EnterText 4.

DONALD CRUMMEY

African Banditry Revisited

Eric Hobsbawms Bandits is a powerful and beguiling work of historical imagination

attentive to the rural poor, their capacity for political action and their potential as

participants in larger processes of social change.1 It combines an enthusiasm and

affection for popular heroes of the countryside with an extraordinary range of

examples across space and through time. Central to Hobsbawms argument is that,

from time to time, bandits rise from the level of criminality and vendetta to become

vital articulators of the cause of the rural poor and actors on their behalf. In short, they

become social bandits, the prototype for which, in the anglophone world, is Robin

Hood, who stole from the rich and gave to the poor. Criticism seems pedestrian by

contrast with Hobsbawms sweep and verve. However, as Hobsbawm points out, in

the introduction to the first two editions of the book, Africa is conspicuously absent

from his gallery of heroes, a shortcoming which he addresses by reference, in the first

paperback edition, to the antics of Ghanaian cocoa smugglers in the 1960s, and, in the

second edition, to the careers of the Mesazghi brothers, Eritrean bandits who found

themselves swept up into anti-British politics of the 1940s.2

There was no great rush of Africanist scholarship to respond to Hobsbawms

invitation and challenge. Ed Keller was pretty quick off the mark with a 1973 article

Donald Crummey: African Banditry Revisited 11

 EnterText 4.2

in the Kenya Historical Review,3 and Allen Isaacman followed in 1977 with an article

in the Journal of Southern African Studies.4 They were isolated figures until the

publication in 1984 of Richard Caulks Bad Men of the Borders: Shum and Shifta in

northern Ethiopia in the 19th Century,5 and in 1986 of Banditry, Rebellion and Social

Protest in Africa.6 Banditry, Rebellion attempted to assess the state of the field, to

locate banditry and social banditry within a context that included African

historiographys standing interest in protest and rebellion, to make some contributions

to the literature, and to suggest some programmatic directions. How does its

assessment look after thirteen years?

In the introduction I asserted the criminal character of unreconstructed

banditry and suggested that Africanists would do well to follow the lead of the

literature on the social nature of crime in eighteenth-century England and take more

seriously the issue of criminality. The introduction also encouraged Africanists to take

up Hobsbawms challenge and engage with banditry. It suggested, on the basis of the

case studies that followed, that banditry was, indeed, a significant African

phenomenon, but that approaching it, initially, as rural social criminality would prove

more productive than privileging the concept of the social bandit. It appeared then that

there was evidence of social banditry in Africa, or at least of its myth, but Banditry,

Rebellion argued that it would be unproductive to assume social banditry at the

beginning of any research project.

I think that those judgments have stood up well to such literature as has

emerged since 1986. I would like, first, to review the arguments contained in

Banditry, Rebellion, then look at subsequent advocates of the analytical utility of

social banditry, and finally review the material which has appeared since 1986

bearing on banditry and heroic criminality in Africa.

Donald Crummey: African Banditry Revisited 12

 EnterText 4.2

The contributors to Banditry, Rebellion were less dismissive of the social

bandit than some of the collections critics have suggested, but, for the most part, did

express difficulty in retrieving social banditry from their recalcitrant sources.7 They

were not reluctant to engage with the concept. Moreover their views on banditry were

rather more nuanced than has generally been recognised. Hobsbawms strongest critic

in the collection was probably Ralph Austen. Austen argued that the social bandit

concept was essentially inapplicable to African conditions, but proceeded to suggest

that Hobsbawms emphasis on the social meaning of criminality had been unduly

neglected by Africanists and laid out some new ways in which they might more aptly

conceptualise it.8 At bottom his analysis rested on the view that, historically, African

societies are best seen as segmentary, and lacking the internal or external frontiers

which make banditry possible. He noted that the European language of conquest often

construed African resistance as forms of banditry, but that it would be wrong-headed

to take that language at face value. On the one hand, we already have the concept of

primary resistance to deal with the phenomenon, and, on the other, the contesting

parties shared no institutions or repertoire of values, an essential pre-condition for

banditry of any kind.9 Yet Austen stood alone among the contributors to Banditry,

Rebellion in the extent to which he separated himself from Hobsbawm and in arguing

that Africanists needed a new typology for understanding the political significance of

social deviance.

At the opposite end of the analytical spectrum, and closer to the ground which

I occupy myself, Ray Kea argued that the frontiers denied by Austen in fact

characterised a succession of states and societies on West Africas Gold Coast, at least

from the fifteenth century, and demonstrated that both the European and African

language sources available for the region contain a rich vocabulary of banditry.10

Donald Crummey: African Banditry Revisited 13

 EnterText 4.2

Nevertheless, he, too, found few heroic examples and concludes with a reference to

Gramsci, Every trace of independent initiative on the part of subaltern groups should

therefore be of incalculable value for the integral historian, which I take to be an

endorsement of the view that the social context and meaning of criminality are

valuable terrain for the historian to explore.

Rob Gordon, like Ray Kea, had no problem in finding bandits, in his case in

twentieth-century rural Namibia, and he demonstrated persuasively that rural crime is

an illuminating prism through which to approach the various stages of German and

South African incorporation of the people whom they called Bushmen.11 Each stage

was marked by changing forms and incidents of crime in the countryside. Bushmen

appear in early colonial legal records primarily for their offences as farm labourers.

However, starting around 1911, the incidence of stock theft and other kinds of rural

crime swell to a high in the 1920s, and Bushmen appear as the principal culprits.

Thereafter, Bushmen stock offences slowly decline. Gordon argues that the rise in

stock theft coincided with and was a response to European settlers increasingly

occupying Bushman land and turning it into cattle ranches. The slow subsequent

decline in stock theft Gordon connects with the steady assimilation of Bushmen as

farm labourers. Banditry, in twentieth-century Namibia, was indeed a form of social

crime.

One of the more interesting references to banditry in Banditry, Rebellion was

by Alison des Forges in her study of a 1912 rebellion in Rwanda.12 Rwandan society

in 1912 recognized three major groupsthe Abahutu cultivators, the Abatuutsi

pastoralists, and the Abatwa hunting people. Des Forges finds the category of bandit

useful in discussing the political and social role played by a group of Abatwa led by

one Basebya. Basebya and his followers, skilful archers, had become the personal

Donald Crummey: African Banditry Revisited 14

 EnterText 4.2

guard of Rwabugiri, a formative ruler who died in 1895. Basebya opposed

Rwabugiris successor and withdrew his men into a vast swamp in northern Rwanda

where they lived by hunting and plundering the local cultivators, their power and

influence increasing following a famine in 1905. By 1907 the demands of German

colonialism, channelled through the Rwandan court of Rwabugiris successor, was

generating increasing opposition in northern Rwanda, and substantial numbers joined

Basebya. He and his group proclaimed no political goals and attacked all who had

not paid for their protection, elite and ordinary people alike. But their life outside the

law made them by definition enemies of the ruler 13 In 1912 Basebya hitched his

wagon to the rising star of a religious medium turned royalist pretender and became

his key supporter. Basebya now organised his growing band into two groups, one of

Abatwa, one of cultivators and pastoralists. Des Forges reflects that it is impossible

ultimately to determine the motivations of the Abatwa in supporting rebellion against

colonial oppression, for, while they had good reason to hate the European-supported

Rwandan court, they plundered other supporters of the pretender. In April of that

same year the Germans successfully attacked the pretender and a few days later

captured and executed Basebya. His band passed from the record. Des Forges is

persuasive, pace Austen, that bandit usefully describes the role that Basebya and his

followers came to play. Moreover, social banditry seems at least possible as an

interpretation of their participation in the rebellion of 1912.

Terence Ranger explored the meaning of banditry within the framework of

Zimbabwes guerrilla war for independence which started in the late 1960s and led to

independence in 1980.14 He noted that there was some background to social banditry

in earlier colonial Southern Rhodesia, primarily through the existence of independent

bands which haunted the eastern borderlands with Mozambique. He also noted that

Donald Crummey: African Banditry Revisited 15

 EnterText 4.2

poaching flourished in Crooks Corner, the borderlands linking Southern Rhodesia,

South Africa and Mozambique, evidence documenting the South African side of the

border bearing on the first three decades of the twentieth century and on the

Rhodesian side for the late 1950s. Here we have rural folk, criminals in the eyes of the

state, and, apparently, heroes in the eyes of their farming neighbours. One of them

achieved the status of a social bandit, Rhodesias Ned Kelly, who frequently

presented meat to his temporary African neighbors.15 However, most of Rangers

chapter dealt with the period of guerrilla warfare, which he used to explore the

differing forms of legitimacy which armed bands achieved or failed to achieve.

Bandits are armed bandsguerrillaswithout the sanction of a political party, whose

legitimacy is recognised by the peasants. These bandits achieve social status if and

when they acquire legitimacy in the eyes of those same peasants. Ranger claimed that

such groups existed, their exemplar being Obert Dhawayo, who plundered only

European property and who was viewed by local people as a sort of Robin Hood.16

Fernyhough, Caulk and I all explored Ethiopia, an African example rife with

banditry, an institution embedded in an agrarian, hierarchical, class-divided society.

Indeed, Ethiopia has given us the vernacular term by which bandits are known

throughout north east Africasheftaa term which I parse below. Bandits are

commonly encountered in the literature, both indigenous and foreign, dealing with the

Ethiopian highlands in the nineteenth and early twentieth centuries. Ethiopia,

particularly in the later nineteenth century, was going through the strained process

whereby the monarchy re-asserted itself as a central actor on the national political

stage. European imperialism was an ever-intensifying force in the region, but it

impinged little on the Ethiopian countryside, with few exceptions apart from Eritrea,

Donald Crummey: African Banditry Revisited 16

 EnterText 4.2

until far into the twentieth century. The forces generative of banditry in Ethiopia were

primarily indigenous and the cultural understanding of shefta entirely so.

Caulk first drew attention to Ethiopian banditry in a 1984 article.17 He reviewed the

evidence concerning shefta in the Tegrea-speaking areas of northern Ethiopia in the

nineteenth century, starting with the case of one Balgada Araya of the 1840s. He

noted that most shefta also received appointments (shum = appointee), that one of

them actually seized the throne, becoming Emperor Yohnnes IV, and argued for an

intimate relationship between the Ethiopian nobility and the institution of banditry or

sheftent. On the political motives of noble Ethiopian shefta, Caulk cited the revealing

letter written by Dbbb Araya to the British at Aden in 1888. I treated with the

Italians and I submitted to them thinking that all other Abyssinians dependent on them

will be under my orders. I wished to be the only Chief without having other Chiefs on

my side; they made other Chiefs, that made me angry, and I decided to desert them.18

Nor did Caulks sources paint a more favourable picture of common bandits, who

made their livings by cattle rustling and raiding caravans. Popular culture did not

make any pretense that their exploits served social justice by settling accounts (pace

Hobsbawm).19

In his contribution to Banditry, Rebellion, Caulk dealt with the rich case of

Baht Hagos, leader of an 1894 rebellion against the Italians in Eritrea, who started

his career as a bandit (the term shefta is used in both Amharic and Bahts native

Tegrea), driven by a blood feud to take refuge in wild country. He formed a band

and proved adept at raiding and plunder. Nevertheless, the position from which Baht

led his revolt was an Italian-appointed one.20 In 1889 he started a connection with the

Italians, then still based at the coast in Massawa, and not long after received their

appointment as a local ruler. In 1894 he revolted. Originally motivated by a growing

Donald Crummey: African Banditry Revisited 17

 EnterText 4.2

sense of his own political insecurity, he soon found himself articulating a host of local

discontents against Italian high-handedness, no longer as a bandit, but as rebellious

appointee.

My chapter in Banditry, Rebellion explored themes first established by Caulk

in his 1984 article. My point of departure was semantic, an analysis of the words

shefta (= bandit) and sheftent (= banditry), which, as I pointed out, derive from the

Amharic root shfft, he rebelled.21 A shefta then is a rebel, in some respects a

primitive rebel, although in a sense somewhat different from that intended by

Hobsbawm in his earlier work, Primitive Rebels, out of which Bandits emerged.22

Hobsbawms rebels were primitive by reference to the more secular, centrally focused

rebels of the developing socialist movements of the European world. Mine were

primitive in lacking all but the most elemental ideological motivation and in

rebelling, primarily, in the interests of their own social and political advancement.

The article noted that both the Ethiopian and European language sources supported

the view that sheftent covered much the same linguistic terrain as the English

banditry, and went on to discuss two of the most famous shefta of nineteenth-

century Ethiopia, the future emperors Twodros II and Yohnnes IV. Sheftent in the

nineteenth century, it thus appeared, was most saliently an institution for career

mobility of the Ethiopian nobility.

Yet the interest and challenge of Hobsbawms Bandits is to illuminate the

obscure dynamics of rural protest and resistance. My chapter took up this issue,

exploring peasant political behaviour as revealed by the same sources. I ought to have

been more explicit about the language involved. The European references are

unproblematic. In my review of the Ethiopian language sources, I looked for

references to what, in Amharic are called the balagr, farmer, country person,

Donald Crummey: African Banditry Revisited 18

 EnterText 4.2

peasant.23 To be sure, the term is used loosely, but, then, so is peasant. The

nineteenth-century sources give us a number of examples of peasant/balagr

political behaviour, but establish no meaningful progressive connection between

peasants and the institution of banditry.

Fernyhoughs review, although replete with colourful incidents and characters,

yielded much the same result.24 The twentieth-century information is much richer than

the nineteenth, leading to subtler observations, and Fernyhough found the

involvement of all social groups in banditry. Whereas my nineteenth-century sources

bore primarily on the leadership of shefta bands, Fernyhough had more information

on their membership and found a significant proportion of it to be of peasant origin.

He noted how banditry particularly flourished on the national borders, in and around

internal administrative boundaries, and in rugged terrain. He also drew attention to the

extraordinary amount of banditry around Lake Tana. Fernyhough explored the social

and political conditions which fostered banditry, pointing out the potent coincidence

of natural disasters and internal military conflicts in the decades surrounding the turn

of the twentieth century. He suggested that soldiers disbanded following the Battle of

Adwa and that the suppression of internal revolts was a prime source of bandits.

Bandits rustled cattle, raided for slaves, and plundered caravans, and even bandits of

lowly origin plundered indiscriminately. Banditry also served career mobility.

Whereas in the nineteenth century the very throne was at contest, in the twentieth

century the political ambitions of the shefta now focused at the provincial level and

lower.

On the issue of social banditry, Fernyhough noted how bandits in both Eritrea

and Ethiopia became caught up in political struggles, starting with the Ethiopian

resistance to the occupation which followed the Italian invasion of 1935. In the 1940s,

Donald Crummey: African Banditry Revisited 19

 EnterText 4.2

the Mesazghi brothers committed themselves to the struggle to rid Eritrea of British

rule. Thus, in the epic national struggles against twentieth-century European

imperialism Ethiopian bandits played out the role of the social bandit. But

Fernyhoughs summary judgment on Ethiopian banditry was harsh: For most

peasants, sheftent was a burden of tribute and fear.25 Nor does the evidence which

Caulk and I reviewed lead to a significantly different conclusion.

The aura of social bandit does surround Kassa Haylu who seized the imperial

throne as Twodros II in 1855.26 Kassa was the son of a titled nobleman and half-

brother of a powerful nobleman of the 1830s. Sources closest in time to Kassas early

career describe a rebellious nobleman, who in the early 1840s launched his career on

Ethiopias western frontier, supporting himself as a highwayman and plunderer of

folk who lived beyond the pale of Ethiopian society. He raided these same people for

slaves and reduced them to the status of serfs. He entered a bandit subculture,

marrying the daughter of another shefta (whom he subsequently divorced), and

leaguing with yet others. His official chronicler recounts an incident from these early

years in which Kassa provided relief to Christian peasants suffering from natural

calamity. He gave them lots of money and directed his soldiers to clear the forest

and break land on behalf of the peasants. It seems much more an act of noblesse

oblige than of social banditry, given that Kassa went on to assume the imperial throne.

A later source has glossed this incident, claiming The peasants of Qwara loved

him all the peasants became shefta with Lej Kassa, and young men joined him from

as far away as Dmbeya and Chelga.27

The aura of Kassa as social bandit increases in direct proportion to the

distance we remove ourselves from what we actually know about his career. If this is

not the case with all social bandits, it is certainly the case with many. Hobsbawms

Donald Crummey: African Banditry Revisited 20

 EnterText 4.2

prototype, Robin Hood, may be taken as a case in point, his very historical existence

uncertain, his personage and role the construct of generations of tradition. Yet surely

this is a phenomenonthe remarkable tendency for agrarian societies through time

and across space to construct models of heroic vindication and redistributionin

itself worthy of attention. Yet such figures are not widely manifest in the literature

about Africa.

I would summarise the contributions of Banditry, Rebellion to the question of

banditry and social banditry in Africa as follows. Most contributors took precisely the

line which Austen argued against. Kea and des Forges found banditry in the

hierarchical societies with which they deal, the one on the West African coast and its

hinterland, the other in the East African Great Lake region. Southern Africanists,

dealing with states which radically alienated African property and delegitimised

African values and customs, reported banditry in Namibia, Zimbabwe and

Mozambique. Ethiopianists found it throughout the northern Ethiopian highlands as a

product of historic Ethiopian society and its state. Social banditry they all found

elusive, although the Southern Africanists, Ranger and Gordon, found the concept of

considerable utility. Des Forges showed how a bandit leader provided critical support

for one of the leading rebels against the oppression of early German rule. The

Ethiopianists reported the myth at work in the nineteenth century and noted that in the

crises of the 1930s and 1940s involving questions of national survival and

independence, bandits did participate in political resistance, lending their support to

the emperor, in one case, and to a secular political party in the other. Enough self-

justification; what have Africanists had to say about banditry following the

publication of Banditry, Rebellion?

Donald Crummey: African Banditry Revisited 21

 EnterText 4.2

First of all, banditry remains a popular pejorative with which people seek to

discredit others. A typical example is the Economist, which on August 8, 1992

published a story entitled Banditry versus politics, which argued that, in those days

of transition in South Africa, no one was really ruling the country. The larger event

was a general strike called by the African National Congress, but within that epic

contest, the Economist chose rather to focus on the story of two journalists who, their

car having been hijacked, were left for dead by their assailants. Neither the

government, whose security forces were in the area, nor the ANC, who had instructed

its followers to help journalists, were able to protect them: Banditry, not politics,

rules.28

Allen Isaacman, in a major review article dedicated to peasants and rural

social protest in Africa, addressed social banditry.29 He suggested that there were two

schools of thought on this question. One school of thought rejects the penchant of

colonial regimes for dismissing most forms of rural protest as deviant criminal

behavior. The other school of thought, partially, but not wholly associated with

Banditry, Rebellion, was not so clearly characterised. Isaacman noted that Austen

dissociates himself from the concept but that some contributors to the volume did find

social banditry useful. Isaacman did not come to terms with those chapters of

Banditry, Rebellion, particularly the Ethiopian ones, which document African

banditry, but which find little incidence of social banditry. He comes close to

suggesting that any rural phenomenon which the European colonial rulers of Africa

called crime counts as social banditry, which, in turn, would lead to the odd position

of having social bandits without really having bandits in the first place. The

introduction, and most (probably all) of the contributors to Banditry, Rebellion

rejected the penchant of colonial regimes for dismissing most forms of rural protest

Donald Crummey: African Banditry Revisited 22

 EnterText 4.2

as deviant criminal behavior. Indeed, one of the main contentions of the introduction

was precisely that we should be sceptical of claims of criminality in the context of

colonial Africa, and that, where it might reasonably be held to exist, we should

explore its social content and meaning. Banditry is a special case, prior to, and

inclusive of, social banditry. Banditry, the institution addressed by Hobsbawm, is

embedded in rural societies, and bandits are recognised as such by rural folk. Local

languages have specific terms for bandits and if the criminals in question are not

locally recognised as bandits, and labelled as such in an African language, then

bandits they are not, but something else again. Banditry, Rebellion argued that

banditry pure and simple did exist in Africa and that we should be attentive to it and

to its social meaning. If this institution came to serve some larger social or political

end, well and good, but its prior existence had first to be established.

Isaacman cited Catherine Coquery-Vidrovitch and her 1988 book, Africa:

Endurance and Change South of the Sahara, to suggest that social banditry was a

fairly widespread phenomenon in Africa. I find the reference unpersuasive, at least so

far as her case is revealed by her index for banditry.30 She, too, moved directly to

social banditry without first passing through banditry.31 Her book has a section (pp.

183-8) on Passive Resistance and Social Banditry. Noting that standard

anthropological treatments of Africa have eliminated the concept of marginality from

the history of Africa, she continued, This amounts to a claim of uniqueness for

African societies, since social banditry exists throughout the rest of the world, from

Sardinia to the Ukraine, from Java to Brazil. Since social bandits are a universal

phenomenon then they must be found in Africa, a syllogism which assumes what it

should demonstrate. She then proceeded to collapse the categories of social banditry

Donald Crummey: African Banditry Revisited 23

 EnterText 4.2

and banditry, drew prominently on Isaacmans 1977 article, and referred in large part

to the same incidents as did Isaacman himself.

Nonetheless, interesting writing on banditry and on the social dimensions to

criminality in Africa there has been. Fernyhough has revised the views he put forward

in Banditry, Rebellion, now arguing that Ethiopian banditry was less dominated by the

noble class than the papers in the 1986 collection allowed, and evinced more explicit

peasant participation, that it was an institution more subversive of the social order. It

is worth quoting his conclusion at some length. His earlier interpretation, he now

argued:

diminishes the extent to which peasants turned outlaw to escape the

worst hardships of their rural milieu. Secondly, peasants played
leading roles as shifta and were not always subordinate to nobles and
gentry. Thirdly, the stress on the interdependency of nobles and
peasants within sheftenat deprecates the very close links between
peasant brigands and local communities. The latter assisted bandits and
in turn the authorities demonstrated their recognition of this support in
the ways they tried to suppress brigandage. Rural support was
particularly assured when peasants could see no crime in the incident,
which led the shifta to his new calling, but it collapsed, as Bahta Hagos
discovered, when the bandit was repressive. Fourthly, peasant bandits
were often so explicit about their grievances and the unjust demands
on their communities that their protest, though couched in terms of
familiar moral claims, edged towards being reformist. As bandits
joined successive peasant revolts after 1941, that protest became
clearer and more radical, though often the peasant-bandit voice
remained contradictory, and one among many. Nevertheless, active
shifta involvement in rebellions in the latter years of imperial Ethiopia
cannot be squared with the assertion that banditry helped diffuse
revolt, not least because the number of bandits who made it into the
ruling elite after 1900 was trifling, and their example exceptional.
Ironically, by igniting peasant resentment even the coopted, repressive,
bandit may have played his part in politicizing the weak and humble.32

Outside of Ethiopia, scholars of North Africa have paid the most serious attention to

banditry. David Harts Banditry in Islam looks at five cases from Morocco and one

from Algeria as well as one from Pakistan.33 While he had no problem in finding

bandits, he found practically no redistributors, the only case being reported in sources

Donald Crummey: African Banditry Revisited 24

 EnterText 4.2

at several removes from their original. On the other hand, banditry in both Morocco

and Algeria could not avoid being caught up in struggles to resist European

colonialism, and, in this sense, assumed social dimensions.

Banditry in Morocco was localised in two mountainous regions, the Jbala in

the north, and the Atlas, in the centre and south. The Jbala gave rise to one Ali I-Bu

Frahi, Ali the Six-Fingered, in the first decades of the nineteenth century.34 Harts

account here rests, in turn, wholly on the account of the Englishman, John

Drummond-Hay. Ali was a very physical man and found himself in a number of

scrapes, eventually emerging as a highwayman leading a gang which waylaid

caravans on the main road between Tangier and Tetuan. The sources attribute to him

an eloquent defence of his activities, in which he compared himself with the sultan,

who also broke the Prophets law by rapine and extortion. Ali enjoyed a Robin Hood

reputation. He never molested the poor and enjoyed the active support of local

villagers who fed him and invited him to their weddings. However, he achieved such

eminence as to warrant a sultanic edict against him and he was eventually betrayed,

punished, and killed. Finally, as Hart notes, Ali was already becoming legendary

when Drummond-Hay recorded his story.

Other Moroccan bandits reveal fewer redeeming features. Bandit gangs

operated in the same area as Ali at much the same time. They robbed travellers, raided

livestock, and enslaved children.35 The poor, because of the feebleness of their

resistance, were favoured targets. The most famous of these bandits was Mawlay

Ahmad al-Raysuni, ruthless and very successful.36 In the first decade of the century he

made a profession out of kidnapping Europeans (and Americans) for ransom. This

made him not only rich, but also famous, and he was appointed governor of Asila. In

the circumstances of rising European colonial pressures in the area, Raysuni

Donald Crummey: African Banditry Revisited 25

 EnterText 4.2

temporised, refusing to join the resistance led by bin Abd al-Krim, who eventually

defeated and incarcerated him in 1925, when he died.

Raysuni and his fellow bandits of the Jbala were Arabic speakers. The Berber

of the High Atlas also put forth some bandits of their own. None evinced a

redistributive ethic, although one did involve himself in the Moroccan struggle to rid

themselves of the French. Bandits were known as iqattaem n-ibriden, cutters of the

road, and around 1930 two of them became particularly famous, w-Abbu and u-

Khuya. They bound themselves together by a socially recognised covenant and then

set about robbing and murdering, their victims being all from neighbouring tribes.

Eventually, u-Khuya betrayed w-Abbu, leading him into an ambush, from which he

himself did not escape alive, the ambushers punishing him for breaking his covenant

with his fellow.37 A few years later, in 1933, the nearby Ait Murghad were brought

under French rule. One of them, Zayd u-Hmad, refused to surrender and took to the

hills. He had already for some years established himself as a raider, and in the

summer of 1934 he began to waylay colonial officials, plundering them of payroll

money. However, he found that his own people would no longer provide him with

refuge, so he turned to jihad. Again he attacked officials of the French administration,

killing a few, before he turned to attacking Muslims who were collaborating with the

French. His attacks were successful enough to provoke a manhunt, and at the

beginning of March 1936, he was killed by one of his former companions.38 The

resistance of Zayd u-Hmad was reflected twenty years later in the career of yet

another Berber bandit of the Atlas, Hmad Uhansal, whose brief career, in mid-1953,

took place in a larger atmosphere of nationalist agitation. Hmad killed a number of

Frenchmen before the French secured his betrayal by offering a large reward. Like

Donald Crummey: African Banditry Revisited 26

 EnterText 4.2

most of the other Moroccan bandits, there is a national edge to the story of Hmad, and

in that sense he touches on social banditry.

Neighbouring Algeria, with its large population of French settlers and

extensive alienation of the lands of the indigenous, might be supposed to be more

fertile ground for social banditry. Banditry there was, and with social overtones. The

first noted bandit, Muhammad u-l-Hajj Abdun-Kabylia, emerged in the aftermath of

a great rebellion of 1871 in the Kabylia.39 w-Abdun was convicted of murdering the

president of a local council. In 1884 he was condemned to imprisonment on Devils

Island, off the coast of South America, from which he escaped and returned to

Algeria, where he entered the forest. In 1891 he was joined by a brother and a

nephew. His main targets were Algerians who acted against them. The French

attributed multiple murders of other Kabyles to him. Multiple attacks in 1892 and

1893 meant that it was harder and harder for the French to staff local offices or to

recruit Algerian police. W-Abdun did enjoy substantial support from local people,

but he also multiplied his enemies among them, and early in 1894 his nephew was

captured, and in 1895, w-Abdun himself went on trial in Algiers. He was publicly

executed in May 1895.

The First World War was the setting for the career in the Aures of Misaud bin

Zilmad, who headed up a band of army deserters and malcontents.40 He took over this

role from his brother. On the night of 14-15 October 1917, bin Zilmads band attacked

the village of Fum Tub, an attack which drew the attention of the French. Until 1919

bin Zilmad confined his activities to attacking Algerians, some of them in revenge for

his brothers death. In that year, the French began large-scale operations against him

and other bandits. Although he survived at least one battle, he was eventually tracked

down and killed in March 1921. Bin Zilmad did show a concern for justice, curbing

Donald Crummey: African Banditry Revisited 27

 EnterText 4.2

the exploitation of the poor by the rich and by limiting the liberties taken by local

power holders, but he paid little attention to French settlers in the region. So, of the

two Algerian bandits, bin Zilmad more clearly demonstrates a social dimension.

Nathan Brown takes up the question of banditry in modern Egypt, arguing

that, essentially, it was an invention of the Egyptian state as it struggled to strengthen

and modernize its position vis vis, on the one hand, the recalcitrant Egptian

countryside, and, on the other, the occupying British.41 To be sure, banditry was not

wholly a state invention, rather an increase in rural crime was turned by the Egyptian

state into a national crisis justifying new institutions, Commissions of Brigandage, for

intervention in the countryside, institutions outside the control of the British. In the

1880s there does appear to have been one of a number of upswings of robbery and

plundering in the countryside. Rather than the classical fellaheen, the perpetrators

seem to have been, firstly, bedouin, and secondly, immigrants from Upper Egypt,

Nubia and the Sudan. These bedouin had long ceased to be nomads but retained a

social identity distinct from that of the ordinary countryfolk. The robbers of the 1880s

operated in gangs ranging from as few as six up to sixty members and took part in

highway robbery, estate raiding and rustling. Some set up protection rackets targeted

at the peasantry. Brown claims that there is no evidence that these bandits enjoyed a

privileged relationship with the peasants or acted with any consciousness of the needs

and interests of the peasantry. Nor is there evidence that they achieved heroic status in

the eyes of the peasants. Rather, such evidence as we have, suggests that peasants

looked on bandits with fear. Not unlike the Ethiopian situation, Egyptian banditry

was a weapon not of the weak but of the strong.42 And it proved, ultimately, to be

a weapon whereby the British extended their control in Egypt, seizing control of the

Donald Crummey: African Banditry Revisited 28

 EnterText 4.2

Commissions of Brigandage from the Egyptian government and taking them into their

own hands.

Gwyn Campbell writes of the importance of banditry in nineteenth-century

Madagascar, depicting an institution arising from the dynamics of indigenous

societies and their interactions much as the Rwandan and Ethiopian kingdoms also

generated banditry.43 Banditry in Madagascar was intimately bound up with

resistance to incorporation by an indigenous state. Not the least of the strengths of

Campbells account is the rich Malagasy vocabulary for banditry which he reports.

Madagascar banditry clearly had a social dimension, but that dimension was

originally generated by the oppressive actions of the Merina kingdom, and only at the

very end of the nineteenth century did it engage with European imperialism.

Campbell identifies three distinct strands of indigenous banditry: traditional cattle-

stealing, anti-Merina resistance, and the reaction of ordinary farmers within the

Merina empire to oppressive taxation.44 Cattle-stealing and slave raiding were

institutionalised activities of two non-Merina ethnic groups in Madagascar. Anti-

Merina resistance generated a different kind of banditry along the external and

internal boundaries of the Merina state, which, by the mid-1820s, had assumed its

definitive shape, controlling the islands central plateau and corridors leading through

tropical rainforest to the principal port on the islands north-west coast. Bandits

haunted the efitra, a no-mans land ringing the plateau. The efitra had historically

been the home of bandits and their bands were swelled in the nineteenth century by

refugees fleeing imperial Merina rule. Finally, Campbell argues that brigandage arose

within the Merina state as a protest against the states forced labour policies, starting

in the 1820s, and expanding with the escalation of forced labour in 1869. The Merina

Donald Crummey: African Banditry Revisited 29

 EnterText 4.2

state played a crucial role in determining both the nature of brigandry and its

timing45 by its efforts to monopolise land and labour.

The policy of forced labour, fanompoana, coopted young men into either the

army or factories. Terms of service were harsh and disruptive of local agricultural

production. Protest in turn was met with harsh repression that fed the flight to the

efitra. The states role in stimulating banditry, in Campbells view, is reflected by the

fact that brigandry tended to reach epidemic proportions during and immediately

following any period of major or prolonged military conflict, for war both produced

and left behind armed bands of uprooted individuals.46 However, he notes in addition

that banditry also increased at moments of state strength, which in turn meant

increased fanompoana. Banditry drew young men from agricultural societies during

the off-cycle of the agricultural year, one that coincided with an increase in

commercial activity. Bandits rustled cattle, enslaved, and destroyed rice silos. There is

little evidence that they re-distributed their plunder to local villagers, in spite of the

fact that they maintained close ties to them, but they did express resistance to the

overweening demands of the Merina state. Through most of the century their focus

was the Merina state. However, from the late 1880s the plateau bandits were

transformed into guerrillas fighting a state that violated ancestral values and allied

itself with foreign missionaries and mining firms. French imperialism tried to prop up

this regime, and, in so doing, ignited the Menalamba revolt of 1895-97, in which

former bandits played prominent roles.47 Here then is a persuasive example of

banditry, generated by indigenous dynamics, evincing the venality of ordinary

banditry, but at the same time embodying social protest and feeding into rebellion.

An article by George Simpson returns the discussion of African bandits to

north-east Africa and to Ethiopia.48 Bandits are really incidental to Simpsons paper,

Donald Crummey: African Banditry Revisited 30

 EnterText 4.2

the primary focus of which is on colonial decision-making; nonetheless it does

confirm the importance of Ethiopia as a generator of bandits in the African Horn.49

Following the definition in the early years of the twentieth century of the borders of

north-east Africa through a series of bilateral treaties between Ethiopia, on the one

hand, and Britain, Italy and France on the other, the governments of the region were

slow to establish administrative control of their outlying territories. Ethiopian

territorial expansion, unlike that of the European powers, involved the settlement of

garrisons of Abyssinian peoples in newly-subjugated territories, garrisons only

loosely controlled from Addis Ababa and supported, not by salaries from the national

government, but by levies on the local populations.

The establishment of borders meant the creation, in the European colonial

territories, of lush fields adjacent to Ethiopia, only nominally policed, open to the

Ethiopian border garrisons, who plundered them for cattle, ivory and slaves. Northern

Kenya and eastern Sudan proved temptations which the Ethiopians could not resist.50

These plundering bands, with more than a little justice, were looked on as banditry by

the administrations of Kenya and the Sudan. Simpson addresses a crisis of cross-

border raiding occasioned in 1913 by the terminal illness of Emperor Menilek and the

consequent weakening of central control from Addis Ababa, but, in truth, the regional

problem was broader and more longstanding. The British consul in southern Ethiopia,

Arnold Hodson, described the situation as follows: Since 1920 the inroads of

poachers and raiders from Abyssinia into the East Africa Protectorate had become

more and more frequent, and had made life intolerable both for the natives, who were

robbed of their cattle and often murdered into the bargain, and for the

administration.51 Simpson argues that persistent Ethiopian cross-border raiding led

eventually to the strengthening of the British imperial presence in Kenyas Northern

Donald Crummey: African Banditry Revisited 31

 EnterText 4.2

Frontier District. However, at the same time, his article draws our attention back to

the association of banditry with Ethiopia and to the venal, rather than heroic,

dimensions of banditry.

At the same time as bandits have received modest attention in the literature

dealing with Africa, African historians have also explored social dimensions to

criminality.52 Space precludes an exploration of these contributions, but I would like

to conclude with a brief discussion of poaching, a highly social form of rural

criminal activity, and one of particular salience in East and Southern Africa in an

era of heightening environmental consciousness. There have been two particularly

useful contributions.

Ed Steinhart explored traditions of hunting in Kenya during the twentieth

century, establishing hunting to have been an activity of greater economic and cultural

importance among the peoples of eastern Kenya than has hitherto been recognised.53

He went on to discuss the rise of big game hunting in Kenya, in the early years of the

century, and the institutionalisation in the interwar years of the white hunter.

Hunting brought to world attention East Africas remarkable resources of large

mammals, which eventually became a cornerstone of Kenyas lucrative tourist

industry. In the 1930s, with a remarkable rise in conservationist thinking in colonial

circles, stimulated in part by the American Dust Bowl, the foundations of a national

park system were laid. The system itself came into existence in the years immediately

following World War II. Up to this time, African hunting had survived, in rather more

limited form. But the creation of the national parks meant the criminalisation of

African hunting and major campaigns to eradicate it. In a classical irony, an important

device for undercutting the economic attraction of hunting was the hiring of poachers

as salaried gamekeepers. An activity which had provided protein for the diet of local

Donald Crummey: African Banditry Revisited 32

 EnterText 4.2

people, and revenue through the sale of game trophies and ivory, now became illegal

in the eyes of Kenyas rulers. These rulers, in the 1950s, were, of course, European

colonialists, but their policies were embraced holus-bolus by their African successors,

who found themselves with their own poaching wars to wage.

Kenneth Kaunda, president of the independent African republic of Zambia,

was an avid conservationist and promoted policies similar to those of Kenyas, but

without success. Clark Gibson has explored the collapse of anti-poaching legislation

in Zambia from the standpoint of national institutions. His point of departure is the

anomaly that Zambia was an authoritarian one-party state with an executive leader

with strong conservationist views, but nonetheless failed to pass legislation reflecting

those views.54 The economy of colonial Northern Rhodesia and the early independent

state of Zambia was based on the mining and exporting of copper, but the creation of

the one-party state in 1972 coincided with a collapse of copper prices. Local people

turned to the one resource still accessible to them of value on the world market

ivory and rhinoceros hornsand through the 1970s and 1980s poaching ran at a very

high level, with local markets in game meat also springing up. The activity was a

criminal one, but Kaundas attempt to enact laws which would carry punishments

sufficient to deter it, ran aground in the national legislature, where politicians

reflected local values and refused to cooperate. The scale of poaching was

prodigious75% of Zambias elephants were killed off between the late 1960s and

1989, and the black rhinoceros population of the Luangwa valley went from an

estimated 50,000 in 1972 to fewer than twelve in 1992. Thus, in the latter half of the

twentieth century poaching in Africa has increased exponentially. World attention so

far has focused on the effect on wildlife, but it is not difficult, at the same time, to

perceive the social dimension to this expression of rural criminality.

Donald Crummey: African Banditry Revisited 33

 EnterText 4.2

Africanists still have plenty of terrain to cultivate in exploring the social

dimensions of rural criminality. Hobsbawms Bandits draws their attention to the

phenomenon of the gang, remarkably uniform in its scale and composition over vast

stretches of territory and time. While the sources for, and possibly the incidence of,

banditry in Africa increase with the advent of European imperialism and the

establishment of colonial regimes, Africanists would be mistaken in understanding

bandit activity in Africa purely as a function of resistance to European colonialism.

They would do well to take seriously the spaces created by earlier states on the

continent in which the bold, the disaffected and the ambitious flourished.

1
This is a slightly revised version of a paper originally delivered to the Annual Meeting of the Social
Science History Association in Fort Worth, Texas, November 1999, as part of a panel entitled Eric
Hobsbawms Bandits: A Thirty Year Retrospective. The principal editions referred to here are: Eric
Hobsbawm, Bandits (Harmondsworth: Penguin, 1972, and 2nd ed., 1981). A third edition published by
Weidenfeld and Nicolson (London, 2000) appears to be identical to the 1981 edition. The literature
referenced in this revisitation appeared between 1986 and 1999.
2
See the Preface to the Penguin editions (1972 and 1981), 14. The 2000 edition repeats the 1972
preface verbatim.
3
E. J. Keller, A twentieth century model: the Mau Mau transformation from social banditry to social
rebellion (Kenya Historical Review 1.2, 1973), 189-205.
4
Allen Isaacman, Social banditry in Zimbabwe (Rhodesia) and Mozambique, 1894-1907: an
expression of early peasant protest (Journal of Southern African Studies 4.1, 1977), 1-30.
5
International Journal of African Historical Studies, 17.2, 1984, 201-227.
6
Donald Crummey, ed., Banditry, Rebellion and Social Protest in Africa (London: James Currey,
1986).
7
Thoughtful reviews would include: Stanley Cohens review article, Bandits, Rebels or Criminals:
African History and Western Criminology (Africa 56.4, 1986); Richard Rathbone in the Journal of
Southern African Studies 14.1(1987), 167-168; Allen Isaacman in the American Historical Review 93.2
(1988), 472-473; John Lonsdale in the Journal of African History 29.1 (1988), 111-113; and Tony
Emmett in the International Journal of African Historical Studies 21.3 (1988), 540-541.
8
Chapter 4: Ralph Austen, Social bandits and other heroic criminals: Western models of resistance
and their relevance for Africa.
9
Austen, 91.
10
Chapter 5: Ray Kea, I am here to plunder on the general road: Bandits and banditry in the pre-
nineteenth century Gold Coast. The Gramsci quotation is on p.127.
11
Chapter 8: Robert Gordon, Bushman banditry in twentieth-century Namibia.
12
Chapter 14: Alison des Forges, The drum is greater than the shout: the 1912 rebellion in northern
Rwanda.
13
Ibid., 318.
14
Chapter 17: Terence Ranger, Bandits and Guerrillas: the case of Zimbabwe.
15
Ibid., 377.
16
Ibid., 384.
17
Richard Caulk, Bad Men of the Borders: Shum and Shifta in Northern Ethiopia in the 19th century
(International Journal of African Historical Studies 17.2, 1984), 201-27.
18
Caulk, Bad Men, 214.

Donald Crummey: African Banditry Revisited 34

 EnterText 4.2

19
Ibid., 218. Caulk goes on to quote a parody of a prayer which rustlers offered up: O God, give us
the property of old weak men, the property of the blind and limping, the property of orphans and
women.
20
Chapter 13: Richard Caulk, Black snake, white snake: Baht Hagos and his revolt against Italian
overrule in Eritrea, 1894.
21
Chapter 6: Donald Crummey, Banditry and resistance: noble and peasant in nineteenth-century
Ethiopia.
22
Eric Hobsbawm, Primitive Rebels. Studies in Archaic Forms of Social Movements in the 19th and
th
20 Centuries (Manchester: Manchester University Press, 1959).
23
J. Baeteman, Dictionnaire Amarigna-Franais suivi dun Vocabulaire Franais-Amarigna (Dire-
Daoua, 1929), 449; Amsalu Aklilu, Amharic-English Dictionary (Addis Ababa, 1979Eth. Cal.), 102.
24
Chapter 7: Timothy Fernyhough, Social mobility and dissident elites in northern Ethiopia: the role
of banditry, 1900-69.
25
Ibid., 163. Fernyhough subsequently revised his conclusion in an undated paper, Interpreting
Ethiopian Banditry: A Revisionist View. For a summary, see below.
26
This passage rests on Banditry and resistance, 138-9. The authoritative account of Kassas career
is Sven Rubenson, King of Kings: Tewodros of Ethiopia (Addis Ababa, 1966).
27
Luigi Fusella, ed., Yat Tewodros Tarik (Rome, 1959), translated as La cronaca dellImperatore
Teodoro II di Etiopia in un manoscritto amarico (Annali dellIstituto Universitario Orientale di
Napoli, ns, vol. 6, 1957), 3. The author of this chronicle is unknown.
28
Economist, August 8, 1992, South Africa. Banditry versus politics.
29
Allen Isaacman, Peasants and Rural Social Protest in Africa (African Studies Review, 33.2, 1990),
1-120. He addresses social banditry on pages 38-9, in a section entitled Getting Even: The Politics of
Retribution.
30
The index is actually, banditry, social, suggesting either that simple banditry doesnt exist or that
it is a phenomenon of no interest. Catherine Coquery-Vidrovitch, Africa: Endurance and Change South
of the Sahara (Berkeley: University of California Press, 1988).
31
The starting point for Akira Kurosawas masterwork, Seven Samurai (1954), is the victimisation of
peasants by bandits in pre-modern Japan.
32
Interpreting Ethiopian Banditry, no pagination.
33
David M. Hart, Banditry in Islam: Case Studies from Morocco, Algeria and the Pakistan North West
Frontier (Wisbech, Cambridgeshire: Middle East & North African Studies Press, 1987).
34
Hart, Banditry, 6-12.
35
Ibid., 13-19.
36
Ibid., 20-26.
37
Ibid., 30.
38
Ibid., 31-34.
39
Ibid., 39-40.
40
Ibid., 41-44.
41
Nathan Brown, Brigands and State Building: The Invention of Banditry in Modern Egypt
(Comparative Studies in Society and History 32.2, 1990), 258-81.
42
Brown, Brigands, 260.
43
Gwyn Campbell, The Menalamba Revolt and Brigandry in Imperial Madagascar, 1820-1897
(International Journal of African Historical Studies 24.2, 1991), 259-91.
44
Ibid., 265.
45
Ibid., 267.
46
Ibid., 269.
47
Ibid., 290.
48
George L. Simpson, Frontier Banditry and the Colonial Decision-Making Process: The East
Africas Protectorates Northern Borderland Prior to the First World War (International Journal of
African Historical Studies 29.2, 1996), 279-308.
49
Donald Crummey, Banditry and resistance, 135.
50
See, in addition to Simpson, Abdussamad Hajj Ahmad, Gojjam: Trade, early Merchant Capital and
the World Economy, 1901-1935, unpublished Ph.D. dissertation, University of Illinois, 1986, and
John Lamphear, The Scattering Time: Turkana Responses to Colonial Rule (Oxford: Clarendon Press,
1992).
51
Cited in Simpson, Frontier Banditry, 288.
52
See particularly, P. Scully, Criminality and Conflict in Rural Stellenbosch, South Africa, 1870-
1900 (Journal of African History 30.2, 1989), 289-300; Mark Beittel, Mapantsula: Cinema, Crime

Donald Crummey: African Banditry Revisited 35

 EnterText 4.2

and Politics on the Witwatersrand (Journal of Southern African Studies 16.4, 1990), 750-6; and D.
Johnson, Criminal Secrecy: the Case of the Zande Secret Societies (Past and Present 130, 1991),
170-200.
53
E. I. Steinhart, Hunters, Poachers and Gamekeepers: Towards a Social History of Hunting in
Colonial Kenya (Journal of African History 30.2, 1989), 247-64.
54
Clark C. Gibson, Politicians and Poachers: The Political Economy of Wildlife Policy in Africa
(Cambridge, 1999).

Donald Crummey: African Banditry Revisited 36