approximately 200-300
slaves per annum were exported from the region of the lower Senegal, through Saint
Louis and Goree. The lower Senegal, for Searing, would include Waalo, Kajoor and
Bawol. James F. Searing, 'Door of No Return' (unpublished book manuscript), ch. 2.
The estimate of I00-200 slaves exported per annum from Kajoor is my own based on
Searing's figures.
Doumet, 'Memoire', 44. According to a mid-nineteenth century account, the Damel
received two-thirds of the plunder from pillaging but only one-half of that gathered in
warfare. Carrere and Holle, De la S#n?'gambie franf aise, 72.
55 As the French administrator Jean-Gabriel Pelletan observed: 'The blacks value
highly the horses of the Moors. I have seen black Princes give up to ten or twelve slaves
for a horse, and make a raid expressly for the purpose of paying for the horse.' Jean-
Gabriel Pelletan, Memoire sur la colonie franfaise du Senegal et ses de4pendances (Paris, An
IX), 55 n. i i. Thus through political violence guns and horses found their way into the
hands of slave warriors and nobles throughout Kajoor. According to a mid-nineteenth
century account, the Damel exercised a sovereign control over the supply of gunpowder,
doling it out to his subordinates just before political violence was to be initiated. Carrere
and Holle, De la S#n#gambie franfazse, 7I-2.
numbers of slaves, particularly from 1760 when desert warriors helped

secure the authority of the Damel and in exchange received the right to
pillage within Kajoor for one month per year.56 And following successful
large-scale slave raids into Waalo in 1775, desert warriors began to raid with
greater impunity in the Wolof states. Unfortunately it is not possible to
determine the volume of slaves taken in desert raids that was destined for
North African markets as opposed to the Atlantic slave trade. Some
eighteenth-century writers described the sale of Black slaves by Bidan to
brokers in the Atlantic slave trade and made little or no comment about the
northward flow of slaves.57 On the other hand, Silv. Meinard Xavier
Golberry, writing from his experiences in Senegal during the mid-I78os,
indicated that the Whites had a major market for the captives in North
Africa:
They [the desert people] are merchants; they travel great distances in the Sahara
in all directions; they pillage for slaves on the banks of the Senegal and the Niger,
and they sell them [the slaves] on the shores of the Mediterranean ...58
The existence of a significant western Saharan and Maghribine demand

for Senegambian slaves casts new light on the Atlantic slave trade from
Senegambia in the eighteenth century. The volume of slaves from the Wolof
states exported into the Atlantic sector decreased during the second half of
the eighteenth century, from approximately 300-400 per year in the period
I700-50 to 200-300 per year in 1760-90, when prices were rising over the
course of the century. This suggests that the critical inputs into the political
economy of slave-raiding during the era of the Atlantic slave trade came not
from the Europeans and that the Black African rulers and princes satisfied
their demand for European guns and gunpowder, and for cloth and other
imported goods, with a smaller expenditure of slaves in the second half of the
eighteenth century than in the first. Indirect confirmation of this price
elasticity of demand comes from the report of a Wolof river captain who
related details of his interview in I775 with the Damel:
The Damel was speaking to me one day [at his compound] in Kajoor. I was saying
to him, 'you don't gather as many slaves for sale as you did previously'. He said
to me 'I am going to explain why. It is because at the present time I receive for one
slave what I used to get for five ;.... He made me see that we were fools to pay I20
bars for slaves, which is the going price.59
56 Interviews with Mokhtar wuld Hamidun, ii April I98I, at Nouakchott, I-2, and
with Ahmad Baba wuld Shaykh, March I982, at Nouakchott, I7-I8; Paul Marty,
L'emirat des Trarzas (Paris, I919), 8o. The period of years during which this right to
pillage (with the accord of the Damel) was in force is not known.
5 See, for example, Dominique Harcourt Lamiral, L'Affrique et le peuple affriquain
(Paris, I789), 237-65; ANFSOM D.F.C. Senegal 82, no. 82, Ctes d'Afrique: Traite des
noires, Par M. de la Jaille, 2 June I784, ff. 27-8.
58 'Ils sont marchands; ils executent de tres-grands voyages dans le Zaarha qu'ils
traversent dans toutes les directions; ils font des pillages d'esclaves sur les bords du
Senegal et du Niger, et ils vont les vendre sur les rivages de la mer Mediterranee...'
Silv. M\leinard Xavier de Golberry, Fragmens d'un voyage, en Afrique (2 vols.) (Paris,
An X), i, 30I-2.
5 ' Le Roy Damel me dit un Jour Chez Luy a Chajort [Kajoor]. Je Luy disois tu ne fais
plus autant de Captifs que Les autres fois, il me fit repondre Je Vais t'Expliquer
pourquoy; C'est qu'apresent Je recois pour un ceque Je recevois autrefois pour Cinq ... il
In fact, the revenues from tribute, pillage and warfare seem to have left
Black rulers at times with a surplus of expendable slaves. This would go
some distance toward explaining the fabulously high price of one hundred
slaves, one hundred cattle and twenty camels that one 'Black prince'
(probably the Damel of Kajoor) was willing to pay for a single, if spectacular,
desert horse.60
From the second half of the seventeenth century through the eighteenth
century, to the east of the coastal Wolof states, the volume of slaves generated
by political violence was probably much greater than along the Senegambian
littoral. Along the south-eastern frontier of the western Sahara, the horse and
slave trade was intertwined with the large-scale political violence launched
from the desert edge and from Morocco which rolled Black African
settlements from the Tagant and saw the rise to military dominance of desert
forces in the Hawd. And farther to the south, from Galam and Bambuk, slave
exports to the Maghrib apparently continued at high volume throughout the
eighteenth century. From Fuuta, however, the massive rupture of north-
bound slaves was greatly reduced by the successful Toroodbe revolution in
the mid-I77os.
IN THE AFTERMATH OF THE ABOLITION OF THE ATLANTIC SLAVE
TRADE
The horse and slave trade did not stop with the abolition of the Atlantic slave
trade. But with the closing of the Atlantic slave market, the slave-raiding
states of the western savanna were left without the means directly to purchase
imported European goods. For the Black African rulers and princes, it was
still possible to obtain guns, cloth and other imported goods through the sale
of slaves, but it was now necessary to sell their slaves to desert traders in
order to do so. The collapse of the Atlantic market thus strengthened the
position of the desert traders who purchased slaves. And it brought
immediate repercussions for the slave-raiding states because the desert
traders were able to exploit their newly bolstered, dominant market pos-
ition.61 Comparing the slave trade of the northern Senegambian states before
abolition with the slave trade after abolition, Lieutenant Colonel Maxwell
testified in i 8 i I:
me fit entrevoir que nous Etions des duppes de payer Les Captifs I20 Barres qui est le
prix actuel Toutes ces raisons se Passerent Chez Luy Dans Un Voyage que je fit en I775.
Lieu situe a 6o Lieues Du Bord de La Mer.' ANF C6 I 7, I2 May I78I, Capitaine Guiof.
A similar view of the behavior of the Damel can be found in ANF, C6 i8, Remarques:
Etat et aperSu des esclaves qui peuvent retirer les Nations de l'Europe de la Cote
Occidentale d'Afrique. I783 [unsigned]: '... celui de Cayor ne se determinant a faire des
pillages que quand il en a de grands besoins.'
60 Pruneau de Pommegorge, Description de la Nigritie (Paris, I789), I6-I7.
61 As the Brak of Waalo wrote plaintively to the French governor Schma
nous a acheter des chevaux dont nous avons grand besoin pour repousser les attaques de
l'enemi, car vous devez le savoir, il nous est impossible de rien faire sans chevaux'. See
the letter from the Brak to Schmaltz quoted in ANFSOM, Senegal II, dossier 2, Memoire
Sur l'Etat de la Colonie du Senegal jusqu'au dix Septembre I8I9, P. Valentin, habitant
de St. Louis (Senegal).
11 AFH 34
As there still exists a considerable demand for slaves amongst the Moors who
receive them in exchange for Horses, Cloths and other Merchandize and convey
them into the interior of their Country for Sale, there is no difficulty hitherto in
getting rid of them and the numbers in all probability are not lessened.62
And in I8I8 Gaspard Theodore Mollien visited the towns of Ganjool and
Kelkom, which had been recently devastated by the Damel. He commented
explicitly on the increase in warfare and pillage that was necessary to support
the Damel's court:
The abolition of the slave trade was a grand gesture, generous, worthy of an
enlightened century, but illusory, until we force the Moors to subscribe to it. Let
me explain: the Blacks sell to the Moors the slaves which are forbidden to the
Europeans to buy; and the king, who had sufficient revenue to maintain his court
by selling two hundred prisoners to the French at 6oo francs apiece, at present
makes more frequent incursions against his neighbors, and against his own
subjects, to double the number of his slaves and to have the same revenue, becau
the Moors pay only half as much for slaves as the Europeans did.63
The depredations of the Damel of Kajoor continued well after the ending of
the Atlantic slave trade, sometimes with the explicit purpose of gathering
slaves to purchase horses.64 And even as late as I854, the Kajoor elite was
pillaging Wolof villages with impunity and selling the victims into desert
slavery.65
Many of these slaves must have moved north along the trans-Saharan
route which ran along the Atlantic littoral up to Wad Nun in the Moroccan
Sahara, the point at which the slaves moved into the control of Moroccan
traders.66 In the late I840s, a trade spur opened up that linked Kajoor with
Wad Nun via the Adrar and apparently supplemented the coastal route.
General Faidherbe provided some details on a single caravan from Wad Nun
that travelled this spur and arrived at the Trarza escale. The caravan was
composed of 200 camels and i 50 mares, of which half had already been sold
along the way.67 These 75 mares were apparently destined for the markets
of Kajoor, and although the quality of the horses is not commented upon
62 PRO, CO 267/29 Answers to the Questions proposed to Lieutenant Colonel

Maxwell ... I January I 8 I I.
63 Mollien noted that when he visited Ganjool, the town had the appearance of having
been recently pillaged. The huts were burned, and much of the population had fled. The
Damel of Kajoor had come demanding that the village offer up 83 slaves to him. Mollien,
L'Afrique occidentale en i8i8, 39-40, 48, and (for the quoted and translated text) 76.
64 Ibid. 75; ANFSOM, D.F.C. Senegal, carton 83, no. I47, Rapport de M. de
Mackou... i6 mars I820.
65 ANFSOM, Senegal et dependances XIII, dossier 32, P. Bancal et al., 'Deuxieme
petition adressee a M. le Gouverneur du Senegal', Saint-Louis, I I February 1854, p. 6.
Even by the mid-nineteenth century, full-sized horses, used by the chiefs, continued to
be led into Kajoor by desert traders. Carrere and Holle, De la Sinegambiefranfaise, 74.
66 According to this account, '... the people of El Giblah [Gibla] sometimes go far to
the southward ... whence the Arabs obtain black slaves, in the proportion of 3 or 4 slaves
for each horse. These slaves are sold again at Wadnoon [Wad Nun]'. Major Rennell, 'An
account of the captivity of Alexander Scott among the wandering Arabs of the Great
African Desert', Edinburgh Philosophical J7ournal, iv (I821), 229.
67 Louis Faidherbe, 'Renseignements geographiques sur la partie du Sahara comprise
entre l'oued Noun et le Soudan', Nouvelles Annales de Voyage, 6eme serie, v (1859),
146-7. These other 75 horses were likely also sold for Black slaves, but these slaves do not
figure in the calculations which follow.
explicitly, at the reigning price of I 5 slaves per pure Arab steed, this would
translate into i,I25 slaves for export to the Saharan and North African
markets. And indeed, this calculation would exclude the sale of truly fine
horses. Faidherbe, as late as the late i850s, for example, noted that 'a fine
horse' could bring 25 to 30 slaves in the desert market.68 At the other end
of the range, if all of these horses were desert-bred, they would have brought
between three and four slaves each, and thus the value of the horses in this
single caravan would have been between 225 and 300 slaves.
THE ENDING OF THE HORSE AND SLAVE TRADE
On the heels of a crisis in the gum arabic trade, the French conquered the
lower valley of the Senegal river (I854-8).69 The French presence in the
lower Senegal blocked easy access by the desert horse and slave traders to the
Wolof states. For the lower Senegal the era of the horse and slave trade was
drawing to a close. As Faidherbe noted presciently in I859:
Marabouts from the Tirns, the Adrar, and the Trarza come to buy slaves in Kajoor
in exchange for their horses and [then] take them to Morocco. This migration,
which is leading to the depopulation of our colony, will soon cease, thanks to our
[new] situation in Senegal.70
The French were also successful in preventing large-scale desert raiding

expeditions into the Wolof states or Fuuta,71 and in this regard, too, they
helped staunch the northward flow of slaves. By the i86os the states of the
Atlantic littoral were far from demilitarized, but from this point onward the
Wolof cavalry forces apparently rode almost exclusively upon the local
crossbreeds.
Along the middle and upper valleys of the Senegal river, horse cavalry
continued to be central to the wide-scale religious and political violence of
the jihads of the second half of the nineteenth century. In the upper valley,
where the jihad of Umar Tall wrought awesome destruction, most of the
war-horses used in the religious and political violence seem to have been the
adapted crossbreeds. Umar Tall purchased his cavalry remounts in Segu;
68 Louis Faidherbe, 'Les Berberes et les Arabes des bords du Senegal' Bull.
Societe de Geographie, 4 eme serie (I854), IOO. Although beyond the scope of the present
article, it is important to note that religious teachings were sometimes exported south
along with the horses, and at least on one occasion the horse and slave trade was linked
to the establishment of a religious center in Senegambia. In one celebrated instance
Shaykh Bu Naama, the father of Shaykh Bu Kunta, went to Kajoor in the early nineteenth
century to sell horses, on the order of Shaykh Sidi Lamin, a disciple of Shaykh Sidi
al-Mokhtar al-Kunti. According to Kunta tradition, Shaykh Bu Naama was shocked at
the spiritual condition of the people of Kajoor and received permission to settle there.
And thus the desert horse trade figured in the establishment of the Bu Kunta settlement
near Thies in southern Kajoor. See Paul Marty, 'Le groupement du Bou Kounta', Revue
du Monde Musulman, xxxi (19 I5/I6), 4I5.
69 James L. A. Webb, Jr, 'The trade in gum arabic: prelude to French conquest in
Senegal', J. Afr. Hist., xxvi (I985), I49-68.
70 Faidherbe, 'Renseignements geographiques sur la partie du Sahara...', I43-4: 'Les
Marabouts du Tiris, de l'Adrar et des Trarzas viennent acheter des esclaves dans le Cayor
en echange de leurs chevaux et les emmenent au Maroc. Cette immigration, qui tend a
depeupler notre colonie, cessera bient6t, grace a notre situation au Senegal.'
71 ANFSOM, Senegal II, dossier 4, La Colonie du Senegal et des dependances,
A. Vallon, I2 August i86i.
11-2
Ahmadu Shaykhu oversaw royal stud-farms in Banamba.72 Along the middle

valley, combatants in the internal conflicts of Fuuta Tooro in the i870s
needed a steady supply of horses and at least some were probably obtained
from the desert.73 And by the late nineteenth century, under the conditions
of increasing aridity that forced the desert frontier south, Soninke-bred
horses became an important element in the trade; these horses were sold to
remount the cavalry of Samori Ture.74
By the turn of the twentieth century, desert caravans from the Adrar and
southern Morocco were still making their way toward Saint-Louis du
Senegal to sell horses, principally mares, to local merchants. These horses
often arrived in poor condition and had been poorly used; they brought
250-400 francs; the merchants fed and cared for them for two years and sold
their colts as 'river horses.' From I898 through i900, as noted earlier, the
French bought I 2-I5 of the desert horses per year. But with the French
occupation of the Trarza beginning in i 90 i, desert caravans stopped
coming.75 Horses from the Hawd, however, continued to be sold into
Senegambia - sometimes as race horses - and into the French Soudan well
into the twentieth century.
SUMMARY
Following the late fifteenth- and sixteenth-century cavalry revolution in Sene-

gambia, the horse and slave trade became a major sector of the desert-edge political
economy. Black African states imported horses from North Africa and the western
Sahara in exchange for slaves. Over time, under conditions of increasing aridity,
the zone of desert horse-breeding was pushed south, and through crossbreeding
with the small disease-resistant indigenous horses of the savanna, new breeds were
created. Although the savanna remained an epidemiologically hostile environment
for the larger and more desirable horses bred in North Africa, in the high desert
and along the desert fringe, Black African states continued to import horses in
exchange for slaves into the period of French colonial rule.
The evidence assembled on the horse trade into northern Senegambia raises the
difficult issue of the relative quantitative importance of the Atlantic and Saharan/
North African slave trades and calls into question the assumption that the Atlantic
slave trade was the larger of the two. Most available evidence concerns the Wolof
kingdoms of Waalo and Kajoor. It suggests that the volume of slaves exported
north into the desert from Waalo in the late seventeenth century was probably at
least ten times as great as the volume of slaves exported into the Atlantic slave
trade. For both Waalo and Kajoor, this ratio declined during the first half of the
eighteenth century as slave exports into the Atlantic markets increased. The
second half of the eighteenth century saw an increase in predatory raiding from the
desert which produced an additional flow of north-bound slaves. For Waalo and
Kajoor - and probably for the other Black African states of northern Senegambia
- the flow of slaves north to Saharan and North African markets probably
remained the larger of the two export volumes over the eighteenth century. This
northward flow of slaves continued strong after the abolition of the Atlantic slave
trade and was only shut down with the imposition of French colonial authority.
72 Doutressoule, L'elevage au Soudan Franfais, 220; L'eilevage en A.O.F., 64, 245.

73 David Robinson, Chiefs and Clerics: The History of Abdul Bokar Kan and Futa Toro
I853-I89I (Oxford, I975), I22.
74 ANS, IG 3I0, Kayes, Renseignements historiques, geographiques, et economiques
sur le cercle de Kayes, 30 March I904. I would like to thank FranSois Manchuelle for
communicating this reference to me. 7 Pierre, L'elevage, 79-80.
View publication stats

The Horse and Slave Trade Between The Western Sahara and Senegambia

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Horse and Slave Trade Between The Western Sahara and Senegambia

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Article in The Journal of African History · July 1993

The user has requested enhancement of the downloaded file.

THE HORSE AND SLAVE TRADE BETWEEN THE

Colby College, Waterville

THE CAVALRY REVOLUTION IN SENEGAMBIA

" Law, Horse, 6-7.

wadan, or 'horse from Wadan'.'" In addition, Tukulor tradition indicates

16 Mission Catholique de Saint Joseph de Ngazobil, Guide de conversation fra

20 Crone, Voyages of Cadamosto, 30, 33, cited in Law

common cultural elements to the broader region. The Black Africans of

CROSSBREEDING, HORSE MORTALITY AND SAHELIAN CAVALRY

26 Boulegue, Le Grand yolof, 155-73, esp. I56-62.

were able to reproduce themselves. The difference probably lay in the

30 Doutressoule, L'e'levage en A.O.F., 240. The categorization of the pony herds of

Marcel Leger and L. Teppaz, 'Le "Horse-Sickness" au Senegal et au Soudan

Table i. Annual horse mortality of the colonial cavalry in the French

Year I 882-3 I 883-4 I 884-5 I 885-6 I 886-7 I 887-8

Source: C. Pierre, LYlevage dans l'Afrique Occidentale Franfaise (Paris, i906),

38 Mission Catholique, Guide de conversation, 63.

41 Interview with Mokhtar wuld Hamidun, i9 December I980, at Nouakchot

significant change in the regional distribution of military power. In the

COMPARISON OF THE ATLANTIC AND TRANS-SAHARAN SLAVE

locally bred horses in the late nineteenth-century French Soudan), Waalo

" ' Relation du Sieur Chambonneau, commis de la compagnie de Senegal, du voyage

~~~~~~t C;.c; X VX X X h._ Q .

1-4 4--J 4-1

4-J ::s ct U CZ C ct 4-J

4 Archives Nationales de France (hereafter, ANF), C6 i I, Rapport de I4 Juin I736,

numbers of slaves, particularly from 1760 when desert warriors helped

The existence of a significant western Saharan and Maghribine demand

IN THE AFTERMATH OF THE ABOLITION OF THE ATLANTIC SLAVE

62 PRO, CO 267/29 Answers to the Questions proposed to Lieutenant Colonel

THE ENDING OF THE HORSE AND SLAVE TRADE

The French were also successful in preventing large-scale desert raiding

Ahmadu Shaykhu oversaw royal stud-farms in Banamba.72 Along the middle

Following the late fifteenth- and sixteenth-century cavalry revolution in Sene-

72 Doutressoule, L'elevage au Soudan Franfais, 220; L'eilevage en A.O.F., 64, 245.

You might also like