 Exploring the Articulation of Governmentality

and Sovereignty:The Chwaka Road and the Bombardment of

Zanzibar, 1895–1896
 Geoffrey Ross Owens
Abstract

In 1896, Khalid bin Barghash seized Zanzibar’s throne, and Lloyd William Matthews
ordered the bombardment of the palace, killing 500 of Khalid’s followers. This article
contrasts the bombardment with earlier incidents in which British-led police were
threatened and attacked. The British administration attempted to avoid overt
confrontation with the perpetrators of these earlier incidents. Utilizing Foucault’s and
Agamben’s theories of sovereignty and governmentality, I argue that these tactics of rule
were applied tandemly to different categories of colonial subjects, and served to produce
and reproduce these categories. This case study reveals creative and complex relationship
between colonial power and knowledge.

Introduction
In 1896, upon the death of Sultan Hamid bin Thwain, his cousin Khalid bin Barghash
attempted to claim the throne of the British Protectorate of Zanzibar. He and a large
number of followers staged a coup d’etat against the British government’s preferred
candidate, and they barricaded themselves in the Sultan’s Beit al Hukum palace, facing
the Indian Ocean. Lloyd William Matthews, “First Minister to His Highness the
Sultan,” ordered British naval ships stationed near Zanzibar to bombard the palace,
killing 500 of Khalid’s followers and destroying the palace (Lyne 1905:202; Thompson
1984 [1961]:64).

This event has been reported by historians, and even celebrated by poets, as a contest
between an encroaching colonial administration and a “pretender” to the Zanzibar
throne (Lyne 1936:147; Mutafi 1957:66; Bennett 1978). Yet upon careful consideration,
Matthews’ decision to resort to a naval bombardment might seem an exaggerated
response. By 1896, the British had already established de facto control over the internal
and external affairs of the Sultanate of Zanzibar. And this was by no means the first
time a member of the royal family had tried to seize the throne. The history of the
Zanzibar sultanate in the late nineteenth century had been punctuated by coup
d’etat attempts, most of which had been resolved through British intervention (Lyne
1905:54; Reute 1888:108). Khalid, for whom this was the second attempt to seize the
throne, was outgunned by British naval forces, and had expressed his willingness to
negotiate by sending an envoy to the British Consul General (Lyne 1905:196; 200).
Having established control of the military, police, revenue expenditure and other
aspects of Zanzibar’s infrastructure, one might not necessarily have expected Matthews
to have chosen such a decisive response to Khalid’s attempt.

Both the 1896 bombardment and the events leading up to it should be of interest
because they provide a lens through which to examine the complex and seemingly
paradoxical nature of colonialism as both an organized system of governance and an
exercise of political will against colonized subjects. More specifically, these incidents
bring into focus the creativity of colonial power in its relationship to colonial
governance. A number of scholars have previously utilized Foucault’s concept of
governmentality (Foucault 1991) to argue that colonialism is a kind of statecraft, whose
principal aim is to generate forms of knowledge about subject peoples (Cohn
1996; Wagoner 2003), enabling colonizers to identify, register and discipline subject
peoples in terms of new or preexisting European colonial constructions of domesticity
(Comaroff and Comaroff 1992), morality (Stoler 1992) and “civilization” (Hall 2002).

Colonial forms of knowledge drew upon a European disposition to divide subjects

into distinct races, ethnic groups or “tribes,” often conceived of as occupying
geographically bound territories within a colony (Comaroff and Comaroff 1992;
Ferguson and Gupta 2002). Such territorially-bound groups resembled taxonomic
categories employed in the natural sciences (Pels 1997:175). With such classification
came a list of qualities or traits allegedly possessed by subject people, which rarely if
ever took into regard the individuality of those subjects. But colonialism was not merely
artfully organizing and governing territories and their subjects. Colonialism invariably
involved conquest, a violent process of subjugation of non-Western peoples that
entailed a simultaneous dismembering of earlier precolonial political institutions, and
their subsequent rebuilding as apolitical bureaucratic appendices of European-
controlled administrations (Berman and Lonsdale 1992:21–30; Pels 1994:326–329).

Certainly such characteristics fit well with what is known and understood of the
turbulent history of the British protectorate of Zanzibar. However, if by the late
nineteenth century British colonial rulers had already developed an effective “art of
government,” modeled upon the transformation of the political economy in Europe
between the seventeenth and early nineteenth centuries, and perfected in colonial forays
in the New World, Ireland and India, then why did the British see the need to resort to
guns so late in the history of their colonial empire? The simplest explanation might be
that Lloyd Matthews’ system of indirect administration of Zanzibar was only tenuously
established, and the process of colonization incomplete: in Foucault’s theory of Western
Europe prior to the seventeenth century, the sovereign was “external” to the principality
he governed, connected to it only through “manipulating relations of force” needed to
identify and eliminate potential threats to the person of the sovereign (Foucault
1991:90).

On the other hand, Matthews responded to another set of earlier incidents in a

distinctly different manner from the subsequent coup. Beginning a few months before
Hamid bin Thuwain’s death, Hamid had built up a personal military retinue which, in
the words of one historian, was involved in “minor hostilities with British‐led police”
(Bennett 1978:178). Although the nature of these hostilities hasn’t hitherto been well
documented, it appears that they began when members of a crew building a road
between Zanzibar City and the east coast village of Chwaka were surrounded and
threatened by individuals allegedly working for Sultan Hamid bin Thuwain. These
events culminated in clashes between police and members of Hamid’s military guard
along the Chwaka Road. Hamid’s death and Khalid’s coup followed on the heels of
these events. In a series of memos nearly accidentally preserved by his biographer,
Matthews went to great lengths to call these incidents a problem of “bad feelings”, a
matter to be settled “quietly”, or a “row.” Though Matthews threatened some of the
perpetrators with arrest, evidence suggests that he had every intention of avoiding
armed confrontation. Arguably, Matthews’ choice of words should be viewed as
integral to an artful strategy in which he wishes to control the population of the outlying
settlements of Zanzibar. Political acts usually threaten colonial rule, while criminal ones
tend to be perceived to be more individualistic and thus less threatening (Stoler
1987:50). If the administration could maintain that such events were unrelated
individual criminal acts, political order was maintained, reducing the need for the
spectacle of violent suppression.

Whether we treat the palace coup and hostilities on the Chwaka Road as related or as
separate, these attempts to quietly control geographically widespread acts of hostility by
colonial subjects preceded a major act of violence. At the very least, the incident seems
to suggest that manipulative relations of force and the art of governance were
interwoven in colonialism. Although Foucault used the concepts of “sovereignty” and
“governmentality” as the beginning and ending of his historical narrative that began in
the sixteenth century, and more or less complete by the early nineteenth, he also
recognized that disruptive acts of state violence were and are integral to the tactics
employed by post-Enlightenment states, along with surveillance, new forms of spatial
organization, and biopolitical control of populations of individuals. Though the modern
nation-state may wish to use force sparingly, it is nonetheless a ubiquitous part of the
tactics of the modern state, even when it remains latent.

Even before Foucault wrote of the concept of sovereignty, it was already integral to
the academic discussion of political history and anthropology, often represented as a
power that sits above or beyond a politically demarcated territory or subject people
(Ferguson and Gupta 2002: 990). There are also forms of what Biolsi (2005) calls
“modular” sovereignty, in which entities smaller than the nation-state appropriate
Western discourses on individualism and rights to assert autonomy from a larger power
(Comaroff 1995: 214–225). Each of these uses is consistent with Foucault’s description
of the sovereign as being largely external to his subjects. There is however a still
different use of the concept sovereignty that does not place power above, beyond, or
outside of a system of law and governance. Agamben (1998) argues that sovereignty is
not a condition in which an individual or state enjoys a monopoly over or autonomy
from power. Instead, following the model of social philosophers Hannah Arendt
(1958) and Carl Schmitt (1985), sovereignty is a force that lies within the law, and yet
enjoys the power to decide when there is an exception to the law—a power that has not
been superseded by other modes of political control, but instead lies at “the very heart”
of the modern state (Agamben 1998:106). Agamben goes on to suggest that subjects of
sovereign power are stripped of their political personhood, and live the “bare life” in the
thrall of sovereign power. Hence, it is not the sovereign, but the subject whose body
exists beyond the pale of true political life and the rule of law (Agamben 1998:182–
183).

In order to better understand the efficacy of colonial rule, one must view it not simply
as an ideal structure, but as a dynamic set of practices in which tactics of
 governmentality and sovereign force were situationally employed in different ways on
different categories of colonial subjects. Foucault speaks of how governmentality
turned subject peoples into deindividualized “populations,” while Agamben cites the
examples of the concentration camp and banishment as the quintessential
manifestations of sovereign power on subject individuals. But such tactics were not
employed willy-nilly against all kinds of subjects. This article builds upon the ideas put
forth by Stoler (1992; 1995) that neither the colonizers nor colonized were
undifferentiated, reified categories. Rather, they were themselves internally
differentiated, resulting in unique and shifting forms of cultural interaction in the
colonial setting. The events in Zanzibar in 1895 and 1896 suggest that racial and ethnic
forms of constructed knowledge of colonial subjects governed and circumscribed which
tactics were appropriate against which categories of subjects, producing and
reproducing differential categories of colonial subject. This article will describe how the
Wahadimu people of Zanzibar, classified as a “tribe” living in a “reserve” in the eastern
portions of the island, were most clearly subject to forms of what Foucault describes as
governmentality, and treated as a population whose protests against a government-
sponsored road building and construction projects were deemed as criminal acts, subject
to police action, but not as a threat to colonial rule.

In contrast, those classified as “Arabs,” especially members of the royal family, had a
different relationship to colonial power, and were thus subject to different tactics. The
sultan of Zanzibar was a recognized agent of indirect rule, under the auspices of the
traditional “Arab State” of Zanzibar (Bennett 1978) founded by Hamid’s and Khalid’s
grandfather. Yet, the British wielded the ultimate power to decide when and under what
conditions Zanzibar was to be ruled by the Arab royalty. The act of insubordination on
the part of a “pretender” to the throne resulted in a form of sovereign intervention by
British colonial officials. Though Khalid’s goal was to press his claim to serve as sultan
and agent of the British protectorate, he was overcome by lethal force and banished
from Zanzibar. Khalid represented what Agamben has called the modern homo sacer, a
man who survived the guns, but spent the remainder of his life in exile.

In order to build this argument, I draw almost exclusively on memoirs, biographies

and archival documents written by members of the colonial administration of Zanzibar,
especially the works of Robert Lyne, biographer of Lloyd William Matthews. While
such documents do not necessarily give voice to the interests, needs and concerns of
contemporary Zanzibaris living under colonial rule in the late nineteenth century, they
offer unambiguous testimony to the machinations of those in power, and clear
illustrations of the manner in which colonial governmentality sovereign power operated
at the time.

The Advent of Political Domination

The sultanate of Zanzibar was originally founded in the early nineteenth century by
Seyyid Said, who came from Muskat in what is now Oman. Having considerable
commercial and military interests in this part of East Africa, he decided to relocate his
capital from Muskat to Zanzibar in the 1830s. As sultan of Oman and Zanzibar, his
realm profited through its role as conduit of the trade in slaves, agricultural products,
ivory and other raw materials coming from the east African hinterland to other parts of
the world (Sheriff 1987). On the island of Zanzibar itself, Said and his fellow Omani
Arabs were successful in establishing cloves as a principal export crop (Cooper 1983).
Zanzibar’s dominance in the cultivation and export of cloves lasted well into the mid-
twentieth century. By Said’s death in 1856, he had established a widespread and
influential East African commercial empire (Gray 1962).

Said’s successors faced the growing influence of European powers in the region,
culminating in the partition of the African continent into spheres of European colonial
influence in 1885 (Coupland 1965). Britain’s influence in particular was most keenly
felt in Zanzibar and Oman, motivated in part by the fact that a substantial portion of the
financial backing and marketing of Zanzibar’s mercantile and plantation trade was
provided by Indians, who were then considered British subjects. Britain successfully
intervened in a succession dispute between Said’s eldest sons, negotiating a treaty
allowing one son to rule Oman, and the other Zanzibar. Through the middle of the
nineteenth century, Britain’s influence further expanded due to the purportedly
humanitarian intervention in the Indian Ocean slave trade. By 1873, Britain had
managed to force upon Sultan Barghash bin Said a treaty banning the transport and sale
of slaves in territories under Zanzibar’s sphere of control.

Around 1890, when Sultan Seyyid Ali ascended to the sultanate, a community of
approximately 150 Germans lived and worked in Zanzibar, representing the largest
European presence on the island. By playing off the Sultan’s fear of German
encroachment, the British persuaded Ali to formally request British protection from
Germany, on the condition that he be allowed to keep his throne. Under the Hegelioland
Treaty of 1890, the British and Germans agreed that the former could declare Zanzibar
a protectorate in exchange for a strategic North Sea island (Bennett 1978:162–169).
Lloyd William Matthews was declared “First Minister to His Highness the Sultan”, but
effectively ruled. Gerald Portal was given the post of British counsel, and put in charge
of the sultanate’s finances. British officials presided over other government offices.
Portal gave the Sultan a set of royal estates, and a fixed annual salary equal to one‐third
of total state revenue2.

In theory, Britain agreed to protect the Sultan’s dominion from German

encroachment, since Germany was known not to preserve local governments in its
colonies. In practice, the British simply formalized dominance over Zanzibar’s affairs
which they had already begun to establish in the mid-nineteenth century (Lofchie
1965:56). Although there was considerable distrust and conflict between the colonial
powers, there was also comradery between the Germans and British in dividing up
Africa into spheres of European influence. Matthews assisted the Germans in
suppressing the coastal rebellion in the region controlled by the German East Africa
Company, and they were able to negotiate by treaty when conflicts ensued. Zanzibar,
declared a free port in 1890, was the stopping point for Europeans engaged in military
and missionary activity, and the political leaders often socialized both in Zanzibar and
on the coast.
 Seyyid Ali, in contrast, found himself isolated, though very much aware of his
eroding control of events. He saw that his salary did not meet his expenses, and realized
that his rule was controlled by his ‘First Minister’ Lloyd Matthews, and Portal, the
British Consul. A number of the Sultan’s followers asked that Matthews be removed
from his office, so Ali wrote an appeal to then Prime Minister Gladstone:
He does not consult me in any matter; he has cut off the allowances of people of mine
who have had the allowances since the time of my ancestors, so that the people’s
hearts have been turned against me owing to these things. Unpleasant things have
continually emanated from him, most of which I have endured, until now the rope
round my neck has become too tight, and my patience is tried, and I say to you and
Her Majesty the queen, who is noted for justice, that I do not want general Matyus
[sic] as governor in my country which is under my charge, and he is not experienced
in them. None of my ancestors were dependent on anyone like general Matyus, and I
shall be glad if you magnanimously remove this man from my service.
(Lyne 1936:125)

When Portal received word of the Sultan’s request, he immediately decided that the
Sultan’s advisors put him up to it. He threatened that the Sultan himself would be
removed if he did not punish the advisors involved (Bennett 1978:177). On March 3,
1893, Seyyid Ali died and was replaced by Sultan Hamid bin Thwain. Hamid at first
cooperated with the British protectorate administration, but according to Hardinge, he
too resented his powerlessness:
Now and then he exhibited a touch of dry humour, and at a later period in his reign,
when he had begun to realize the more galling aspects of his own position as a
protected puppet prince, I had occasion to present to him a distinguished naval officer
who had explored the Polar regions north of Greenland, and who described to him the
short dark days of winter, the midnight sun in summer, the white bears and the
Esquimo dogs. ‘I presume,’ he said gently and yet a little dryly, ‘that the interesting
region which you visited now enjoys, or at least will soon do so, the blessings of a
British protectorate!’
(Hardinge 1927:98)

Hamid went so far as to seek the help of the French consulate, but was rebuffed
(Bennett 1978:177). He then began building up a small military force under his own
personal authority. The administration permitted him to pay his palace guard using of
his own expense account. By October of 1895, he controlled approximately 1000
guards and followers, whom he stationed at his palaces in Zanzibar City, and at Dunga
in the center of Zanzibar Island (Lyne 1936:174).

At the time the Sultan was building up his personal force, a rebellion took place on
the mainland of British East Africa. In February 1895, a succession dispute ensued with
the death of Sheikh Salim bin Hamisi of Takaungu, a town situated between Mombasa
and Malindi in present day Kenya. Under British pressure, Rashid bin Salim was
selected as new sheikh, although his cousin Mubarak bin Rashid, whose father preceded
Salim as sheikh, refused to recognize his rule. In June 1895, Mubarak and Aziz bin
Salim, Rashid’s brother, were given a summons to submit to Rashid’s rule. Three days
 later, Matthews sought to enforce this summons with a force of 300 naval bluejackets
and 200 “native” troops from Zanzibar (Hardinge 1927). “Only superior arms and,
eventually, heavy reinforcements,” allowed the British to defeat the rebels (Koffsky
1977:182)3.

Thus, while the Sultan was amassing armed retainers between Zanzibar and Dunga,
Matthews was forced to send his own troops away to the African mainland to attack
Mubarak and Aziz. In a report to Parliament by Captain A. C. Raikes, who was in
command of “His Majesty the Sultan’s” forces, the army consisted of eight battalions
with a total of two European officers, twenty nine “native” officers and 513 rank and
file. It was these forces that were sent to the mainland to fight Mubarak. The only other
force remaining to carry out administration policy on Zanzibar Island was the police.
For the “native district” of Zanzibar City and environs, the police consisted of one
captain, seven lieutenants and three hundred and sixty six men, who were armed with
truncheons, but were partly trained in the use of rifles (Raikes 1895:136). Matthews
also had at his disposal the British navy, which was engaged in protecting the coast and
islands from illegal slave trading. The naval ships however were of only limited
assistance in inland situations.

The Chwaka Road: Colonial Development and the

Creation of the Reserve
One of Foucault’s lasting contributions was his meticulous descriptions of Western
institutions such as the prison (Foucault 1977). Originally envisioned as an enlightened
system of reform for those engaged in criminal activity, the prison was transformed into
an instrument of discipline and surveillance of the bodies of those incarcerated.
Ferguson (1994:8–10; 194–199) found that Foucault’s discussion of the manner in
which institutions diverged from their designers’ intents also sheds light on problems of
development practice in the non-Western world. Like the Western penal system, which
seems perpetually to “fail” to accomplish its aim of rehabilitating criminals, so too
development projects nearly inevitably fail in what appears to be their principal target:
the alleviation of poverty and increase in productivity of peasants in the developing
world. But while prisons fail to produce law-abiding citizens, and development projects
fail to alleviate poverty, they often have consequences that may be viewed as desirable
outcomes from the perspective of the powerful individuals who promote these schemes.

As with many instances in the Western penal system, and still is with the case with
postcolonial development projects, many colonial-era development projects like the
Chwaka Road project resulted in broad forms of civil unrest that interfered with the
completion of the project. Though one might view such unrest as a failure of planning,
it is also possible to infer that the construction of the Chwaka Road and the problems it
engendered actually contributed to the colonization process in unexpected ways. The
tactics employed in response to these forms of unrest may have facilitated and
reinforced the colonially constructed status of outlying residents of the protectorate as
part of a population “inclusively excluded” from political life in the colony. And the
road, once completed, made it possible to both send police and military officials to
these outlying populations, and facilitate their transformation into a pool of transient
laborers in the burgeoning colonial plantation economy of western Zanzibar.

When a plantation economy based upon the cultivation and export of cloves was
established on the western portion of Zanzibar Island in the early 1800s, the Arab
newcomers expropriated agricultural land in the western lowland belt of the island,
forcing many of the Swahili‐speaking residents of the island to escape to the stony
eastern coral region (Sheriff 1991:112). According to John Gray, one‐time chief justice
and historian of colonial Zanzibar, the eastern and southernmost parts of the island were
a “reserve,” protected from direct contact with the Arabs and Indian settlers who
occupied the more fertile portions of the island (Gray 1977:148). Other ethnic groups of
Zanzibar and the British considered the Wahadimu, who lived in these regions the
indigenous population of Zanzibar Island (Middleton 1961:8–9).4

The Wahadimu were thus viewed as a population living in a “reserve,” physically

and politically separated from political life in Zanzibar by “mangrove swamps and coral
rag jungle, almost impassible” (Lyne 1936:169). But with the development of the
Chwaka Road, as it was proposed in 1894, the road would connect the eastern port of
Chwaka to the port of Zanzibar City, previously inaccessible over land because of the
jagged coral rocks that cross the center of the island and allow easy access to the richest
plantation belts in the north and south of the island. As an additional benefit, the
Wahadimu, “once they are in direct touch with the town markets will find their wants
increase, and of necessity be forced to work harder and cultivate to a greater extent
those parts of the rocky belts which bring forth red peppers in abundance, and other
products” (Portal 1895:124).

By August of 1895, a road had been completed between the town of Zanzibar and
Dunga, located about 11 miles east of the town (Anonymous 1896:158). Within about
three to four miles of Dunga, the more fertile belt gave way to rocky coral soil:
Starting from the north eastern end of the capital and passing Walezo and the river
Mwera, which it crossed by a bridge, it reached, after traversing rich mango, banana
and clove plantations, some twenty miles from Zanzibar, the large pile, known
afterwards as Dunga Palace... a little further on, at Ndajani [Ndijani], the cultivated
region is exchanged for the coral soil, on which only tangled bush could grow...
(Hardinge 1927:540)

The road crew was made up of Christians, probably former slaves employed from the
Universities Mission to Central Africa station in Mbweni, and under the leadership of
Joseph T. Last, a Church Missionary Society officer who was appointed as
commissioner of slavery for Lloyd Matthews’ government in 1894 (Firminger
1896:170; Lyne 1936:169). In addition to clearing bush, grinding stone and applying it
to the road, they planted milestones at appropriate intervals. The milestones were
rectangular stone blocks, about four feet high and one and one-half feet wide, with an
inscription in both Arabic and English, and capped with an eight inch pyramidal top
piece. The portion of the road between Dunga and Chwaka was probably finished
around April of 1896, although people were passing as early as December of 1895
(Zanzibar Gazette 1896). According to Lyne, the Wahadimu resented the fact that the
government was building a road in their territory, and began breaking the milestones.
So, Matthews offered a R50 reward to anyone who was able to capture the vandals. He
also threatened to arrest the local headmen of villages situated along the road (Lyne
1936:210).

When Matthews acted, the headmen called off the milestone breakers. We know little
more about these people or their motives. However, if we were to speculate that their
action was an outward and visible sign of an inward and festering resentment towards
colonial authority, is it possible that there is an explanation? Fortunately, there are a
number of descriptions of the region available, and it should be possible to piece
together an explanation for their acts of vandalism.

Hardinge, the Consul General at the time of the coup, complained that the jagged
coral rocks cut through his thickest leather boots, and Fitzgerald mentions that the
locals of the area used “specially‐made coir sandals for traversing this barren coral
country” (Hardinge 1927:114; Fitzgerald 1898:540). To the administrators and
explorers who described the country upon which the road was to be built, the land was
“barren,” “impassable,” where “only tangled bush could grow.” It was a common
practice in colonial territories to treat such land as available for public works; an
individual occupying and using such land would gain proprietary rights over it, while
abandoned and unused land would be considered available for public works projects.
However, for the people who lived in this region, and the coral land was an integral part
of their agricultural system.

Not only was this allegedly barren country capable of supporting people, in the
eastern zones of Zanzibar Island, the Wahadimu “were self‐sufficient to a considerable
extent” (Sheriff 1991:112). The stony coral regions were cultivable, and farmers
successfully break up the coral with heavy iron tools, and plant crops in the limestone
gravel. At the end of the nineteenth century, the Wahadimu were making a lucrative
living amongst the coral, growing crops which could only have been used for exchange
on the world market. Consider, for example, the following description of the precise
region where the Chwaka road would later be located, based on Fitzgerald’s journey in
1892:
At 8:30 AM I came upon another small chili plantation belonging to the Wahadinu
(sic) people; the ground was so covered with a jagged coral outcrop bleached white by
the sun... Tobacco with very small leaves and pinkish‐white flowers seemed as much
grown as the pepper plant; pumpkins and tomatoes also flourished
(Fitzgerald 1898:540).

Such significant quantities of chili, tobacco and tomatoes must have been produced
for markets in the city of Zanzibar and beyond. Lyne observed that “the bulk of
Zanzibar chillies [sic] go to London, though New York exercises a controlling influence
in the market.” Because they were difficult to start in a nursery, and to harvest, it was
difficult for a plantation owner to grow chilies profitably (Lyne 1905:253). Therefore,
people were left to cultivate them on individual holdings. As Fitzgerald proceeded, he
also came upon what must have been a piece of kiambo garden land where more
traditional use‐value crops were cultivated:
 As I proceeded, cultivation reappeared, with clumps of cocoanuts, scattered mangoes
and fields of rice and maize, and very fine and luxuriant fields of chillies grown round
the houses of a Wahadinu village
(Fitzgerald 1898:540).

Portal mentions that produce was carried by local people on their heads, to the town
(Portal 1895:124). Even food crops like rice and maize were sold to Indian merchant
middlemen. Therefore, this country, described by administrators as a barren coral and
bush covered land, managed to provide for the people a substantial living, tying them
economically to the larger community and world markets. The notion that the
Wahadimu lived in a barren reserve was contradicted to some extent by evidence that
they were fully immersed in a global market economy.

Of course, not all of the bush and coral land was being utilized as agricultural land all
of the time. The Wahadimu practiced shifting cultivation, and it is possible that the bush
that Hardinge observed may have once been a cultivated field, now fallow. But even in
the stoniest areas where the soil could not support crops, the Wahadimu must have
claimed proprietary rights based upon a communal rule of land tenure (Middleton
1961). Thus, economically, the coral country was a socially and economically viable
community, which regularly interacted with the greater Zanzibar society.

But the attitudes expressed by colonial administrators that the land was barren
became a rationale for ignoring the claims of cultivators whose land would be traversed
by the road. The road crews built right through kiambo land belonging to local people,
as one administrator, who conducted research in Chaka reported in the 1940s:
From another source I heard of the demolition, by a government road gang some 35
years ago, of the wall on two sides of a cultivator’s field, which obliged him to
abandon the place. The laborers were no doubt from Zanzibar town, to whom one
stone was as good as another, but it illustrates the circumspection which is necessary
when working in new conditions among a strange community.
(Pakenham 1947:22)

The events being referred to may actually relate to the road building project of the
late 1890s. It is possible that, what Matthews saw as a criminal act of vandalism was in
fact motivated by fear on the part of locals that their land was being confiscated and
rendered uncultivable.

In spite of the actions of some members of the community, the roadbuilders seem not
to have been deterred by the presence of human activity in the “barren” coral country.
In Agamben’s theory of sovereignty, the paradigmatic model for the bare life is not the
prison, asylum or factory, but the camp, a location where refugees and the politically
dispossessed exist in a state of “inclusive exclusion,” where an apolitical disregard for
the interred represents the norm (Agamben 1998: Norris 2005:270–271). In the colonial
context, one might be able to substitute the idea of the reserve for the camp, since it
both exists by administrative fiat, yet is often viewed as the location of a group
“superfluous” to the colonial state, outside the political life of the colony itself
 (cf. Blecher 2005:730). The laborer exists within the reserve, and has but the most
tenuous connections to the colony to which he provides labor— only a body in service
at the convenience of the colonial plantation owner.

The Town of Chwaka and Dunga Palace

The town of Chwaka, on the eastern coast of Zanzibar Island, was the terminus of the
proposed road. Hardinge called Chwaka a “miserable hamlet” (1927:114). In spite of
Hardinge’s rather negative assessment, it was an important trading center in the 19th
century. But of more direct interest is that Chwaka and nearby Dunga housed a
substantial number of Hamid’s personal military retinue, and this area of the Chwaka
Road became the staging area of the clashes between these forces and British police.
Lyne preserved a number of memos hastily written by Matthews during this period, for
the purpose of illuminating the latter’s skills as a “master of the Orient.” But when these
memos are arranged chronologically, they also reveal evidence of a widespread and
coordinated contestation between the British and Zanzibaris. What is striking is the
extent to which Matthews attempts to refrain from military intervention, in
contradistinction to the events surrounding the palace coup several months later.

Prior to the building of the road, Chwaka communicated with Zanzibar City and
other regions by sea, and it was a trading depot for plantations and villages on the east
side of the Island. Towns such as Chwaka consisted of several villages, and each village
had several wards. Chwaka had a single Ijumaa Mosque, and a number of smaller ones
for daily prayers. The pre‐Omani rulers used to appoint a Sheha or local headman, and
the Zanzibar Sultan also recognized the Sheha, and he appointed a First Sheha,
or Sheha wa serikali, and a second assistant Sheha, or Sheha wa mji. Although a few
Arab or Indian shopkeepers lived in the towns, kinship was often the basis of settlement
(Middleton and Campbell 1965:30). However the Sheha was undoubtedly aware of the
Sultan’s military buildup, as some of his military forces had been housed at Chwaka in
“browned roofed whitewashed buildings” (Zanzibar Gazette 1896:1).

At the terminus of the road, the crew began building a palace for the Sultan, a
sanatorium and convalescent house for the benefit of Europeans. The parts for the
palace and sanatorium, imported from Norway, were assembled in the style of a
Norwegian chalet (Zanzibar Gazette 1895), and the buildings still stand in Chwaka
today. Matthews himself seemed to have a personal financial stake in revenue generated
by the proposed convalescent home, because he sought financial capital from
Europeans, including his own brother, to build and manage it (Matthews 1895). On
May 28 1895, Matthews went to Chwaka to inspect the progress on the road. There is
some reason to suspect that Matthews anticipated a confrontation, because according to
the June 1 letter, Matthews sent a Comoro sergeant and four policemen to assist Last
(Lyne 1936).

Lyne does not preserve any correspondence between the June and the end of
September, when the soldiers loyal to the British administration were being sent to
Kenya to attack Mbaruk bin Rashid. But the correspondence resumes on September
27th because of a conflict between Last and the Sheha of Chwaka. Lyne speaks of an
incident where the Sheha of Chwaka tried to intrude in the building of the sanatorium.
Matthews considered sending police that day, but hesitated so as not to “look
aggressive” (Lyne 1936:171). Lyne suggests that the Sheha was the “Sultan’s man,”
and was emboldened by the Sultan’s buildup of a personal force (1936:170):
Occupy and build at once on the spot at once and carry on with all other sites even if
you put up only four poles, one at each corner of the house... If the Arab will build on
a site chosen by yourself [Last] and state in writing that he will be guided by any
regulation that comes into force, he may, with your consent, build‐ but not without
(Lyne 1936:170–171).

On September 29th, Matthews wrote three memos to Last. The first mentions that the
Sultan ordered the first and second Sheha to go to the Town. Matthews goes on to say
that the Sultan appeared to be “very annoyed,” presumably by the actions of the Sheha.
Matthews instructed Last to write a letter to the Sultan, saying that you are certain that
if it had not been for your workmen you would have been attacked...I want him [the
Sultan] to know that forty or fifty men surrounded you, and all that they said about
Wazungu etc; and their threatening attitude (1936:172–173).

Almost fifty years after the incident, a British administrator interviewed residents of
Chwaka, and learned that the building of the palace and bungalow drove away residents
of an entire ward. Lyne’s account makes no mention of the fact that residents were
forced to move. Nor does it mention that as the “Sultan’s man” and local leader, the
Sheha, probably would have had the task of articulating the people’s resentment to the
government:
Shamba-la-Pwani has quite a story of its own. It begins after the wall and gateway
marking the northern end of the premises of the Sultan’s bungalow, and extends
northwards along the coast for about 300 yards and to a depth of some 50 yards
(mainly between the low escarpment and the beach), with a further depth up to 150–
200 yards of good garden land west of it... The community had increased to about a
score of huts when, after the death of Umbaya Vuai and his brothers, Mjengwa Usi
moved to Chwaka and the whole village followed him. This return migration occurred
about the year 1900 and the exodus is said to have been completed within the space of
about a month: the houses were demolished and the building poles sold as firewood if
unserviceable for rebuilding. No doubt this precipitation was occasioned by the
completion of the main road and of the lower and Norwegian bungalows and the
commencement of the sultan’s bungalow on their very doorstep.
(Pakenham 1947:16)

As mentioned earlier, Lyne calls the Sheha of Chwaka the Sultan’s man, and it is
likely that the individuals who threatened Last were among the troops loyal to the
Sultan. However, let us speculate that, for a moment, that the Sheha is his own man,
and that, although he is allied with the Sultan’s play to rid the islands of British rule, he
has his own personal and class interests within his district. It is possible that the Sheha
and other local leaders, alienated and angered by a callous British administration, and
seeing that the Sultan was attempting to resist that administration, might ally
themselves with him and his troops.
 One final important stop on the Chwaka Road was the settlement of Dunga. Located
east of Zanzibar City, the Sultan had built a palace at Dunga, and began to discreetly
hide part of his personal military force there, away from the watchful eye of the British
administrators. Lyne does not leave record of any correspondence between the
September incidents at Chwaka and December of 1895. In fact, Last had a large
celebration on Christmas day at Dunga palace, and Hamid’s soldiers engaged in mock
battles to entertain the revelers:
Now took place one of the events of the day a display of military evolutions by the
household troops of the Palace, in which imaginary attacking forces were repulsed by
an imaginary fire from imaginary guns, all being characterized by intense earnestness
and calculated to impress the onlookers with the potentialities of the defending force,
which if the scene were translated into reality would be in no way a contemptible one.
(Zanzibar Gazette 1896:2)

These scenes had already been translated into reality when, on December 18, the
sultan’s forces “came to blows” troops loyal to the administration. In this clash,
eighteen of the Sultan’s guard and 23 government troops were injured (Lyne 1936:178).
On December 29th, Lyne mentions that the “mischief” had “spread” to Dunga. In
another memo, Matthews assures Last that his men would be “more than sufficient” to
quell the conflict. Matthews tells Lyne, with a touch of unintentional irony: “There can
be no organized attack on Dunga. It is merely a bad state of feeling. But I will send you
more rifles and ammunition” (Lyne 1936:173).

Incomplete as they are, what is striking about both Matthews’ direct correspondence
about these incidents, and Lyne’s recollections, is that there seems to be a mix of
concern that violence might ensue; yet, each also attempts to downplay the gravity of
the situation. Matthews’ control of discourse was almost as significant as his control of
the army and police in turning threatening political movement into a seeming pause in
history. When personally confronted by the British consul, Hamid denied an impending
attack on British forces and offered to reduce the size of the palace forces. According to
Lyne:
Could there be smoke without fire? To which the Sultan replied “that smoke was
sometimes occasioned by dust, and that dust was gossip of the streets” which neither
of them should deign to notice; but nevertheless he would reduce his force.
(1936:178)

Matthews’ actions, and the manner in which he defines the terms of the problem, are
illustrative of the principals of governmentality in the context of colonialism. Britain
instituted what it termed a system of “indirect rule” over subject peoples—delegating,
in theory at least, important responsibilities of governance to representatives of
precolonial political institutions they found at the time of conquest. And the sultanate of
Zanzibar was an example of such a political institution. In practice however, the British
did not merely delegate, but controlled critical aspects of social, political and economic
life of the sultan and his subjects. The ordinary people of Zanzibar, who attempted to
express their grievances against public works and colonial rule, were dismissed as
mischevious, or at worst, criminal—hence depoliticizing their acts against
representatives of the British administration.

About eight months after these incidents, in August of 1896, Hamid bin Thuwain
died, and the palace coup commenced. According to Lyne, Matthews felt inclined to
“shoot Khalid on the spot”, but instead chose to withdraw from the palace. Khalid sent
notices to foreign consulates stationed at Zanzibar requesting that they recognize him as
sultan. Acting British Consul Basil Cave and navy Admiral Rawson sent an ultimatum
to the sultan, and began landing marines near the palace. Lyne provided his own
perspective on the decision to bombard:
The triumvirate—Cave, Rawson, Matthews—were no doubt influenced by the attitude
of the Foreign Office, which from the first had been kept informed of the progress of
events by cable. In answer to one cable asking “Are we authorized in the event of all
attempts at a peaceful solution proving useless, to fire on the Palace from the men-of-
war?” Lord Salisbury replied, “You are authorized to adopt whatever measures you
may consider necessary, and will be supported in your action by Her Majesty’s
Government. Do not, however, attempt to take any action which you are not certain of
being able to accomplish successfully.
In the face of this telegram, characteristic of the British Government at its best, it is
not surprising that Cave and his advisers decided against half-measures. Since the days
of Pangani and Tanga the atmosphere had been charged with anti-European feeling,
which had grown in intensity under the malignant influence of Hamed bin Thuwaini.
Forbearance, interpreted as weakness, would have let loose the forces of disorder and
postponed the reckoning which some day must have come
(Lyne 1936:145–146).

Having received permission from the foreign office, Rawson began to bombard the
palace. Khalid managed to escape, and was able to find his way to the German
consulate. He was eventually allowed to leave Zanzibar, and spend the remainder of his
life as an exile in Dar es Salaam, located in German East Africa (Lyne 1905:199–204).
The bombardment punctuated a very important phase in the history of Zanzibar.5

Khalid’s exile, in contrast to the incidents near the Chwaka Road, represents a
manifestation of the sovereign rule inherent in colonialism. Though British
administration was purported to be indirect, they retained the ultimate decision as to
whether or not the throne of Zanzibar remained in the hands of a successor to the first
sultan of Zanzibar. And they also had the power to declare exceptions—situations in
which the rules of those traditional claims to authority were to be suspended. In
Agamben’s political theory, Homo sacer is an ancient category of person, who could be
killed but not sacrificed. He is stripped of his personhood, and life is “bare” in the sense
that he is forfeit by anyone without sanction. If we were to transfer this notion to a more
secular realm of colonial politics, then banishment would represent this idea well.
Khalid lost his status within Zanzibar as a member of the royalty through the sovereign
act initiated by Lloyd Matthews. He went into exile—dispatched to the sphere of
German influence across the Zanzibar Channel, where he spent the rest of his life in Dar
es Salaam.
Conclusions: The Creativity of Governmentality and
Sovereignty
It is certainly not new to argue that forms of governmentality and acts of political
sovereignty were employed in tandem in colonial situations like that of Zanzibar in the
late 1890s. Dirks (1992:4) has shown that colonialism’s success was predicated upon
both superior arms and vastly complex “cultural technologies” of rule that facilitated
the process by which colonial subjects internalized their subjugation as a natural and
inevitable part of their everyday existence. What distinguishes this particular Zanzibar
case study is the stark manner in which acts sovereign power and governmentality were
differentially applied to two different categories of colonial subjects. Khalid, classified
as an “Arab,” was subject to sovereign intervention through lethal force. For the
Wahadimu, the problem was that they were viewed as a tribe living on a reserve,
separate from the colonial political economy, and in need of police intervention so as to
be under surveillance and in regularized contact with the rest of the colony. The
assumption that they were in a separate sphere belied the fact that prior to the
construction of the Chwaka Road, they were already fully immersed in transoceanic
produce markets and had hosted members of the Sultan’s military guard within their
territory. These incidents demonstrate the efficacy of power to produce and reproduce
cultural categories of person in a colonial context.

When scholars apply Foucault’s concept of governmentality to colonialism and

postcolonial development in non-Western settings, they base such an assertion on the
principle that colonialism is an extension of post-enlightenment forms of liberal
governance. In the period in which this form of governmentality first emerged,
European powers had already begun colonial expansion, in both the New World, and in
the British case, in Ireland and India. To a considerable extent, the emergence of
governmentality and colonialism historically overlap, and should thus have a
relationship to one another. Or viewed another way, one can argue that tactics of
governmentality were so encompassing and successful in Europe that colonial powers
exported their tactics of governance to their colonies, through which Western
governmentality expanded into something global in reach and scope. Hence, whatever
the relationship between early forms of colonial conquest and European
governmentality, they clearly emerged in tandem, and thus it is plausible to infer
resemblances between colonial administration and Foucault’s specific use of the
principle of governmentality as a model to analyze Western European polities.

Even in the most liberal forms of the modern democratic state, there remain segments
of the population deemed unworthy or unfit to enjoy the full constellation of juridical
rights and privileges associated with governmentality (Dean 1999:134). Such unfit
segments are often subjects of sovereign intervention. What differentiates liberal from
totalitarian states is not the presence or absence of sovereignty or govermentality, but
the manner in which they are articulated vis a vis different segments of the population.
In the most illiberal states, like Nazi Germany, this formula was taken to an extreme,
and “the administration of life and the right to kill entire populations” made it essential
for the perpetuation and well-being of one race to be predicated on the annihilation of
other races (Dean 1999:141). Thus in any form of post enlightenment state,
governmentality and political sovereignty is constructed upon the edifice of
differentiating subjects into different categories who are themselves subject to
differential degrees of governmentality and sovereign intervention.

The colonial state represents an interesting intermediate category because its entire
existence is premised upon being an “alien other” to the Western nation-state (Young
1994:44). As members of alien races, “tribes” and nations, colonial subjects were
perpetually outsiders to the metropole, even when their lives were inextricably tied to
the metropole’s political economy. Britain typically instituted what it termed a system
of “indirect rule” over colonial subjects—delegating important responsibilities of
governance to representatives of precolonial political institutions they found at the time
of conquest. And the sultanate of Zanzibar was an example of such a political
institution. However, the British did not merely delegate, but controlled critical aspects
of external international diplomacy, internal finances and the economic life of the sultan
and his subjects. So, colonial subjects embody what Agamben described as sovereign
subjects—they were “inclusively excluded” from full participation in the juridical
privileges of the populations of the colonial metropole.

Agamben’s theory is thus also useful in understanding how the general infrastructure
of colonial power affected colonial subjects. When Agamben attempts to cite historical
examples to illustrate examples of political sovereignty on subject peoples, he drew
disparate ones from many historical eras, including the Homo Sacer and Pater
Familias of ancient Roman jurisprudence, the Medieval European decree of
Banishment, and the modern concentration camp. All of these institutions have a
resemblance—but the examples and the historical contexts to which they belong are
also quite different from one another. This might suggest that what Agamben describes
as sovereign power is not a unitary force, but might creatively serve to fragment and
fracture the forms of political personhood when exercised over a subject people. Thus,
when differentially distributed over subject peoples, sovereignty leads to differential
degrees of participation in political life. In this way, sovereign acts help create and
shape manifestations of colonial identity, in ways that turn out, in the end, to serve
power in useful ways.