Abstract: Most of the Swahili port cities that occupied the western rim of the medieval
Indian Ocean long-distance trading system were founded on islands. Dating from as
early as the 10th century CE, Lamu and Zanzibar have become ‘museumified’ as World
Heritage sites, while other island port cities, such as Kilwa and Pate, are now uninhabited
ruins. Mombasa Island, however, despite numerous calamities, is an increasingly important
commercial hub and gateway into East Africa. This study aims to determine how some
intrinsic benefits of islandness have shaped the settlement patterns and architectural
forms that embody this continuous process of urbanisation. A typological analysis
serves to explore Mombasa Island’s layers of spatiality and morphology. More than
any other East African city, this island reflects the synthesis of the distinctive
settlement traditions of the Swahilis, Portuguese, Omanis and, subsequently, the
British colonisers, through to its current state as a dynamic, modern urban centre.
Compact, complex, and culturally diverse, its unique island concepts offer a wide
range of contemporary urban and architectural solutions.
Keywords: Colonial island urbanism, Indian Ocean impacts, island cities, Kenyan coastal
architecture, Mombasa Island, port cities, Swahili Coast
Cite as: Steyn, G. (2015) ‘The Impacts of Islandness on the Urbanism and Architecture
of Mombasa’, Urban Island Studies, 1, 55-80.
The Impacts of Islandness on the Urbanism and Architecture of Mombasa
The Senegalese scholar, historian Sékéné Mody Cissoko (1986: 17), regards just three
instances of urbanisation in pre-colonial, sub-Saharan Africa as “significant”: Sahelo-Sudanese
cities, Yoruba-Benin cities and the Swahili coastal settlements. The Yoruba-Benin
city type developed in situ as an adaptation of territory and village traditions
and, as regards appearance and functions, demonstrates considerable continuity with
the rural countryside. The first and third types were cities linked to long-distance
external trade and introduced new architectural and settlement forms. Cissoko writes,
“Despite the decline of these [Swahili] cities at the end of the Middle Ages, the
urban tradition remained alive on the east coast.” No East African urban entity is
more representative of this resilience than Mombasa Island.
The legendary archaeologist, Neville Chittick, “characterised the Indian Ocean as
the largest cultural continuum in the world” (qtd. in Sheriff, 2014: 1). For several
millennia and certainly long before Portuguese intervention at the end of the 15th
Century, trade, social interactions and cultural exchange took place all along its
rim (Sheriff, 2014: 2-3). While Chaudhuri (1985: 10) depicts Indian Ocean trade on
a trans-continental scale, Sheriff (2014: 14, 18) points out that delimiting the
“cultural area” to the maritime belt around the rim is possibly more realistic, picturing
the rim as an “intricate network of islands and peninsulas.” Chaudhuri (1985: 3)
was correct, however, in proclaiming that “means of travel, movements of people,
economic exchange, climate, and historical forces created elements of cohesion,”
adding: “The wind system known as the monsoons brought the whole area within the
operation of a single global variable” (Chaudhuri, 1985: 23) While the predominantly
island-based Swahili port cities constitute the western edge of the medieval Indian
Ocean trading system, they nevertheless represent the centre of the Swahili world
Figure 1: Indian Ocean trade as imagined by Chaudhuri, and the Swahili realm of influence
Sujit Sivasundaram (2015: 2) laments, “Islands were critical in birthing our modern
world, and yet they have often been forgotten in our accounts of world history.”
However, with Africa having the second lowest density in number of islands (Sicking,
2014: 490), and considering the significance of the Swahili who, as a pre-colonial
urban society, were primarily settled on islands, it is impossible to ignore the
impact of islandness in any study focused on the East African Coast.
In his in-depth study entitled Mombasa: An African City, Harm De Blij (1968: 49-52)
describes Mombasa as “an island in more than the real, physiographic sense.” He notes
its uniqueness in the East African context mainly because its “island-focussed urban
pattern may well be exceptional for British-influenced Africa.” Aspects that can
be directly attributed to its islandness include high population density, considerably
less distinct land-use zoning, less class segregation, closer proximity (between
functional elements) and more building types than in other African cities. In fact,
De Blij (1968: 65) claims, “the residential areas of Mombasa present almost every
conceivable variation existing in sub-Saharan Africa.”
Figure 2: Swahili Coast in its historical and geographical Indian Ocean context (author’s
Since the early centuries CE, merchants from the Arabian Peninsula and the Persian
Gulf, travelling with the seasonal monsoons, have been visiting the East Coast of
Africa. The subsequent mingling of Arabs (and, purportedly, Persians too) and locals
produced the Swahili culture that emerged in the course of the 10th century. This
culture was predominantly Islamic, mercantile and urban, concretised by its distinctive
stone houses and towns. These towns stretched from Mogadishu in the north to Sofala
in the south, a 2400 km expanse of coastline, also referred to as the Swahili Coast
Over time the Portuguese, Omanis and thereafter British and German colonisers built
their own settlements along the coast, all in their own distinctive urban and architectural
typologies. After independence, the Kenyans – arguably informed more by globalisation
than a quest for regional identity – added their own. Together with indigenous vernacular
forms, these collectively constitute what Robert Baron (1994: 263) calls an “inherited
urban morphology.” The Swahili Coast is hot and humid throughout the year, but although
it straddles the equator, the average temperatures are moderate because of the ocean
This study investigates the way in which the intrinsic nature of islandness has been
shaping Mombasa Island’s settlement patterns and architectural forms. The central
argument is that Mombasa Island was initially settled because of the ancient Indian
Ocean trade and that its resilience and ultimate development into an island city
are mainly due to it being a spatially defined, safe and strategically located Indian
2. Method and sources of information
Louis Sicking (2014: 491) reports that “the preoccupation of the social sciences
with contemporary problems related to islands” has dominated this field of research.
At the same time, Godfrey Baldacchino (2008) and Pete Hay (2006), amongst others,
identify a number of challenging issues inherent to studying islands, pertaining
to objectivity, taxonomy, terminology and identity.
Methodologically, this study is essentially a typological analysis. The technique,
widely accepted in architecture and urban design, relies on reducing drawings and
diagrams to their relevant essentials. In one of the most authoritative manuals on
the topic, Precedents in Architecture, Roger Clark and Michael Pause (2005: v) state
that this technique focuses exclusively on the “formal and spatial realm of architecture.”
Julia Robinson identified four ways to classify the typology of environments (1994:
185). Architects tend to focus on physical properties (the configural type) and are
often criticised for neglecting how environments are made (the genetic code), how
environments are used (the functional mode) and how environments are understood (the
associational mode). However, these aspects are always intrinsic to the configural
type. As Anthony King (1994: 142) argues, the physical and spatial environments are
themselves significant, since they are the settings for social, cultural and political
taxonomies: “The material reality of the built environment, the physical and spatial
world in which those systems have been constituted, is an essential part of the discourse.”
In other words, configural types are the essential elements of urban and architectural
forms and imply time, space, history and cultural conditions.
The research method comprises the cross-referencing of some of the prominent benefits
intrinsic to islandness: territorial benefits, defence benefits and transport benefits
(as defined by Adam Grydehøj, 2015: 3) with Mombasa’s positive urban attributes (as
per De Blij). By chronologically tracking these aspects, the resulting grid is a
convenient morphological matrix to guide research (Table 1). The research question
then becomes simply: How have the benefits inherent to islandness been shaping the
built environment on Mombasa Island over time?
The author collated information from a plethora of publications, including books,
articles, dissertations and reports. Two recent study trips to Mombasa Island provided
Table 1: Research framework (compiled by author).
3. Mombasa Island in context
Mombasa is an island in the Indian Ocean and Kenya’s second largest city after the
capital, Nairobi. By the 12th Century, it was an established Swahili trading town,
first described by the Arab geographer Al Idrisi in 1151 and subsequently by the
Moroccan traveller, Ibn Battuta, who visited it in 1331. The inhabitants, white and
black Moors (pure Arabs and half-castes), according to Justus Strandes (1961: 79)
were Swahilis. Ivory, leopard skins and slaves were among the exports to India and
the Arab Peninsula, while porcelain, metal artefacts and cloth were amongst the imports.
Mombasa at that time was the centre of the Indian Ocean cloth trade (Pearson, 1998:
Vasco da Gama arrived in 1498: Because the city, like most other Swahili settlements
(but unlike Malindi, which provided a pilot to guide the Portuguese to India), was
reluctant to cede its trade to foreign overlords, the Portuguese attacked and destroyed
it in 1505, 1526 and again in 1589. The Omani Arabs (historically key players in
the Indian Ocean trading system) subsequently defeated the Portuguese and conquered
Mombasa in 1698. Apart from a short Portuguese interlude (1728–1729), it remained
occupied by Omanis, albeit by the rogue Mazrui clan from 1735-1737. The local people
helped Sultan Sayyid Said of Oman to oust the Mazruis; he subsequently granted Mombasa
rights and privileges denied other coastal settlements under Omani hegemony. The
Sultan, who had moved his seat of state from Muscat to Zanzibar in 1840, directed
the Omanis’ main interests: the export of slaves and large clove-producing plantations,
which also relied on slave labour.
The Sultan ‘leased’ Mombasa to the British in 1887, and it became the capital of
the British East Africa Protectorate, but Kenya achieved independence in 1963. The
World Bank (2010: 16) recently declared unambiguously, “The Port of Mombasa is the
largest in East Africa and a vital gateway for imports to Kenya and its neighbouring
4. Territorial benefits
Arab traders probably founded the first settlement on Mombasa Island during the 11th
Century: Its original Arabic name is Manbasã (Battuta & King, 2005: 21). Since most
other Swahili cities, such as Lamu, Takwa, Pate, Pemba, Zanzibar and Kilwa, were
also located on islands it seems as if their founders were aware of the benefits
of settling on small littoral islands (Hoyle 2000: 3). As Catherine Coquery-Vidrovitch
(2005: 127) suggests, “These sites were obviously chosen to protect traders travelling
by sea.” This is not surprising: Peter Mitchell (2005: 113) maintains that in the
11th Century, “several interlocking networks emerged connecting East Africa with
other parts of the Indian Ocean world.” Being an integral part of this trading system,
these islands were linked to others such as Hormuz, Bahrain, Aden, the Comoros, Mauritius
and Madagascar (Rockel, 2014: 104).
Regarding the relationships between Swahili communities and mainland tribes, Michael
Pearson (1998: 20) believes that, “Social, political, and economic relations were
all, whether friendly or hostile, advantageous or not, overwhelmingly with their
fellow Africans to the west rather than with the people and places across the sea.”
There is very little evidence to support his position. More credible is Abdul Sheriff
(2014: 32), who states simply that littoral communities on the rim of the Indian
Ocean depended more on other ports across the ocean than on their hinterlands for
“their livelihood and prosperity,” and adds that they also often had more in common
with these. Therefore, in the case of Mombasa, “the islandness of the region and
the role of the sea,” to quote a phrase by Grydehøj et al. (2015: 8), cannot be overemphasised.
Colin Breen and Paul Lane (2003: 479) stress unequivocally that, in the case of Mombasa,
“the sea as a facilitator of trade, communications and resources has been an underlying
constant in a period of continual cultural flux.” In addition, as Rosemary McConkey
and Thomas McErlean (2007: 101) contend, “It would be difficult to view any part
of Mombasa as anything other than a maritime landscape. All aspects of the cultural
heritage could be regarded as having been directly or indirectly influenced by association
with the sea, and in many cases sites found on or near the coastline have been deliberately
placed to exploit the maritime potential.” Finally, Ali Mazrui and Ibrahim Shariff
(1994: 81) emphasise, “Kimvita, the dialect of the Swahili people spoken in Mombasa,
Kenya, is overwhelmingly associated with Arab ethnicity and Islam.”
According to Grydehøj (2015: 3), it is “relatively easy” to conceive small islands
as territory, and they are “exceptionally strong as places” (Grydehøj’s italics).
It is, therefore, apart from security reasons, quite logical for the Swahili ruling
class, merchants of Arab or mixed African ancestry (Coquery-Vidrovitch, 2005: 131)
to claim the island as its territory. The Swahili conception of civilisation means
urban living (Sheriff, 2014: 29), and within the identifiable spatial containment
of the island, Swahili culture and urbanism could develop and mature.
Grydehøj (2014: 184) mentions, “There are different kinds of island cities, most
obviously: 1) strongly urbanized small islands and archipelagos; as well as: 2) major
population centres located on largely rural islands or archipelagos.” Mombasa is
an example of the former, and, geographically, is particularly clearly defined as
an estuary island with only a narrow creek separating it from the mainland. Lamu
and Zanzibar constitute examples of the latter.
The proximity of Mombasa Island could affect the mainland in ways that were not possible
for the other Swahili Islands. As an offshore island, Zanzibar had to project its
power to the coast where it relied on representatives, maliwali (singular: liwali
in Kiswahili), to maintain control. Lamu, on the other hand, situated on an archipelago
containing many islands and many Swahili towns, had to compete not only for trade,
but also for territorial influence.
Although Coquery-Vidrovitch (2005: 91) insists that “Islamizised cities seemed to
want to turn their backs on their hinterlands,” this relationship is not clear at
all for Mombasa. Ibn Battuta observed (Battuta & King, 2005: 22), “There is no cultivation
of grain among the people of this island; food is brought to them from the Sawãhil
[coast],” De Blij (1968: 29), however, emphasises that gardens outside the town walls
“provided the immediate need for foodstuffs.”
Today, Mombasa County has a population of about one million. The population density
can be mapped as a Y radiating from Mombasa Island at the junction (Figure 3). The
population density along the main road to the interior is also high because of access
to transport. This Y-shaped territory is essentially Mombasa’s Umland. Whereas the
strip along the coast is fertile and semi-humid, the region behind it is semi-arid.
Mombasa Island’s hinterland comprises Nairobi (480 kilometres away), the fertile
highlands, and the area around Lake Victoria, including Uganda and northern Tanzania.
Two bridges and a ferry link the island to the mainland. Its port remains a prominent
feature, as do an international airport and a railway line. These gateways, together
with tourism, financial services, an oil refinery and a cement factory, form the
major economic centres.
5. Defence benefits
Territorial benefits and those related to defence are, as Grydehøj (2015: 4) argues,
“closely related.” Coquery-Vidrovitch (2005: 132) remarks, “The cities that grew
the most [towards the end of the 15th century] were Mombasa and Pate, thanks to […]
a safer environment and the fact that they controlled trade between the Persian Gulf
and the Indian Ocean.” Mombasa Island has been a sanctuary from its earliest days.
Kai Kresse (2007: 45) recounts that a Sheikh Mvita, a Swahili of purported Shirazi
(Persian) descent, moved to Mombasa from the coast further north (after which the
settlement was named for him). Subsequent to his arrival, there was a “continuing
influx from other towns and regions along the coast,” due to instability caused by
marauding Galla (Oromo) and Zimba from present day Somalia.
Strandes (1961: 60) highlights the fact that the town was surrounded by a wall just
prior to the first Portuguese attack. However, Swahili walls were of doubtful defensive
value: They were only about two metres high and no more than a half-metre thick,
and from personal observation by the author, they had no crenellations, arrow slits
or gun ports. Swahili towns were like the gated estates of today. The island geography,
not the wall, was the line of defence.
John Middleton (1992: 7), whose book Port Cities and Intruders is seminal in the
field, asserts, “Mombasa, in particular, was always the great prize for invaders
such as the Portuguese and the Omani Arabs: whoever controlled Mombasa controlled
the coast.” Although the creek separating Mombasa Island from the mainland could
protect it from raiders, it was vulnerable to seaborne artillery attacks. Therefore,
in an effort to entrench their monopolistic but rather tenuous control over the East
Coast, the Portuguese built Fort Jesus.
Fort Jesus is the oldest surviving construction in Old Mombasa (Rhodes, 2014: 121).
While the coast south of Mombasa was controlled from Ilha de Mocambique, Pearson
(1998: 135) notes that Fort Jesus was intended as “a cornerstone of Portuguese policy
on the northern coast.” Even so, although construction started in 1593, Fort Jesus
was completed only in the 1630s (Figure 4). Namdi Elleh (1997: 150-153) describes
Fort Jesus as a “sophisticated bastion.” It was designed by an Italian architect,
João Batista Cairato: Elleh praises its “anthropomorphic qualities, which echo the
humanistic attitude of the Renaissance artists and architects.” It represents a style
of fortification that the Portuguese recreated in Africa and the East.
Figure 4: Fort Jesus, 1593 (photos by author) .
The Omanis took Fort Jesus in 1698 and made some alterations. The British used it
as a prison, arguably the ultimate gesture of political control. Mombasa Island’s
defence benefits remained relevant right into the twentieth century; a British Admiralty
defence scheme map of 1913 shows beacons, submarine cables and potential landing
sites (Figure 5).
Figure 5: Defence scheme map, covering Mombasa Island and surrounding estuaries (http://www.bl.uk/onlinegallery/onlineex/maps/africa).
6. Transport benefits
Many island cities existed as either political or commercial centres (Grydehøj, 2015:
3). Mombasa has always been both. At the pinnacle of Swahili power, just before the
Portuguese conquests, there were about ten independent Swahili city-states, with
some, certainly including Mombasa, actually controlling the adjacent mainland.
Geographical features contributing to Mombasa’s status as “the most powerful city-state
on the coast” before the Portuguese era (Ogot, 1999: 368) were not just the seasonal
patterns of the monsoon winds (Pearson, 1998: 51-4). They also included Mombasa’s
safe harbour (Middleton, 1992: 6) and its central locality on the Swahili Coast as
well as its relative proximity to Lake Victoria Nyanza, Africa’s largest lake. Since
waterborne transport has always been an efficient way of moving goods and people,
economic development favours small islands with good harbours. Mombasa, as an estuary
island, is perhaps the quintessential “easily administrable and defensible [interface]
between sea and mainland supply routes” (Grydehøj, 2015: 4).
Coquery-Vidrovitch (2005: 329) found that colonisers were attracted to existing cities
which offered a good site and a dynamic and diversified urban society. What Europeans
changed “in accordance with their outward-looking interests” were transport and trade
networks. This is entirely evident in the case of Mombasa Island. British infrastructure
development commenced with the building of Kilindini Harbour and the Kenya-Uganda
Railway (Figure 6), which began construction in 1895 (Rhodes, 2014: 121), rejuvenating
Mombasa’s economy (O’Connor, 1983: 200). The British created an integrated East African
urban system of which both Mombasa and Nairobi were part (O’Connor, 1983: 244).
Figure 6: The Mombasa-Uganda railway (author’s drawing).
De Blij wrote in 1968 that “Mombasa’s immediate hinterland makes little contribution
to the traffic of goods reaching the island; it is estimated that less than five
per cent of such traffic arrives by means other than rail” (De Blij, 1968: 49) Today,
a significant portion of the cargo to and from Mombasa is transported by truck to
Nairobi and beyond. Finally, Mazrui (2001: 111) reminds us that Africa is “culturally
intermediate between Europe and Asia.”
7. Mombasa Island city densification
Grydehøj (2014: 185) classifies Mombasa as one of the “densely populated small islands.”
Maps made during the 1600s clearly indicate the settlement of the Portuguese as a
walled compound adjacent to and directly north of the fort, the area known today
as the Old Town (Figure 7). The distinctly separate settlement marked as Cida de
dos Mouros (English: Town of the Moors), about a kilometre to the north of the Portuguese
compound, shows the Swahili settlement at that time, in the area today known as Mzizima
(Aldrick, 1995: 11). After the Portuguese conquest, the Sultan of Malindi, a Portuguese
ally and bitter enemy of the Mombasa rulers, moved into the existing Swahili town,
known then as Mvita (Aldrick, 1995: 12).
Figure 7: Mombasa maps by João Teixeira, 1649 (Axelson, 1998: 43); Antônio de Mariz
Carneiro, 1639 (http://www.mariapereiraweb.net/?area=historia); António Bocarro,
Mzizima is now the site of the Coast General Hospital, and there is no evidence of
the Swahili stone town. However, excavations of the ruins of other Swahili towns
have been informative, since although the Swahili towns were competitors in long-distance
trade, they shared a common culture and building tradition. They were small, neighbourhood-sized,
compact, urban entities consisting of flat-roofed stone houses and mosques, surrounded
by wattle and daub huts. They were true, sharply demarcated enclaves, and stone houses
were rarely built outside the walls in rural settings (Figure 8). Mosques seem to
have been the centre of social events and interaction.
Apart from mosques, the medieval Swahili stone houses were the first tangible manifestation
of Indian Ocean relationships and were continuously constructed along the whole East
African coast for the major part of the 2nd millennium CE. They are now generically
referred to as Lamu or Lamu-type houses, because that locality is where the only
functioning examples are found. However, it is at the sites of ruins that the original
form is most legible. Peter Garlake (1966: 5) insists that the major elements still
visible at the abandoned sites “are so similar to each other as to be valueless in
a typological study.” Swahili towns feel distinctly Arab, with all the spatial characteristics
one expects to find in an Arab-Islamic town: the narrow lanes (from one to 1.5 metres
wide), the blank façades, and the dead-end alleys. The Lamu house was also a symbol
of magic and power and only the patrician families were allowed to own them.
Figure 8: Typical Swahili town footprints (author’s drawing).
Figure 9: Mombasa Island in 2015 (author’s drawing).
From 1698 onwards, the Omanis simply built over the Portuguese precinct between Fort
Jesus and Mvita and, together with the Indian traders, defined the Old Town as it
exists today, a Muslim enclave covering 47 hectares. Although there is a modest conservation
area, UNESCO has not listed it, unlike Lamu and Stone Town. The fact that it is relatively
intact reflects the resilience and robustness of the Omani-era architecture and urban
form. De Blij (1968: 29) writes that “’Old Mombasa’ never came close to filling the
available space on the island” and never exceeded 60 hectares. Interestingly ,he
adds, “Congested as it was, the walled town never sprawled outward along finger-like
arteries.” Coquery-Vidrovitch (2005: 212) writes succinctly, “[During the 1860s]
Mombasa was still huddled on its island and separated from the mainland by a narrow
channel.” However, with the construction of the port, railhead and the Tudor and
Kizinga precincts, the island was rapidly covered (Figure 9). Only then did the town
form cross over the creek and expand onto the mainland.
Although the Arab-Swahili economy stagnated because of the abolition of slavery and
the appropriation of land by the British, some Omani and Indian entrepreneurs profited
from this new speculative, colonial economy. Many of their houses still exist in
what is today the Old Town Conservation Area (Figure 10). De Blij (1968: 29) is adamant:
“Until the late nineteenth century Mombasa had been an Arab town in the true sense
of the word.” Only in the 1920s did the economic centre of Mombasa shift from the
Old Town to what is today the Central Business District. Incongruously, in 1908 the
British considered demolishing the Old Town (Freund, 2007: 76), which, together with
Fort Jesus, are at present the two main tourist attractions on the island, and major
sources of revenue.
Figure 10: Swahili house in foreground left, Omani house behind it, Indian house
The effect of island-based densification can be illustrated by comparing Mombasa
Island to Malindi (Figure 11), with Mombasa’s population density five times that
of Malindi’s. Situated 120 kilometres northeast of Mombasa, Malindi is very much
part of the same historical and cultural landscape and was subject to the same Omani
and British overrule. British colonial town planning, however, imposed not only functional
zoning, but also allocated separate dormitory precincts to tourists and ‘high-income’
households. Such separation requires buffer zones and needs space, which equals sprawl.
Malindi was prominent during Portuguese overrule, at least until the Portuguese relocated
their base to Mombasa. Today, Mombasa is large and important, whereas Malindi is
in decline, except for tourism (Middleton, 1992: 6).
Figure 11: The density of Mombasa Island compared to Malindi, Kisumu and Le Corbusier’s
unbuilt scheme for the left bank of the river Scheldt in Antwerp (author’s drawing).
A comparison with Kisumu, founded in 1903 on the shores of Lake Victoria as the western
terminus of the Mombasa railway line, illuminates the effect of unplanned growth.
Here the population density is considerably higher than that of Mombasa, because
of the existence – and continuous expansion – of the huge, overcrowded slum precincts,
Manyatta and Nyalenda. Since Kisumu cannot expand, being between Lake Victoria to
the west and a mountain range to the north, it is experiencing uncontrolled, informal
urban spread to the east and south-east, expanding into already limited agricultural
The unbuilt scheme for the left bank of the river Scheldt in Antwerp by the eminent
Swiss-French Modernist architect and city planner (1887-1965) Le Corbusier, is also
based on zoning for commerce, culture, recreation and residences. However, Le Corbusier
treated the site as an island and celebrated the water’s edge with a variety of functions,
which certainly promotes choice. He also planned for a range of transport modes to
access the precinct. Further below in this article, a comparison between this plan
and Mombasa Island reveals a number of existing deficiencies that prevail in spite
of, as well as, because of Mombasa being an island.
8. Typological variety
Three Arab and Afro-Arab building types are distinctive to the East Coast (Figure
12). The houses of the original Swahili residents are part of a tradition that emerged
in the medieval period in Lamu, and clearly have Middle Eastern roots. Today, they
are simply referred to as Lamu houses, but they are typologically identical to the
houses of the period all along the Swahili Coast.
Figure 12: Arab and Afro-Arab house types (author’s drawing).
The origin of the contemporary Swahili house, also called a Majengo dwelling, remains
enigmatic. Sheriff (1998: 16) proposes that it is a derivative of the Lamu house
and that it first appeared in Malindi in the latter part of the 19th Century. He
theorises that a central corridor was cut through the oblong rooms while the courtyard
was moved to the rear.
Conceptually, an Omani house in Zanzibar may be described as a roughly square courtyard
building of two or three storeys, in a cuboid, solid form with a flat roof, sometimes
with a crenellated parapet. Towards the end of the 19th Century, they were generally
roofed over with corrugated iron sheeting as protection against the tropical rain
and to reduce direct exposure to the sun. Regularly spaced shuttered windows face
the street and any adjacent open space. An ornately carved door defines the main
entrance. The term ‘Omani house’ is generic, since Indian families owned similar
houses. In fact, Abdul Sheriff (1998: 50), a preeminent authority on the East Coast,
avers that the Omanis brought Indian artisans to the East Coast. It is quite possible
that these artisans, who certainly worked according to patterns, influenced both
layout and the detailing.
Figure 13: Indian shopfront houses (author’s drawing; map from Hart 2007).
During the 19th Century, arising from opportunities offered by Omani hegemony over
the region, substantial numbers of Indian businessmen, administrators, artisans,
craftsmen and clerics settled in some of the more important mercantile towns such
as Zanzibar, Lamu and Mombasa. Most were from Gujarat. Indian shopfront buildings
delineate trading routes – bazaar streets – that introduced linearity into the winding
street patterns (Figure 13). They gave structure and public paths to a pattern of
labyrinthine lanes with dead-end alleys leading into the wards. These routes are
today the ‘main streets’, where the communities shop for their everyday needs. Tourists
(an important source of revenue) tend to restrict their walking to the bazaar streets,
which also serve as references in negotiating these historical neighbourhoods. The
bazaar streets are indisputably the essence of the neighbourhoods’ economic existence
so that, without them, the Old Town might not have become a tourist destination or
even a functioning town in its present form.
The shopfront houses are two to three storeys with flat roofs, a typology derived
from India (Siravo & Pulver, 1986: 53). These shopfront houses are true ‘home-above,
shop below’ types (uppar makan, niche dukan). Many feature four-leaf, bifold doors,
also known as Gujarati doors, which open the shopfront completely for contact with
Figure 14: Views of Mombasa Island today (author’s photos).
The fabric of Central Mombasa, the business district adjacent to the Old Town, along
Kilindini Avenue, consists mainly of three to four-storey buildings right on the
sidewalk, erected in the 1930s during colonial times by the British and by Indian
businesspeople. As Thomas Gensheimer (2001: 22) comments, these buildings continued
the Indian shopfront tradition because of their mix of residences above the ground
level shops. The British lived in Tudor and Kizingo, both Garden City types of neighbourhoods,
described as a “village [layout]” by Anthony O’Connor (1983: 190). Just as Fort Jesus
is a conspicuous reminder of Portuguese presence, so the neoclassical columned and
arcaded institutional buildings, banks and the Law Court are of course prominent
symbols of British colonialism.
Physically, the island resembles a mosaic of cultures and sub-cultures, clearly reflected
in the architectural and urban typologies. Except perhaps for the colonial-era Kizingo
garden suburb, the neighbourhoods and precincts are relatively compact and walkable
and clearly reflect their cultural orientation (Figure 14).
Ironically, the British, in spite of their self-serving intent, imported the one
element that enhances tropical architecture, the veranda. Inextricably associated
with the colonial bungalow, that “tool of Empire,” as Anthony King (1995: 200) describes
it, the veranda quickly became a common pattern, shared by a variety of building
types on both the island and its hinterland. The verandas of Mombasa and its surroundings
are not only pleasant and predictable unifying elements, but also inviting social
threshold spaces as well as comfortable shelter against heat and rain (Figure 15).
Figure 15: Mombasa’s verandas (photos by the author).
The wide range of prevalent building types is an excellent example of a comprehensive
“inherited urban morphology,” to repeat Baron’s (1994: 263) phrase. More than anywhere
else along the coast, Mombasa represents a concentrated and extensive convergence
of African, Swahili, Arab, Indian and European traditions. These seem to cater well
to an equally varied range of needs and social structures (Figure 16). Interestingly,
an ever-present building type in African cities, the makeshift shanty, is quite rare
on Mombasa Island and even its Umland.
Figure 16: Diversity of building types (author’s drawing).
9. Spatial integration
The most tangible manifestation of multi-culturality is the proliferation of places
of prayer for the major religions. In Mombasa, Islam, Christianity and Hinduism have
co-existed for many centuries. The majority of the coastal people are Muslim and,
as one might expect, there are many mosques evenly spread throughout the fabric.
Some, like the Mandhry Mosque in the Old Town, are as old as 500 years. The mosques
are generally architecturally less elaborate than, say, the Anglican Cathedral (built
in 1903) or the Hindu temples. Often, only the presence of a minaret reveals the
function of the former.
Mazrui (2001: 112) believes that “the multiculturalization of Mombasa began quite
early.” Inadvertently the British set the stage for the current nature of multi-culturalism
when the colonial economy upset the demographics: Whereas the inhabitants of Mombasa
Island were previously mainly Muslim, O’Connor (1983: 121) noticed that the port
workers are mostly Luo, a non-Muslim tribe. The colonial enterprises required labour:
Upcountry migrant workers often became Muslims and called themselves Swahilis, believing
that this would confer social respectability. At the same time, the patrician Swahili
families fabricated Persian and Omani genealogies to enhance their status in the
eyes of the colonial administration (Mazrui & Shariff, 1994: 29).
10. Potential for enhancing islandness
The Mombasa County Government recognises that tourist attractions such as Fort Jesus
are an economic asset (MCG, 2013: x-xi): “There is enormous potential in the tourism
and hospitality sector which is yet to be optimally tapped.” A master plan and policies
are needed to “enhance the aesthetic value of Mombasa City as a beautiful tourist
The first issue spoiling urban aesthetics is that some prominent contemporary buildings
in Central Mombasa evidence globalisation in their typical International Style building-as-object
paradigm and disregard for cultural and climatic context; they create pockets of
placelessness. Mutonga Peninah (2015) is adamant that the ‘Islamic’ architecture
of the Old Town responds better to the hot humid climate than does the “modern breed
of architecture […] that barely even respects materialism and the concept of environmental
Grydehøj (2015: 5) advises that, “Island status […] comes with not only spatial benefits
but also spatial constraints in the form of land scarcity, which drives urban densification.”
Land scarcity is being offset by the benefits that “tight clustering” offers in terms
of administrative, residential, industrial, transport and service functions. On Mombasa
Island, that “tight clustering” seems difficult to manage; at this stage, the city’s
planning authorities are under pressure from developers who are eager to build high-rise
residential and office towers, in precincts where there is currently a three-storey
limit (Sanga, 2013). It is clear that businesses consider the island’s central business
district as the prime position. The logical alternative is dense courtyard buildings
in a carpet pattern that can recapture the walkability and ambience of the Old Town,
and achieve the same or higher densities as towers, with the added advantage of providing
positive outdoor spaces that freestanding towers, by definition, cannot do (Rogers
& Power, 2000: 180).
The second issue is that, ironically, Mombasa Island is not experienced as an island.
It offers absolutely no sense of place as an island. Unlike the mainland coast to
the north and south, Mombasa Island has no idyllic palm-lined, white, sandy beaches;
it is a coralline tectonic mass with “steep, often almost vertical cliffs between
7 and 10 m high” (Hoyle, 2002: 187). Brian Hoyle continues:
There is no waterfront causeway, not even a footpath, along most of the maritime
perimeter of the urban Conservation Area; public access to the Old Harbour is available
only at certain points such as Fort Jesus and the Leven Steps. Although certain functional
access points such as the Old Port obviously provide clear exceptions to this general
lack of water-orientated activity, many properties, public and private, appear to
turn their backs to the water.
In order to enhance the island experience it is imperative that a master plan must,
like Le Corbusier’s plan for Antwerp, focus on access to and views of the water,
similar to the “promenades” and “panorama” that Le Corbusier envisaged (1967: 272).
The third problem is access. With only two roads and a ferry linking Mombasa Island
to the mainland, travelling by car is mostly a frustrating and time-consuming experience.
A new bridge linking the island to the south coast is one of a number of planned
development projects, which also include a new container terminal and the southern
bypass (Figure 17). However, it is doubtful if a bridge replacing the ferry will
truly alleviate traffic congestion. Le Corbusier planned an extensive rail and road
“urban shuttle service” network between Antwerp and the new Left Bank City (Le Corbusier,
1967: 273). Similarly, only rapid mass transit will solve Mombasa Island’s traffic
problems. Quoting Cliff Moughtin (2003: 279):
Clearly, an effective public transport system is the basic requirement for the development
of compact, fine-grained sustainable cities of mixed land use, and the foundation
for a network of urbane streets and squares. Anything short of this holistic urban
agenda is superficial, merely treating the symptoms of the ills that beset our cities.
Figure 17: Location map of envisaged Japanese development projects (drawing by the
author after a plan by Katahira & Engineers International, 2015).
11. Lessons learned
South African cities are amongst the “most inefficient and dysfunctional cities in
the world” because of sprawl and segregation (Schoonraad 2000: 220). Since Mombasa
Island represents the typological opposite, it clearly offers lessons in good urbanism.
Grydehøj et al. (2015: 1) maintain that island concepts are useful to “metaphorically
describe developments in urban space.” This requires an “island approach to urban
research,” a convenient and credible theoretical position for the present research.
An island is a clearly defined, bounded place, in contrast to the cities and towns
of sub-Saharan Africa, whose edges generally fray out into the countryside. Whereas
such fluid and blurred boundaries are conducive to fragmentation and urban sprawl,
the geographically bounded space intrinsic to islandness, especially for small islands,
necessitates densification. Urban Growth Boundaries (UGBs) can achieve spatial containment,
emulating islandness (Figure 18). In South Africa, this should allow the gradual
densification and transformation – inside the UGB – of traditional low-density suburbs,
historical townships and the recent RDP (Reconstruction and Development Programme)
schemes, as well as the development of Apartheid-era buffer zones where these cannot
be used as nature reserves for public enjoyment. It might even be advisable to demolish
parts or all of particularly degraded and economically unviable historic townships
that lie beyond the UGB.
Figure 18: Islandness as an urban concept (author’s drawing).
Because of their rigid boundaries and small territories, islands were subject to
intensive processes of cultural encounter, political annexation and settlement, making
them particularly revealing and tragic places to observe the impact of colonialism
and globalisation (Sivasundaram, 2015).
This quote applies in every respect to Mombasa Island. For a tiny place of only 14.1
km², it has had a disproportionately formative impact on the geopolitical and economic
history of the whole East African region.
Mombasa evolved from an early Swahili island settlement, followed by Portuguese,
Omani and British colonial interventions and finally independence into the physical
island city it is today. The architecture and urban forms of Mombasa and its surrounds
clearly reflect all of its historical contacts and impacts. The typological diversity
resulting from the convergence of indigenous, Swahili, Omani, Indian and European
influences is obvious. Some typologies were of course, originally introduced by oppressive
regimes that made no provision for the existing population. They have nevertheless
been pragmatically absorbed into the socio-cultural milieu as part of the inherited
Mombasa was long a defensible site but still remains a trading centre and seat of
economic power. An essential symbiosis with the hinterland that was established by
the Swahilis still exists, as do the long-distance trading networks that were the
raison d’être for the Swahilis’ existence.
Mombasa’s long uninterrupted urban history, its cultural and architectural diversity
and relatively consistent economic prosperity are largely due to the benefits afforded
by islandness, reinforced by the patterns of the monsoon, proximity to India and
the Arabian Peninsula, access to desirable products from the African interior and
its locality on the sea-route to the East. Islandness allows Mombasa to function
as an incubator that allows cultures to evolve – and hybridise – and physical urban
forms to mature.
At various times during the last 1000 years, spanning the urbanisation of Mombasa
Island, the three benefits being studied intermittently dominated (Figure 19). For
all practical purposes, all three have been important, except that the benefit of
defensibility is perhaps irrelevant with the long-distance weaponry of the 21st Century.
Figure 19: The benefits of islandness (author’s drawing).
1. Aldrick, J. (1995) ‘The Old Town of Mombasa’, Journal of the Kenya Museum Society,
2. Allen, J. de V. (1993) Swahili Origins: Swahili Culture and the Shungwaya Phenomenon.
3. Axelson, E. (1998) Vasco da Gama: His Travels through African Waters 1497-1499.
Somerset West: Stephan Phillips.
4. Baldacchino, G. (2008) ‘Studying Islands: On Whose Terms? Some Epistemological
and Methodological Challenges to the Pursuit of Island Studies’, Island Studies Journal,
3 (1), 37-56.
5. Baron, R. (1994) ‘The Inherited City as a Resource’ in T.J. Bartuska & G.L. Young
(eds) The Built Environment: Creative Enquiry into Design and Planning. Menlo Park,
California: Crisp, pp. 253-263.
6. Battuta, I. & N. King (2005) Ibn Battuta in Black Africa. Princeton: Markus Wiener.
7. Breen, C. & P. Lane (2003) ‘Archaeological Approaches to East Africa's Changing
Seascapes’, World Archaeology, 35(3), 469-489.
8. Chaudhuri, K.N. (1985) Trade and Civilisation in the Indian Ocean: An Economic
History from the Rise of Islam to 1750. Cambridge, UK: Cambridge University Press.
9. Cissoko, S. (1986) ‘The African City: A Source of Progress and Modernity’ in A.
Alfonsi (ed.). Le rapport entre ville et campagne en Afrique Occidentale et le transfert
de technologie. Paris: CERFE, pp.1-18.
10. Clark, R. & M. Pause (2005) Precedents in Architecture: Analytical Diagrams,
Formative Ideas, and Partis. New York: Van Nostrand Reinhold.
11. Coquery-Vidrovitch, C. (2005) The History of African Cities South of the Sahara:
From the Origins to Colonization. Princeton: Markus Wiener.
12. De Blij, Harm J. (1968) Mombasa: An African City. Evanston, Illinois: Northwestern
13. Elleh, N. (1997) African Architecture: Evolution and Transformation. New York:
14. Freund, B. (2007) The African City: A History. New York: Cambridge University
15. Garlake, P. (1966) The Early Islamic Architecture of the East African Coast.
London: Oxford University Press.
16. Gensheimer, T. (2001) ‘Cross-Cultural Currents: Swahili Urbanism in the Late
Middle Ages’ in N. AlSayyad (ed.). Hybrid Urbanism: On the Identity Discourse and
the Built Environment. Westport: Praeger, pp. 21-41.
17. Grydehøj, A. (2015) ‘Island City Formation and Urban Island Studies’, Area, 1-7.
18. Grydehøj, A. (2014) ‘Guest Editorial Introduction: Understanding Island Cities’,
Island Studies Journal, 9(2), 183-190.
19. Grydehøj, A., X.B. Pinya, G. Cooke, N. Doratlı, A. Elewa, I. Kelman, J. Pugh,
L. Schick, & R. Swaminathan (2015) ‘Returning from the Horizon: Introducing Urban
Island Studies’, Urban Island Studies, 1, 1-19.
20. Hart, T. (2007) ‘Gazetting and Historic Preservation in Kenya’, CRM: The Journal
of Heritage Stewardship, 4(1). Available at:
http://www.nps.gov/history/crmjournal/Winter2007/article.html. Accessed on 7 July
21. Hay, P. (2006) ‘A Phenomenology of Islands’, Island Studies Journal, 1(1), 19-42.
22. Hoyle, B. (2002) ‘Urban Renewal in East African Port Cities: Mombasa’s Old Town
Waterfront’, GeoJournal, 53: 183–197.
23. Katahira & Engineers International (2015) Study on the Project for Construction
of Mombasa Gate Bridge in the Republic of Kenya. Final report to the Ministry of
Economy, Trade and Industry.
24. King, A. (1995) The Bungalow: The Production of a Global Culture. Oxford: Oxford
25. King, A. (1994) ‘Terminologies and Types: Making Sense of Some Types of Dwellings
and Cities’ in K. Franck & L. Schneekloth (eds). Ordering Space: Types in Architecture
and Design. New York: Van Nostrand Reinhold, pp. 127-144.
26. Le Corbusier (1967) The Radiant City. New York: Orion.
27. Kresse, K. (2007) Philosophising in Mombasa: Knowledge, Islam and Intellectual
Practice on the Swahili Coast. Edinburgh: University Press.
28. Mazrui, A. (2001) ‘Stages of Globalization in the African Context: Mombasa’ in
N. AlSayyad (ed.). Hybrid Urbanism: On the Identity Discourse and the Built Environment.
Westport: Praeger, pp. 111-130.
29. Mazrui, A. & I. Shariff (1994) The Swahili: Idiom and Identity of an African
People. Trenton: Africa World Press.
30. McConkey, R. & T. McErlean (2007) ‘Mombasa Island: A Maritime Perspective’, International
Journal of Historical Archaeology, 11 (2), 99-121.
31. MCG (Mombasa County Government) (2012) First County Integrated Development Plan
2013-2017. Unpublished official document.
32. Middleton, J. (1992) The World of the Swahili: An African Mercantile Civilization.
New Haven: Yale University Press.
33. Mitchell, P. (2005) African Connections: Archaeological Perspectives on Africa
and the Wider World. Oxford: AltaMira.
34. Moughtin, Cliff (2003) Urban Design: Street and Square. 3rd edition. New York:
35. O’Connor, A. (1983) The African City. New York: Africana.
36. Ogot, B.A. (Ed.) (1999) UNESCO General History of Africa: Volume V – Africa from
the Sixteenth to the Eighteenth century (Abridged version). Oxford: Currey.
37. Pearson, M. (1998) Port Cities and Intruders: The Swahili Coast, India and Portugal
in the Early Middle Ages. Baltimore: Johns Hopkins University Press.
38. Peninah, M. (2015) Mombasa: Great City Terrible Place: Future of Island Cities.
Available at http://www.archidatum.com/articles/great-city-terrible-place-future-of-island-cities.
Accessed 2 July 2015.
39. Rhodes, D. (2014) Building Colonialism: Archaeology and Urban Space in East Africa.
London: Bloomsbury Academic.
40. Robinson, J. (1994) ‘The Question of Type’ in K. Franck & L. Schneekloth (Eds.)
Ordering Space: Types in Architecture and Design. New York: Van Nostrand Reinhold,
41. Rockel, S. (2014) ‘Between Pori, Pwani and Kisiwani: Overlapping Labour Cultures
in the Caravans, Ports and Dhows of the Western Indian Ocean’ in A. Sheriff, & E.
Ho (Eds.). The Indian Ocean: Oceanic Connections and the Creation of New Societies.
London: Hurst, pp. 95-122.
42. Rogers, Richard & Power, A. 2000. Cities for a Small Country. London: Faber &
43. Sanga, B. (2013) ‘Why Mombasa Pulled Plug on Highrise Buildings’. Available at:
http://www.standardmedia.co.ke/thecounties/article/2000097827. Accessed 2 July 2015.
44. Schoonraad, Maria (2000) ‘Cultural and Institutional Obstacles to Compact Cities
in South Africa’ in M. Jenks & R. Burgess (Eds.) Compact Cities: Sustainable Urban
Forms for Developing Countries. London: Spon, pp.219-230.
45. Sheriff, A. (2014) ‘Globalisation with a Difference: An Overview’ in A. Sheriff,
& E. Ho (eds). The Indian Ocean: Oceanic Connections and the Creation of New Societies.
London: Hurst, pp. 11-41.
46. Sheriff, A. (1998) Zanzibar Stone Town: An Architectural Exploration. Zanzibar:
47. Sheriff, A. & E. Ho (eds) (2014) The Indian Ocean: Oceanic Connections and the
Creation of New Societies. London: Hurst.
48. Sicking, L. (2014) ‘Islands and Maritime Connections, Networks and Empires, 1200–1700:
Introduction’, The International Journal of Maritime History, 26(3), 489–493.
49. Siravo, F. & A. Pulver (1986) Planning Lamu: Conservation of an East African
Seaport. Nairobi: The National Museums of Kenya.
50. Sivasundaram, S. (2015) ‘Islands and Beaches: The Pacific and Indian Oceans in
the Long Nineteenth Century’. University of Cambridge: Unpublished course material.
51. Strandes, J. (1899/1961) The Portuguese Period in East Africa. Translated by
J. Wallwork. Nairobi: East African Literature Bureau.
52. World Bank (2010) Kenya Economic Update with a Special Focus on the Port of Mombasa.
Nairobi: World Bank.