|UFDC Home||myUFDC Home | Help ||
|Table of Contents|
|Coffee and prosperity in Buganda:...|
|Recent archaeological work in Southern...|
|The role of forts in safeguarding...|
|H. M. Stanley's journey through...|
|The education of African Muslims...|
|The diaries of Emin Pasha -- Extracts...|
|Uganda bibliography 1964-1965|
|Index to Volume 29 (1965)|
ALL VOLUMES CITATION PAGE IMAGE ZOOMABLE
STANDARD VIEW MARC VIEW
|Table of Contents|
Front Cover 1
Front Cover 2
Title Page 1
Title Page 2
Table of Contents
Table of Contents 1
Table of Contents 2
Coffee and prosperity in Buganda: Some aspects of economic and social change
Recent archaeological work in Southern Uganda
The role of forts in safeguarding the Uganda road
H. M. Stanley's journey through Ankole in 1889
The education of African Muslims in Uganda
The diaries of Emin Pasha -- Extracts X
Uganda bibliography 1964-1965
Index to Volume 29 (1965)
Back Cover 1
Back Cover 2
COFFEE AND PROSPERITY IN BUGANDA:
Some Aspects of Economic and
Social Change - W. SENTEZA KAJUBI'. "Z3.!5
RECENT ARCHAEOLOGICAL WORK IN '
SOUTHERN UGANDA - GLEN H. COLE
THE ROLE OF FiRTS IN SAFEGUARDING THE .
UGANDA ROAD - - A.T. MATSON
and J. E. G. SurrroN
H. M. STANLEY'S JOURNEY THROUGH M
ANKOLE IN 1889 NTARE SCHOOL HISTORY SOCIETY .'. Vi
THE EDUCATION OF AFRICAN MUSLIMS IN UGANDA-FELICE CARTER
THE DIARIuA OF EMIN PASHA-ExTRAcrS X: '.
FIRST MEETING OF EMIN WITH STANLEY
25 APRIL-26 MAY 1888 Edited by SIR JOHN GRAY '.W
MAIZE IN UGANDA - - B. W. LANGLANDS ,_l
ENSA ZA KATEBOHA - - H. A. OSMASTON
HILL-TOP IIOLLOWS---A Further Investigation M. PATZ A
The Vegetation of Uganda and its bearing upon land use
(by I. Langdale-Brown, H. A. Osmaston
and J. G. Wilson) - - G. JACKSON :I.:
(by Jomo Kenyatta) -. T. V. SATHYAMURT'HY
The Economics of Subsistence Agriculture
(By C. Clatk and M. R. Haswell) --- M. HALL *
Towns in Africa
(by N. Oram) - - P.H. TEMPLE
The French at Kilwa Island
(by G. S. P. Freeman-Grenville) - M. POSNANSIKY.
Drumming in Akan Communities of Ghana .'
(by J. H. Nketia) - - G. MOORE 23.,
Studies in Speleogy - - M. POSNANSKY 1
SOCIETY NOTES -- - - - --. :-3.
NOTES ON CONTRIBUTORS - - "'
UGANDA BIBLIOGRAPHY 1964-1965
- - Compiled by B. W. LANGLANDS':
FNDEX TO VOLUME 29 (1965) - - --,
THE UGANDA SOCIETY
&.f/0 c- Price Shs. IS/-
The President of Uganda, His Ex(
Dr. MNI. Posnansky
The Hon. Secretary
:.. .... The Hon. Treasurer
The Hon. Editors
The Hon. Librarian
Mr. C. H. M. Barlow
Mr. J. L. Dixon
Mr. R. K. K. Gava
and Mr. S.
: Hon. Librarian.
Messrs. Cooper Bros. & Co.
Corresponding Secretary at Atbale:
H.H. Frederick Mutesa II, K.B.2.,
S" Kabaka of Buganda
R. A. Sir Tito Winyi Gafabusa IV,
c.B.r., Omukama of Bunyoro
Lord Twining of Tanganyika and
Godalming, G.C.M.G., M.B.N.
':. ::': Past Pr
19'3-34 Sir A. R. Cook, C.M.o., o.B.e.
.1934-35 Mr. E. J. Wayland, C.B.E.
s; 1935-36 Dr. H. H. Hunter, C.B.E., LL.D.
1936*37 Dr. H. Jowitt, C.M.G.
'.937-38 SirH. R. Hone, K.C.M.G., K.B.E.,
NI. M.C., Q.C.
..1938-39 Mr. J. Sykes, o.e.E.
1i939-40 Mr. N. V. Brasnett
1940-41 Captain C. R. S. Pitman, C.B.E.,
.t'-1941-42 Mr. S. W. Kulubva, C.B.E.
1:42-43 Mr. E. A. Temple Perkins
1943-44 Mr. R. A. Snoxall
194-45 Dr. K. A. Davies. c.M.G., O.B.E.
N., 1945-46 Mr. G. H. E. Hopkins, o.B.E.
: 1946-47 Mrs. K. M. Trowell, M.B.E.
Mr. S. W. Kulubya, C.B.E. Mr. B.
cellency Sir Edward Mucesa, K.B.B.
Dr. W. B. Banage
Mr. W. S. Kajubi
Mrs. M. Macpherson
Mr. D. K. Marphatia, M.B.E.
Mr. P. Marsh
Mr. A. Mlayanja
Mr. R. J. Mehta, o.B.E.
Mr. B. A. Ogot
Mr. A. H. Russell, M.B.E., D.S.C.
Miss M. Senkatuka
Dr. W. W. Bishop
Mrs. J. Bevin
Dr. J. K. Almond
Mrs. A. E. Luck
Dr. M. Posnansky
Mr. W. A. Trembley (Co-opted)
Mr. A. J. Loveday
Hon. Legal Adviser:
Mr. C. L. Holcom
Mr. R. F. Clarke
Sir John Milner Gray
Mr. E. B. Haddon
Mr. H. B. Thomas, O.B.E.
Professor A. W. Williams, C.B.E.
Mr. E. J. Wayland, c.B.E.
1947-48 Dr. W. J. Eggeling
1948-50 Dr. G. ap Griffith
1950-51 Rev. Fr. F. B. Gaffney, w.F.
1951-52 Professor A. W. Williams, c.B..
1952-53 Sir J. B. Hutchinson, C.M.G., F.R.S.
1953-54 Mr. J. D. Jameson, o.B.E.
1954-55 Dr. Audrey I. Richards, C.B.E.
1955-56 Rev. Dr. H. C. Trowell, o.B.E.
1956-57 Mr. D. K. Marpharia, M.B.e.
1957-58 Mr. M. Barrington Ward
1958-59 Dr. H. F. Morris
1959-60 Professor A. W. Southall
1960-61 Mr. J. C. D. Lawrance
1961-62 Mr. B. E. R. Kirwan, M.B.E.
1962-64 Mr. W. S. Kajubi
K. Mulyanti, O.B.E.
Mrs. J. Bevin
Mr. G. P. Saben
THE JOURNAL OF THE UGANDA SOCIETY
B. W. LANGLANDS
A. W. R. McCRAE
W. A. TREMBLEY
THE UGANDA SOCIETY
Copyright Uganda Society 1965
COFFEE AND PROSPERITY IN BUGANDA:
Some Aspects of Economic and
Social Change - W. SENTEZA KAJUBI
RECENT ARCHAEOLOGICAL WORK IN
SOUTHERN UGANDA - GLEN H. COLE
THE ROLE OF FORTS IN SAFEGUARDING THE
UGANDA ROAD - - A.T. MATSON
and J. E. G. SUTTON
H. M. STANLEY'S JOURNEY THROUGH
ANKOLE IN 1889 NTARE SCHOOL HISTORY SOCIETY
THE EDUCATION OF AFRICAN MUSLIMS IN UGANDA-FELICE CARTER
THE DIARIES OF EMIN PASHA-EXTRACTS X:
FIRST MEETING OF EMIN WITH STANLEY
25 APRIL-26 MAY 1888 Edited by SIR JOHN GRAY
MAIZE IN UGANDA - - B. W. LANGLANDS
ENSA ZA KATEBOHA - - H. A. OSMASTON
HILL-TOP HOLLOWS-A Further Investigation M. PATZ
The Vegetation of Uganda and its bearing upon land use
(by I. Langdale-Brown, H. A. Osmaston
and J. G. Wilson) - - G. JACKSON
(by Jomo Kenyatta) - -
The Economics of Subsistence Agriculture
(By C. Clark and M. R. Haswell) -
Towns in Africa
(by N. Oram) - -
The French at Kilwa Island
(by G. S. P. Freeman-Grenville) -
Drumming in Akan Communities of Ghana
(by J. H. Nketia) - -
Studies in Speleogy - -
SOCIETY NOTES - -
NOTES ON CONTRIBUTORS -
UGANDA BIBLIOGRAPHY 1964-1965
T. V. SATHYAMURTHY
- -M. HALL
- P. H. TEMPLE
- M. POSNANSKY
- G. MOORE
- M. POSNANSKY
- - Compiled by B. W. LANGLANDS
INDEX TO VOLUME 29 (1965) - - -
Uganda Journal 29, 2 (1965) pp. 135-147
COFFEE AND PROSPERITY IN BUGANDA
SOME ASPECTS OF ECONOMIC AND SOCIAL CHANGE
By W. SENTEZA KAJUBI*
The vast majority of the people of Uganda derive their livelihood
directly from the land. Furthermore, among the countries of the Commonwealth
Uganda stands first in the production of green coffee and is second to India in
the production of cotton lint. Plantains, sweet-potatoes, millet, ground-nuts
and a multiplicity of other crops are grown for domestic consumption and
increasingly for circulation in the local markets, and an exportable surplus
of tea and cane sugar is produced under plantation management, mostly by non-
African enterprise. The country still has considerable deposits of copper and
possibly other minerals under the Rwenzori massif. It has apatite, some iron
ore, limestone and a variety of other minerals which, together with a large
hydro-electric power potential, may form the basis for industrial development
in the future. The exchange economic base of Uganda, however, still rests on
two crops, cotton and coffee, which in 1963 accounted for 80.6 per cent of the
value of the domestic exports.
For over fifty years raw cotton was "king", most of it being sold to India,
Japan and the United Kingdom. In 1955, green coffee surpassed cotton as the
principal export item, accounting in that year for 47 per cent and cotton for 38
per cent of the value of the domestic exports. Also in the same year, the United
States of America became Uganda's best customer of green coffee, and indeed
the single most important source of foreign exchange. Despite fluctuating
prices, coffee has maintained first place as Uganda's most valuable export for all
years since 1955 except 1956 and 1961. In 1963, for example, coffee exports
accounted for 53 per cent while lint exports accounted for 28 per cent of the
value of Uganda's total domestic exports earnings, copper and tea providing 7
per cent and 4 per cent respectively. As the main dollar earner and principal
domestic export item, coffee has assumed a pivotal role in the domestic and
external economy of Uganda. There can be no doubt as to the beneficient
effects its success has had on the economy in general and on the living standards
of the people in the main coffee areas in particular. Its rise to first place in the
export list and the concomitant increase in farm incomes, particularly in
Buganda, have helped greatly to widen the exchange base of the economy,
which previously rested almost entirely on cotton.
Export taxes levied on green coffee have in good years contributed as much
as 15 per cent of the total recurrent budget of Uganda. Funds which accrued
to the government through the cotton and coffee assistance schemes during the
buoyant years following the end of the second world war have in the past been
diverted not only to general revenue for the provision of social services, but
also to the capital budget of the government for general development schemes,
and may, thus, have a multiplier effect on the economy as a whole. The pro-
fitability of coffee cultivation and the relatively high income receipts of coffee
136 UGANDA JOURNAL
growers have been instrumental in creating the basis of a substantial market
for imported as well as locally made products. The rapidity with which develop-
ment in this respect has taken place with the development of the cash crop
economy is shown by the fact that the value of imports has risen from 62,168
in 1902-1903, to over 27,900,000 in 1963, and payments to farmers have risen
from 29,000 to over 42,000,000 in the same period.
More significant than the development of trade and the rise of cash incomes,
have been the far-reaching changes which the impact of money and a new
economic order have brought about in the social and economic outlook of the
Baganda. Notwithstanding the old agricultural tradition of the people of Bu-
ganda, the cultivation of crops solely for exchange markets is a foreign in-
novation. The changes which are taking place in the traditional modes of living
and the cultural landscape as a result of the new patterns of occupance present
fascinating problems for those interested in the study of economic development
and cultural change. To the extent that these changes are expressed in the
transformation of the landscape, they become of particular concern to geo-
graphers. A salient feature of this economic transformation is that it is very
unevenly distributed among the different parts of the country. The extent to
which production for subsistence has been superseded by exchange production
varies enormously from district to district. In the fertile lacustrine crescent of
Buganda and Busoga, the early introduction of cotton and coffee into the
economy has evolved a volume of cash incomes far in excess of what is received
in other parts of Uganda. Buganda produces about 92 per cent of the total
robusta coffee exports and about 30 per cent of Uganda's total cotton crop,
although it has only 21 per cent of the land area, and 28 per cent of the total
The cultivation of coffee is a more profitable activity than cotton growing,
and is giving rise in Buganda to a new class of farmers managing relatively
large estates, involving capital investment and the use of hired labour. The
disparity in the tempo of economic growth and in the distribution of land suit-
able for coffee cultivation, has induced migration of population into the favour-
able areas. It appears also that in Buganda cotton is giving way to coffee.
While the amount of robusta coffee produced in Buganda increased by 12 per
cent between 1958 and 1962, cotton production and the acreage under cotton
in Buganda decreased by 10 per cent and 29 per cent respectively during the
same period. In 1962 cotton was grown on 13,500 acres in Buganda as against
20,800 acres in 1958 whereas between these two years coffee acreage increased
from 22,700 acres to 26,300 acres.I A general problem facing the coffee industry
as a whole, and one which is likely to hit Buganda in particular, is the over-
production or under-consumption which has characterized the world coffee
trade in the past and has been aggravated in recent years. This problem is even
more crucial in those countries which produce only a small proportion of the
world's coffee export, but depend upon that commodity for their main eco-
nomic support. Uganda, relying so far on simple techniques of peasant growers,
produces more green coffee than it can readily dispose of in the world markets.
There are thus a number of problems facing the industry in general and the
economy of Buganda in particular. If production techniques are to be improved
and the domestic living standards of the people raised or maintained, it is
necessary to analyse the salient features and problems facing the industry.
The central objective of my paper is to provide a brief survey of the development
TREND OF COTTON AND COFFEE PRODUCTION IN UGANDA
AND BUGANDA 1930-1965
YEAR OF HARVEST
0 I I I
W S Koubi,
I l l 0
138 UGANDA JOURNAL
of coffee growing in Buganda and its relationship to the political, economic
and social development of Buganda.
Coffee was cultivated or gathered in a wild state in Buganda many centuries
before the advent of Arabs or Europeans into this area, and it was significant
in the social life of the Baganda. For example, coffee beans were used in the
ritual of blood-brotherhood (okutta omukago), whereby a coffee berry was
split and each of two men took one half to moisten with his own blood and
gave to the other man to eat, thereby establishing mutual blood bonds between
the two men and their families. It was a sign of respect and hospitality to offer
coffee berries to visitors, and travellers commonly carried some with them to
chew on the way. Nsimbi states in his book: Amannya Amaganda2, that Kintu,
the founder of Buganda, was offered coffee to chew on his arrival in Kyagwe.
Speke states in his diary that on reaching Masaka in 1862, he was offered
among other presents a large bundle of country coffee, which "grows in great
profusion all over this land in large bushy streets, the berries sticking on the
branches like clusters of holly-berries."3 The Sese Island and parts of Kygawe
county are said to have been important centres of early coffee cultivation in
Buganda and a considerable trade in chewing coffee took place between these
areas and the rest of Buganda in the middle of the nineteenth century. At the
same time, coffee from the northern borders of Lake Nalubale (Victoria) was
sold in the northern and central part of what is now Tanganyika. Thus Sir
Richard Burton noted in 1860 that "Karagwah grows according to some,
according to others imports from the northern countries, along the western
margin of the Nyanza Lake, a small wild coffee locally called mwami. Like
all wild productions, it is stunted and undeveloped, and the bean, which-
when perfect is about the size of a corking-pin's head, is never drunk in de-
coction... The people of this country chew it like tobacco, and during visits, a
handful is invariably presented to the guest... In Karagwah a single khete of
beads purchases a Kubabah (from 1 lb. to 2 lbs.) of this coffee; at Kazeh and
Msene, where it is sometimes brought by caravans, it sells at fancy prices."4
Although indigenous to eastern Africa and Arabia coffee spread to south-
east Asia and to South America and was established in these regions as a staple
crop before it was formally cultivated as a cash crop in most parts of Africa.
It may be said that Africa gave the coffee plant to South America and South
America gave the cocoa to Africa. It is only in relatively recent years that
coffee cultivation in the native continent has represented an appreciable pro-
portion of the world's total. The development came naturally as a result of
the political and economic organisation of Africa by countries of Europe,
which drew large sections of Africa for the first time into the web of the world
The newly acquired colonial possessions had produced little by way of
commerce with the exception of tropical products like rubber which were the
results of a collecting economy, and diminishing amounts of ivory and slaves.
In the case of the Uganda Protectorate, government services were sustained by
loans and grants from the imperial treasury. On the other hand, the womenfolk
of these areas were adept at agricultural work, and the advent of the white man
had deprived the men of their traditional activities. The cultivation of cash
crops was not only the most convenient means of absorbing the male labour
force, but also the only way of relieving the British tax-payer of the financial
burden of supporting the protectorate. This would in addition make available
COFFEE IN BUGANDA 139
cheap raw materials and create a market for the manufacturers of the cosmo-
The first crops to be considered for the export market were those which the
local people had grown for their own use. "Coffee", wrote Sir Harry Johnston
in 1900, "yields the most encouraging results under cultivation. Mr. White is of
the opinion (and from the little I know, I agree with him) that the Kingdoms of
Uganda, and the adjoining districts of Busoga and Unyoro and Toro, are
destined to be great coffee producing countries .. but of course to make coffee-
growing possible as a commercial enterprise, the railway must be completed to
the lake and steamers on the lake must gather up the coffee crops for transport
to rail head and thence to Mombasa. Given the railway and the steamers no
other part of tropical Africa could compete with Uganda as a coffee country."5
The first systematic efforts to grow coffee for export in Uganda date from
1900, when the government introduced arabica seeds from Nyasaland where
coffee cultivation was already flourishing. Arabica, because of its earlier crop-
ping, larger bean and higher market value was considered to be the more
valuable species for cultivation than the native robusta, and it was this variety
which was multiplied and issued to European planters and African cultivators
during the initial years. Coffee cultivation proved a great success at first. Plant-
ing was taken up by mission stations and by peasant cultivators in Buganda,
spreading later to Ankole and Bugisu. After 1909 a great expansion occurred in
the acreage, when coffee became an important plantation crop on non-African
estates. Coffee was planted first as a temporary crop among rubber trees, but when
its profitability had been realized, it was planted alone in its own right. It is signi-
ficant to note that during the early years most of the production was in the
hands of non-indigenous producers and that arabica was almost the only
variety which received official attention. At the end of 1922 it was estimated
that there was a total of 47,187 acres under coffee made up as follows :-
17,656 acres of coffee growing alone on European estates, 3,044 acres of coffee
growing alone on Indian estates, 8,894 acres of coffee interplanted with
rubber on European estates, 895 acres of coffee interplanted on Indian estates
and 10,698 acres of native grown coffee. Of the 20,700 acres stated to be
under coffee growing alone 20,245 acres were arabica and only 454 were under
In spite of its popularity and initial success, arabica proved to be unsuited
for cultivation in most parts of Uganda. It was susceptible to disease and pests
and its life was short, particularly in the less elevated parts of the country.
From 1925 onwards more official attention was turned to robusta, which grew
well in the lacustrine fertile crescent of Buganda. The following figures indicate
the development and subsequent decline of arabica on non-African plantations.
Arabica Acreage on Non-African Estates
1910 1915 1920 1925 1930 1935 1950
407 14,386 20,783 17,404 13,449 6,286 3,000
The policy of fostering robusta on the other hand is reflected in the following
table which shows estimated acreage planted under robusta 1925-1960:
140 UGANDA JOURNAL
Estimated Acreage Under Robusta
1925 1930 1950 1955 1960
African .. .. 348 15,818 162,000 291,000 436,000
Non-African.. .. 1,480 5,457 19,000 18,000 23,000
Total.. .. .. 1,828 21,275 181,000 309,000 459,000
The relative decline of non-African plantation agriculture has been attri-
buted to a number of reasons. First the unsuitability of the climate to arabica.
Secondly a leaf fungus (Hemilia vastatrix), which had devastated the coffee
industry of Ceylon about 1870, appeared in Uganda in epidemic form in 1913
and caused a loss of crop that year estimated at about 30 per cent. Fear of results
similar to those which had occurred in Ceylon caused planters to panic and to
turn their attentions to other occupations. Another retarding factor of the
first years was the lack of any means of mechanical transport or draught power
in the country. Transport of the crop to the buying centres was effected by
means of head porterage. Economic factors were also involved. The freight
charges from Kisumu to the coast were high, and could be only offset by good
yields and high prices. When prices were low the returns to the farmers were
too low to cover the cost production. These circumstances discouraged planters
and African cultivators alike. ".. many of their plantations" wrote the Director
of Agriculture in 1922, "are sadly neglected and in many cases cultivation has
entirely stopped." The solution to the transport problem had to await the advent
of the bicycle, the lorry and the arrival of the Kenya and Uganda Railway at
Kampala, all of which have played a very significant part in the collection and
evacuation of produce. Lastly, there was the problem of labour because the
Baganda were not keen on working as unskilled labourers. Furthermore the
Busulu and Envujjo Law passed in 1927 regulating the dues paid by peasants
and giving each Mukopi the right to cultivate such food and cash crops as he
wished, fortified the position of the peasant, and thus reduced further the foot
free labour force in Buganda. The shortage of labour limited the size of plant-
ations to about 200 acres or less, an area considered insufficient to support a
paid manager and at the same time pay sufficient dividends to an absentee
planter. The majority of the plantations which closed down during the 1920's
were absentee-landlord estates, with proportionately high overhead expenses.7
While the government was preoccupied with the promotion of coffee and
other plantation crops such as rubber and cocoa, the Church Missionary
Society through the Uganda Company was busy distributing cotton seed to
peasant growers through their chiefs. The ease with which cotton could be
grown gave it considerable advantages over coffee. It required only simple
skills and a minimum of adjustments in the life of the peasant grower. It ma-
tured in only seven months, as against four years in the case of coffee, and it
could be grown in rotation with subsistence crops. Under these circumstances,
the development of coffee was slow, and it was not until the 1940's that coffee
occupied second place in the domestic export list of Uganda.
We have seen the introduction and establishment of cash crops in the
farming economy of Uganda, and the development of peasant cultivation
against plantation agriculture. Up to about 1940, the Baganda were on the
whole a happy and contended peasant community. E. M. K. Mulira, in his
little book, Troubled Uganda,8 refers to part of this period as "years of un-
precedented progress." The Baganda were proud of the fact that they had
invited (or they thought they had invited) to Uganda, tha Europeans and the
COFFEE IN BUGANDA 141
Christian Missionaries; that their country was a Protectorate, (whatever that
meant), in contrast to Kenya which was a colony, and that they had been
allowed to possess mailo land and to cultivate their plots as they wished. But
in spite of this apparent progress and the praises sung of the Baganda by
early missionaries and travellers, the response of the Baganda to forces of
economic and social change to have been disappointing. Farming continued
to be regarded as a way of life, rather than as a means of making money. The
Baganda continued to pursue subsistence a mode of existence growing cotton
and coffee merely to pay taxes, buy sugar and tea, and a few other necessities of
life. The indivual who wished to advance himself from the lethargic peasant
community, did so by aspiring to become a chief, a clerk, driver or shop-keeper,
rather than by improving his farm. C. C. Wrigley in reviewing this part of the
economic history of Buganda, arrives at the paradoxical conclusion that "one
of the consequences of British rule and economic development in Buganda was
a rustication of the Baganda".9
The introduction of a cash crop economy and the emphasis placed on the
small producer had a disintegrating effect on the fabric of the Kiganda
village community. Contrary to the traditional apportionment of labour,
Baganda men were forced to cultivate the land. Whereas, in the past the tradi-
tional occupations of men, barkcloth-making, blacksmithing, hunting, beer
brewing and drinking, and warfare, were done largely on a communal basis,
the new cash crops were grown on an individual basis. Families settled on
disseminated dwellings on Bibanja, and had little to do with each other. In the
olden days men used to go to chief's home to talk and to hear the chief decide
cases. The chief's home was the centre of intellectual' life in Buganda. Nsimbi
records that in the reign of Sekabaka Mwanga a song was sung emphasising
the importance of court life in Buganda. "Akiika Embuga Amanya Ensonga"
(he who goes to the chief for a chat, learns how to be wise).10 Under the new
economic order, the people could no longer frequent the chiefs nor could the
chief afford to entertain their subjects. Furthermore the chiefs were kept busy
tracking down the peasants who had not paid their taxes. "Akiika Embuga"
is the now title of the Official Gazette of Buganda, thus it is necessary to take
court news to the people, for the people can no longer come to the chief! This
change in the pattern of village life had a moderating effect on the intellectual
climate of Buganda. The new sources of intellectual stimulus such as the schools
and the newspapers brought about a conflict resulting in a reversal of roles.
It was the young rather than their parents who became more knowing. Further-
more, what the children learnt at school, was only remotely connected with the
problems and traditions of their own country. This conflict resulted in the
suspicion of the "educated" person in Buganda. Even today the intellectuals are
not the leaders of political and social thought in Buganda. The Baganda them-
selves were not unaware of this economic stagnation, decline, and lack of stimu-
lus in their intellectual life. In 1937 the late Samwiri Nsubuga wrote a book
and sponsored a series of lectures which lasted for four years under the title
Buganda yekomererayo ("Buganda, the most backward") in which he appealed
to the people of Buganda to revive and to adopt a new way of life in order to
keep up with the changes elsewhere in the country.
The agrarian policy of the government based primarily on the small indivi-
dual peasant holder rather than on the progressive developer was also an
important limiting factor. Development dependent on an agglomeration of
small individual holders could not be expected to move any faster than the rate
142 UGANDA JOURNAL
at which the peasants themselves were willing to adjust and to adapt to new
methods. Development was in many ways tortoise slow. Production increased
not through increased output per unit area or per man-hour but through
expansion of acreage.
This is not to suggest that there has been no progress in Buganda. On the
contrary, the cumulative effect of many years of agricultural extension work, the
stimulating effect of World War II, and the dramatic rise in the prices of cash
crops following the war years had a great impact, and opened up new horizons
for farmers in Buganda. The aftermath of the war was attended by an increase
in political and economic awareness on the part of the peasantry. Several
political movements and farmers' organizations sprang up, such as Abaganda
Abakopi (the Baganda Peasants), the Uganda African Farmers Union (the
precursor of the Federation of Uganda African Farmer) and the Uganda Natio-
nal Congress. The Bataka (people of the land) movement was also resuscitaed
during this period. A number of Luganda newspapers whose general tone was
described by the Commission of Inquirity into the disturbances, 194512 as
"persistent anti-government propaganda aimed at both the Protectorate and
Buganda Governments" were also started. Four "Proclamations of a State of
Emergency" were declared in Buganda between 1945 and 1953. While it is not
intended to reduce all these tensions to a single economic base, it is clear that
there was a striking coincidence between post-war prosperity and post-war
nationalism in Buganda. It is generally accepted that when people are economic-
ally depressed they have little time for politics and their power to organise
themselves is limited. The increase in farm incomes and the concentration of
of wealth in Buganda allowed the purchase of "pirate" taxis, more travel and
greater spread of rumours. Subscriptions to newspapers and the spread of
political ideas stimulated new hopes and new demands.
With this increase in economic prosperity and the spread of new ideas and
nationalistic feelings in Buganda, radical changes in the structure of government
and economic organisation, became inevitable. It will suffice to mention that
the coffee and cotton industries had to be modified, the Lukiiko reconstituted,
and attempts made to accommodate Buganda in Uganda.
Some of the changes taking place in the pattern of farming in Buganda and
the problems these change pose can be illustrated by specific studies of individual
farmers. An examination follows of a typical large coffee estate, a medium sized
farm and a farm with livestock.
An increasing number of farmers particularly in Masaka and south -east
Mengo have began to incorporate livestock in their farming economy, and to
adopt new techniques. Some farmers in Masaka and Kyagwe, which are the
two most important cores of coffee cultivation in Buganda maintain relatively
large acreages ranging from 20 to 150 acres of established coffee grown by
itself. Such farmers employ substantial hired labour and command a high level
of living. Several own large automobiles, have built brick houses with cor-
rugated iron or tiled roofs and maintain separate structures for livestock, for
drying, and for storage of produce. One farmer near Masaka owns two estates,
one at Kyambogo in Kibinge, where he farms 65 acres of coffee and 4 acres of
lusuku. He has erected an estate coffee factory for dry processing, and a pul-
pery for wet processing. He also runs a maize mill at Kyambogo. At Kya-
bayajanja in Mawogola he maintains 38 acres of coffee and a dairy herd of exotic
stock consisting of nineteen adult cattle and six calves. He has cleared and fenced
100 acres of pasture, and wishes to purchase thirty more head of exotic cattle.
COFFEE IN BUGANDA 143
Many farmers have mulched their plots to reduce labour costs on weeding,
and are using artificial fertilizer like sulphate of ammonia to maintain the
fertility of the soil. Everywhere in the fertile crescent of Buganda, bati or cor-
rugated iron roofs are replacing the traditional thatched roofs, and bati houses
are to be seen scattered between the banana groves and coffee shambas on the
hill slopes. A count along the road between Ngogwe and Mukono, showed
that over 90 % of the houses in this area have corrugated iron roofs.
In contrast to the Masaka plantations one may cite the farm of Mr. Yiga,
located in Kyalubu village in the Gombolola of Sabawali, about ten miles west
of Masaka, as representative of the small peasant farm. Mr. Yiga's farm is
slightly larger than the average for Buganda but like many of his fellows in his
village and elsewhere in Buganda he is a tenant farmer, the land belonging to a
Muganda landlord who lives in another village. Mr. Yiga moved on to the
present farm from Mawokota County (western Buganda) because he wanted to
settle down and grow enough food for his own needs. The farm is five acres
large and on it the family grows matoke, (plantains) coffee, beans, cassava
groundnuts, maize, pineapples, sorghum (for brewing), sweet potatoes, toma-
toes and green vegetables. Matoke forms the main subsistence staple, occupies
the largest area of the farm, and most of the productive efforts of the family are
directed to it. The care of the lusuku is the responsibility of the female members
of the family as is that of the other subsistence crops. Intercropping is practised,
but there is no systematic rotation of crops. Robusta coffee is grown under
mituba (barkcloth trees) and occupies 11 acres or about a third of the area of
the farm. It is the main source of income and yields about 6000 lbs of dried
cherry or about 2 tons per acre valued at about 3000/- annually at an average
price of 50 cents per lb. But Mr. Yiga complained that there are wide fluctua-
tions both in yield and the price of coffee. Subsidiary income is received from
sales of Mwenge (banana wine), sorghum and bananas.
Mr. Yiga and his family form the regular labour force on the farm, but
casual hired labour is also used. Weeding and harvesting are the most import-
ant operations, but Mr. Yiga's farm has been mulched with coffee husks,
thus reducing weeding operations to a minimum. The Muluka chief would like
Yiga to grow some cotton as a matter of government policy, but Yiga has no
more land, and he also knows that coffee yielding 2 tons per acre would pay him
almost ten times more than cotton yeilding 400 lbs of seed cotton per acre at an
average price of 50 cents per lb. Here lies the dilemma of the Buganda farmer.
Yet another type of Muganda farmer is represented by Mr. Erissa Kizza.
This farm is located at Buseke Gombolola Mumyuka, Budu, and is typical of the
larger coffee grower who has incorporated livestock in the farm economy. This
farm is also illustrative of the problems which face the progressive landlord
farmer. Mr. Kizza has 7 acres of coffee, 3.6 acres of lusuku, a fish pond and
has fenced 10 acres of paddocks for exotic cattle. But one half of his 60 acres
of mailo land is occupied by eight peasants who pay him only 8/50 each annually.
The farming destiny in this area is high and it is becoming increasingly difficult
for peasants to secure contiguous tracts of land which could be farmed econo-
mically if mechanical means were to be used. The landlord who would like to
farm his land himself, finds that he cannot evict or regroup his tenants without
paying prohibitive compensation. The small farmer on the other hand who
wishes to improve and enlarge his holding finds that the cannot often do so
beyond the customary three or four acres, and that he cannot mortgage his
farm for development capital. "Here is aprima facie argument", writes Mukwa-
YIGA'S FARM BUDDU-MASAKA
FH Farmner's House
PM H Farrne'r's Mother's House
L Latrine s 9 ss a 4a
H Hut for workers A s 7 A n
T Tank k s s A
CDY Coffee .Drying Yard (Concrete) s
d Drying bed for Cups & Plates
R Tree '
G Groun'dnuts (Peanuts)
M Maize (Corn)
Mut Mutuba (Barkcloth Tree)
on Sweet Potatoes
- 4 Bush /
W specially for
beer (Kisubi) N
G1' T f co co co.A
Co Co Co Co CoQ
Mut *Mut Mut
Co Co Co Co SCALE 1:1800
50 0 SO 100 I50
Co Mut Co Co C Co Mut S -0 [so__
- Mut MutAA 1
W- FROM MASAKA -
~I.SentezaKajubi L E.K,~oz, I. K. Mbszira
4.Senteza Kajubi E. Kigaozi
J. K. Mbazira
9, 92 p
SCHEDUL- OF AREAS
Swamps St+S2 3-84 ACRES
Coffee C1 0.65
C,, 007 ,,
C4 1 63
Homestead H 0-56
Bush B 0-77
Fish Pond F P 0-13
Eucalyptus E 0.80
Plantains(Lusuki)E.L. 3-61 E
Outside, A. O.A 0.74
Short Grass S.G I '12
Paddock P, 1,50 -
P, 1.06 TENAWN
P3 1.62 9
Tenant T, 1 19
T, 723 AA
,, T4 46
T5 2.28 A A
T, 0 99
re &,tf9 5-84
Total 60-00 Acres
146 UGANDA JOURNAL
ya, "for amending the agricultural legislation to provide for improved
agricultural methods for the benefit of the land owner and peasants."13
Another problem facing Buganda farmers springs from their great reliance
on hired labour. Many farmers in Buganda, are merely supervisors of hired
labour, and seldom do any farm work themselves. This situation maintains a
real barrier between the extension work of the agricultural department and
farm practice. Furthermore, the cost of agricultural labour has gone up con-
siderably in the last few years, so that an ordinary farmer can no longer afford
to employ a labourer for the whole day. Employers have reduced the labourer's
daily task to such an extent that a day's task is equivalent to about 2 hours
work, and a labourer is, therefore, able to work for three or four different em-
ployers concurrently. A disproportionately high percentage of Baganda farmers'
earnings is being spent on hired labour, particularly for weeding and harvesting
operations on large coffee farms. The increased cost of labour is due to the
enhanced cost of living resulting from the rising price of imported goods and
the raising of minimum wages in urban areas but it may also be attributed to
to the fact that many of the immigrant labourers who formed the mobile agri-
cultural labour force have in recent years taken to settling down as sedentary
farmers in Buganda, particularly in Budu and Bugerere, where they are growing
cotton and coffee themselves instead of working for Baganda farmers. There is
no doubt that increased returns to the growers in future will result from more
efficient production methods, such as will increase the productivity of the
Baganda farmers themselves, and minimise manpower requirements per unit
area, rather than through a reduction of labour wages, which are likely to
continue to rise. The present trend on the part of progressive farmers seems
to be towards the smaller, well maintained farm involving the actual partici-
pation of the farmer himself, rather than towards the larger plantation type of
farm dependent entirely on hired labour. Coffee growing has proved a re-
markable success in Buganda. But while it will continue to pay Baganda farmers
to continue to grow coffee, there are international restrictions and there is no
certainty that Uganda will be able to dispose of all its coffee on the world
market. While the situation with regard to cotton is more optimistic, production
per acre is so low and the returns per man-hour are so limited that it does not
pay the farmer in Buganda to continue to grow cotton. Baganda are pro-
sperous, but only in terms of the general poverty of East Africa. Incomes per
head are lower here than they are in many parts of Africa for which information
is available. If the people of Buganda want a higher standard of living they can
certainly achieve it. This means sacrificing some attitudes and customary
modes of thought which are prejudicial to economic change. Higher levels of
living must be judged not only in terms of amounts of cash crops exported
and the number and the size of automobiles and television sets imported, but
also in terms of the pints of milk, number of eggs, pounds of meat consumed
by their children, and the kind of life led by the ordinary people. Buganda has
the natural resources and can mobilise the know-how of its people to produce
1. The economic development of the Kingdom of Buganda, Kampala, Kingdom of Buganda
1965, p. 53.
2. M. B. Nsimbi, Amannya Amaganda n'Ennono zaago, Uganda Society, 1956.
3. J. H. Speke, Journal of the discovery of the source of the Nile, London 1863, p. 275.
4. Sir Richard Burton, The lake region of central Africa, London, 1860, Vol. 2, pp. 180-181.
COFFEE IN BUGANDA 147
5. Sir Harry Johnston, Preliminary report by Her Highnesses Special Commissioner on the
Protectorate of Uganda, Great Britain, (Africa No. 6,), Cd.256, 1900, p. 42.
6. Acreage and production figures quoted in this essay are drawn from the Annual Reports
of the Department of Agriculture, unless otherwise stated.
7. W. A. Maclaren, The resources of the Empire, London, Benn, p. 219.
8. E. M. K. Mulira, Troubled Uganda, Fabian Society, Colonial Bureau Pamphlet, 1950.
9. C. C. Wrigley, Buganda, an outline economic history Economic History Review, 10,
no. 1, second series, August, 1957, pp. 60-80.
10. M. B. Nsimbi, Village life and customs in Buganda Uganda Journal, 20, no. 1, March,
1956, p. 30.
11. S. Nsubuga, Buganda nga yekomererayo.
12. Report of the commission of enquiry into the disturbances which occurred in Uganda
during January 1945. Entebbe, Government Printer, 1945.
13. A. B. Mukwaya, Land tenure in Buganda, East African Studies, no. 1, Kampala, E.A.I.S.R.
The author wishes to acknowledge the assistance of Mr. Eric Kigozi in surveying Mr.
Kizza's farm and kind permission granted by the Divisional Agricultural Officer, Masaka,
to reproduce the map of Mr. Kizza's farm. The author also wishes to express especial appreci-
ation for the helpful co-operation rendered by Mr. Yiga and Mr. Kizza.
Uganda Journal, 29, 2 (1965) pp. 149-161
RECENT ARCHAEOLOGICAL WORK IN SOUTHERN UGANDA
By GLEN H. COLE
During a two and a half year period in 1962 through 1964, an archaeologic-
al programme was carried out, the object of which was to investigate the nature
of the Uganda Acheulian and Sangoan industries and especially the nature of
the transition betweenthem.1 Field work in Uganda was undertaken, for the
most part, in the vicinity of Nsongezi at the junction of the Orichinga Valley
and the Kagera River on the Tanzania border. This area was chosen because it
was known from the activities of Wayland, a geologist and former director of
the Uganda Geological Survey, that there occurred at Nsongezi a thick series
of lacustrine and fluvio-lacustrine deposits which contained Acheulian and
It had generally been thought by prehistorians in Africa that the Acheulian
should end with the Middle Pleistocene and many deposits were dated on this
assumption. It was further thought that the end of the Middle Pleistocene was
characterized by widespread desiccation during which period there were few
situations in which deposits accumulated. By the time that clearly recognized
Upper Pleistocene or Gamblian deposits were formed, entirely different stone
industries involving different manufacturing techniques were in evidence.
The nature of the transition was not known but the main industry that was
assigned to this intermediate position, although not yet found in clear strati-
graphic context, was the Sangoan. Very few deposits containing artifacts which
can be attributed to this intermediate period are known. One such series of
deposits is that which had been reported at Nsongezi, and, in fact, for many
miles along the Kagera River in Uganda and Tanzania, and in the Orichinga
Valley. The pattern of post Pleistocene erosion at Nsongezi, just where the
Kagera issues from a gorge into the exposed lake flats, is such that a strongly
developed artifact-bearing rubble zone called the M-N Horizon, is easily
accessible, much of the overburden having been stripped away. Thus the
Horizon was exposed in road cuts and in a borrow pit at the mouth of the
Orichinga Valley (Fig. 1). For these reasons Wayland conducted most of his
Van Riet Lowe (1952) had described an industry from Nsongezi which he
regarded as being transitional from Acheulian to Sangoan, that is, a late Acheu-
lain with elements indicative of an emerging Sangoan. The lowermost of the
Nsongezi Series-as these deposits are now known, following Bishop (in press:
see Howell and Clark 1963; p. 491)-are regarded as being of Middle Plei-
stocene age. This is, perhaps, exclusive of a basal boulder bed which yielded, in
one locality a few miles from Nsongezi, a tooth of a Lower Pleistocene elephant,
presumably contemporary with that deposit. The series is topped by sediments
of unquestionable Upper Pleistocene age, which contain abundant artifacts.
Deposition was certainly continuous in the Lake Victoria basin and it was
hoped, would prove to be more or less so in the Nsongezi area, although one
150 UGANDA JOURNAL
pronounced unconformity, the M.-N Horizon, occurs in the sequence. Now it
has long been suspected that the Sangoan had developed from the Acheulian in
the Congo and contiguous areas including this part of Uganda.2 There seemed
to be no reason why occupation should not have occurred at one place or an-
other in the general area throughout the period so, with the series of deposits
from the Middle and through the Upper Pleistocene, the Nsongezi area seemed
an ideal one in which to investigate the transitional industry.
For a variety of reasons, not the least of which has been the extension of the
range of Carbon 14 dating and the obtaining of some dates applicable to this
period, there is good reason to think that the industrial transition from Acheu-
lian to Sangoan, whether or not it might have occurred during a Kanjeran-
Gamblian interval, certainly did not coincide with the end of the Riss glaciation
in Europe which event is often taken to mark the Middle-Upper Pleistocene
boundary. If the dates in the range from 60,000-70,000 years can be trusted,
one must accept as a fact that a late manifestation of the Acheulian at Kalambo
Falls on the Tanzania-Zambia border is contemporary with an early inter-
stadial of the last glaciation in Europe, and that the Acheulian-Sangoan tran-
sitional period could, then, fall well within the Upper Pleistocene.
The terminology applied to the various artifact occurrences in the Nsongezi
area is, regrettably, in a state of considerable confusion. Wayland had named
the strongly developed artifact bearing rubble at Nsongezi, which he had
discovered in 1930, the M Horizon (Wayland, 1934): M for the Mousterian
culture, to which he thought he could recognize an affinity in the Nsongezi
O'Brien, a prehistorian who was working in the Nsongezi area a few years
later, in the mid 1930s, had discovered another rubble zone in the Orichinga
Valley a short distance north of Nsongezi proper. This rubble contained an
entirely different complement of artifacts and was named, by O'Brien, the N
Horizon (O'Brien 1936): "N" simply because that letter aphabetically succeeds
Shortly after that, in the later 1930s, Van Riet Lowe, a South African pre-
historian who had been invited by Wayland to describe the Nsongezi arti-
factual material, came to the conclusion, or accepted Wayland's observation,
that there were two separate artifact bearing horizon. The name N Horizon was
applied to what was supposed to be the uppermost. Unfortunately we do not
know upon what evidence this conclusion was reached. We do know that Van
Riet Lowe had a vast amount of artifactual material, excavated and selected by
Wayland and, doubtless, had a number of pits and trenches open for examin-
ation althoughit seems the excavations which showed any appreciable separation
of two separate rubbles were filled in long before Van Riet Lowe's Uganda work
(Wayland 1937).3 Only three of these can be identified for certain today. One
is the original borrow pit (Fig. 1) where Wayland had discovered the site. A
second is the very large "Paddock" excavation (Wayland 1934, Plate 48),
worked in the 1930s and reopened in 1953. The other is a small excavation re-
ferred to as the N Horizon Test Pit, worked by Van Riet Lowe. Bishop has
recently shown, (Bishop in press) and it was abundantly clear in my own work
there, that Wayland's M Horizon and Van Riet Lowe's N Horizon, at least as
known from the N Horizon Test Pit, apply to the identical artifact bearing
rubble zone. One can only guess at the reasons for this confusion; it happens
ARCHAEOLOGY IN SOUTH UGANDA 151
that there is a considerable variation in elevation locally of the artifact bearing
rubble and consistent with the ideas they then had that the present disposition
of the rubble represents an old shore-line of Lake Victoria (see, e.g. Wayland
1935) it must have seemed that a difference of elevation of 15 feet or more
between rubbles in two excavations not far apart (i.e. the N Horizon Test
Pit and the borrow pit) indicated two successive deposits. Van Riet Lowe does
not actually claim that two separate rubbles were ever found to occur separated
by any significant amount of finer sediment within a single excavation but
presents in his memoir on the prehistory of Uganda (1952), a composite section
in which he indicates how it was thought to occur. There is reason to think that
Wayland had misgivings about the N Horizon for when he returned to the site
in the 1950s, he caused a deep pit to be sunk adjacent to Van Riet Lowe's
N Horizon Test Pit (P 35, Fig. 1) which is the only situation at Nsongezi
where one knows just where the artifact bearing rubble was regarded by Van
Riet Lowe as being the N Horizon. This pit reached a thick boulder gravel,
presumably resting on bedrock, some 30 feet below the N Horizon, but no
intervening rubble deposits were encountered. Further, he put in an extensive
series of trenches tracing the M Horizon as it occurs in the borrow pit to that
exposed in the Paddock excavation. The latter occurs at the same elevation as
the N Horizon Test Pit rubble. Wayland had another shaft sunk in the Paddock.
This reached bedrock at some 48 feet below the M Horizon with another thick
and coarse gravel deposit at the bottom and no intervening rubble zone (Way,
When Wayland first used the term N Horizon, he applied it to an upper,
Acheulian artifact bearing portion of the M Horizon where that feature was
found to be locally separated by thin deposits of fine sediments. At that time
he had seldom noted a separation of more than a foot or two, and often none at
all, as opposed to the 20 or 30 foot one postulated by Van Riet Lowe (Wayland
1937a)4 In more recent years, Wayland has applied the term to a variety of
situations: In one case, in excavations at Solomon's Gully some eight miles
north of Nsongezi, certain rubbles bearing Sangoan artifacts were named NI
and N2 Horizons (Wayland n. d.). These rubbles are in uncertain stratigraphic
context insofar as the M Horizon is concerned. In another, the term has been
applied to an Acheulian artifact occurrence at Nyabusora involving artifacts
of "very advanced types." This "horizon" is separated by an appreciable
thickness of fine sediments (ca. 35 feet according to Spur, 1955, Fig. 3) from a
weakly developed stone line said also to contain Acheulian artifacts and which
was regarded as representing the M Horizon (Wayland 1954).5 In a third case,
a thin bifurcate rubble stringer in the Paddock lying from 8 inches to two feet
four inches above the M. Horizon, in the section cleared by the writer, has been
called the N Horizon. This seems to have been the only place at Nsongezi
proper where Wayland observed his N Horizon and its two parts were called
N1 and N2. A small collection of artifacts was taken from this excavation on the
occasion of cleaning a fresh section. The uppermost or N2 stringer provided
material which is certainly very late, apparently comparable to that which
occurs in the uppermost portion of the Kafunzo Flat deposits, i.e. late Upper
Pleistocene. This is entirely in accord with Van Riet Lowe who regarded the
artifacts found with this rubble stringer as Middle Stone Age.6 This, however, is
in a peculiar situation where some late channel cutting by the Orichinga stream
was involved and where the M-N deposits themselves were reworked.
Van Riet Lowe believed that, due to erosion, his N Horizon had, in places,
152 UGANDA JOURNAL
come to rest upon the M Horizon and employed the term M-N Horizon or
Complex to embrace both of them. The term actually seems to be Wayland's.
In one place (1937b), he mentions "the separateness of N, as distinct from the.
M-N Complex" but this usage differs from that of Van Riet Lowe for whom
the M-N Complex would be Wayland's M-N Complex plus Wayland's (and
Van Riet Lowe's) N Horizon.7
It is unfortunate that O'Brien's book (1939) had never been seriously con-
sidered and is still today almost entirely ignored (see, e.g. Cole 1963) so that
Van Riet Lowe's memoir, which did not appear until 1952 (although written in
1939) has firmly established the usage in the literature. Thus it is that Bishop
has retained the term in his forthcoming memoir on the Pleistocene sequence in
Uganda, but it should be understood that the term must be taken to mean the M
Horizon of Wayland and the N Horizon of Van Riet Lowe, these being one and
the same feature. This is also the way the term will be used in this paper.
The N Horizon of O'Brien seems not to be represented at Nsongezi, but
there is, according to Van Riet Lowe, another artifact zone called the 0 Horizon
near the top of the sequence, ten feet from the surface (Van Riet Lowe 1952).
It is not known just where this was encountered but, presumably, it was not far
from Nsongezi. Neither is it known if this was supposed to be the 0 Horizon
of O'Brien (1936), subsequently regarded as having been named in error (O'Bri-
en 1937a), or that of Solomon (1939). Wayland (personal correspondence) in
more recent years, at least, has come to think that this 0 Horizon must be the
same as O'Brien's N Horizon. This it is not but, at least, it is much more similar
to 0 than it is to the M-N Horizon at Nsongezi.
THE M-N HORIZON:
The M-N Horizon consists of rubble with artifacts, sometimes with thin
intercalated lenses of finer sediments. The rubble in some places has been
caught up in a stream channel to be converted to a gravel and also includes
well rounded material derived along with the rubble and some artifacts from pre
M-N riverine, or possibly, lacustrine beach deposits. The artifacts fall into two
main groups. One is more or less heavily rolled. In this group, many of the
smaller artifacts, and the larger ones as well which are relatively thin and flat,
(i.e. those with a large surface area in relation to their mass,) have been sorted
out by agents of transport active in forming or reworking the M-N Horizon,
leaving mainly the large, chunky artifacts behind. Unquestionably, artifacts of
very different ages are included in this heavily worn group. Typologically,
some are of the kind that are found in Beds I and II at Olduvai Gorge. Some of
these are pre-Acheulian at Olduvai and there is no good reason why some
at Nsongezi could not be so as well. Most of this group is probably Acheulian,
although older than that of the main Acheulian occupation. Doubtless some
of the later specimens have also suffered considerable wear, perhaps having
been transported from up the Orichinga to be deposited amongst their less worn
contemporaries. Needless to say, little information is to be gained from a group
of artifacts of this nature. At most, one can say that these artifacts provide
good reason to suspect that tool makers have lived in southern Uganda well
back into the Middle Pleistocene, if not earlier.
The bulk of artifactual material to be taken from the M-N Horizon at
Nsongezi is much less heavily worn. In many cases, it would seem that little or
no transport has been involved but some weathering and abrasion, caused by
smaller fragments of rock in transit, has resulted in a slight blunting of sharp
0 0Qo 200 300
Figure 1 Map of the Nsongenzi pre historic site
N SONGE ZI
1 |-N Nor.tW
Figure 2 Wayland's trench 16 at Nsongezi as deepened and extended and pit C13.
(Elevations are in feet above the Kagera River)
ARCHAEOLOGY IN SOUTH UGANDA 155
edges. This slightly worn group includes some of the smaller flakes and tools
but not the very small trimming debris and smallest of the flakes and tools.
(Although unworn Acheulian material including the small artifact component
seems not to occur at Nsongezi proper, such material has been recovered from
excavations a couple of miles north of Nsongezi in the Orichinga Valley).
This material seems to represent a fairly late Acheulian industry and is chara.
cteristically made on a dark blue-grey to nearly black quartzite. This as opposed
to a much higher proportion of light grey quartzite and quartz in the heavily
Also included in this slightly worn group are artifacts which are either
Sangoan or which are reminiscent of Sangoan types. It is this element that is the
cause of much disagreement in the interpretation of the Nsongezi association of
artifacts. It seems to have been these that led Van Riet Lowe to regard that
portion of the Nsongezi artifact association which he assigned to an N Horizon
as being representative of an industry transitional from Acheulian to Sangoan.
A third, minor component of the M-N artifact association is a recent
addition. This is fresh, unworn material, mostly quartz. This dates to the late
Upper and post Pleistocene. In one instance, pottery has been found resting on
the top of the M-N Horizon beneath more than five feet of overburden. This
material, due to the pattern of late Upper Pleistocene and post Pleistocene
erosion has come to rest on the M-N Horizon in favoured situations. Many of
these favoured situations are those where it is easiest to conduct excavation,
much less overburden being present. This unworn element seems to have been
overlooked in earlier work. Van Riet Lowe, in his N Horizon Test Pit, where
this material certainly must have occurred, concerned himself only with the
larger items-pieces that would not pass through a two inch ring. Most of this
late material is quite small.
THE UPPER RUBBLE AND O'BRIEN'S N HORIZON:
Another rubble zone, apparently of rather limited distribution, occurs
not very far above the M-N Horizon, from one to six feet in the several ex-
cavations where it has been encountered. This feature, which for convenience
has been called the Upper Rubble, seems not to have been exposed in any of
the previous excavations at Nsongezi.8 Acheulian material seems not to occur
in this rubble whereas a few Sangoan implements have been found. A finer
rubble and other sediment lying between it and the M-N Horizon contains a
high proportion of phyllite in sharp contrast to the usual sediments found
overlying the M-N Horizon and also has a few Sangoan type artifacts.
The Upper Rubble was first encountered on extending and deepening a
shallow trench of Wayland's (Tr. 16). In this trench the Upper Rubble can be
seen to come to rest on the underlying M-N Horizon (Van Riet Lowe's N
Horizon) (Fig. 2).
An excavation (Cl) undertaken adjacent to the N Horizon Test Pit was
planned to permit recovery of material of the sort that was regarded as transi-
tional between Acheulian and Sangoan. It was found that what Sangoan ma-
terial did occur was confined to the upper portion of the rubble mixed with
Acheulian while the only diagnostic artifacts in the lower portion were Acheulian
There is, then, ample reason to suppose that Van Riet Lowe's transitional
industry, as it was found in the N. Horizon Test Pit, is, in fact, no more than a
mixture of a rather late (and somewhat peculiar) Acheulian and a Sangoan plus
Middle Stone Age material.
156 UGANDA JOURNAL
It is difficult to state just what this Nsongezi Acheulian consists of because
material from these higher lying rubbles as well as isolated artifacts have so
often settled onto and mixed in with the M-N Horizon. Also, the M-N Horizon
prior to later enrichment has often had the smaller artifact component removed
by natural sorting agencies. After the preliminary excavation programme when
the higher lying rubble was discovered, it was possible to conduct several ex-
cavations in situations where unenriched Acheulian material might be ex-
pected. Analysis of the data gathered from examination of the artifacts taken
from these excavations has not yet been completed but the impressions gained
from preliminary study is that Sangoan material does not occur in those portions
of the M-N Horizon which have experienced no later enrichment. Conversely
the Upper Rubble at Nsongezi contains no definite Acheulian, but has a
sparse scatter of diagnostic Sangoan artifacts along with a good deal of non-
descript artifactual material and some Middle Stone Age artifacts which have
worked their way onto the rubble through the process of soil creep and bio-
logical activity, or perhaps, date to the period of erosion which caused the lag
deposit to accumulate.
O'Brien's N Horizon seems not to be the same as the Upper Rubble at
Nsongezi. O'Brien encountered this horizon in several places in the Orichinga
Valley. O'Brien's N Horizon is reported to be separated from the M-N Horizon
by 15 feet of cross bedded river sands (O'Brien 1939). These sands contain a
number of stringers of artifacts in thin rubbles. These artifacts are almost
entirely cores and flakes and were called by O'Brien "Levalloisian" which, in
this case means little more than that he regarded this as a flake industry. The
sediments appear to have been quickly accumulated and no apparent differences
in the artifacts found in the different layers seem to exist. The presence of these
artifacts poses a problem which will not be dealt with here. For present purposes
it is only pointed out that deposits and artifacts which lie between the M-N
Horizon and the Upper Rubble at Nsongezi, and below O'Brien's N Horizon
in the Orichinga Valley, are quire different.9 Further, the Upper Rubble at
Nsongezi is essentially a lag-concentrate and the Sangoan, in part at least,
dates to the period of deposition of the underlying deposits. O'Brien's N Hori-
zon, on the other hand, seems to terminate a phase of depositional activity.
This was certainly the case in the writer's excavations although O'Brien (1937a)
holds that pre-N Horizon erosion of beds overlying the M-N Horizon occurred
in places and that his N Horizon material sometimes rests directly on the M-N
Horizon. Whatever the case, it seems unlikely that very much of the rubble or
any but an occasional artifact can have accumulated as a lag-deposit. The
artifacts are largely unworn and an abundance of small waste flakes are associ-
ated. The Upper Rubble Sangoan artifacts are somewhat worn, apparently
having been derived from deposits since removed by the erosion which formed
the Upper Rubble surface.
Typological considerations might suggest that if the Sangoan component of
the Upper Rubble at Nsongezi and that of O'Brien's N Horizon is, in fact,
of different age, that at Nsongozi is the earlier. At least, the few Sangoan pieces
from the Upper Rubble and underlying deposits seem more similar to Sangoan
implements from the Sango Hills than to those of O'Brien's N Horizon. The
inadequacy of the Upper Rubble sample and the uncertainty of placement of
the Sango Hills in the stratigraphic sequence, however, combine to render this
impression of little value. Whether the Sangoan of Brien's N Horizon proves to
be earlier or later than that from the Upper Rubble, it seems unlikely that
ARCHAEOLOGY IN SOUTH UGANDA 157
there is any very great difference in age between the two. On the basis of a
comparison of my own material taken from O'Brien's N Horizon and of
material from Kalambo Falls where Carbon 14 determinations have been pos-
sible, a date of between 40,000 and 45,000 years ago can be suggested for the
Sangoan of this horizon.
The reason that so few of these deposits are found at Nsongezi probably
lies in the proximity of this site to the Kagera River. Whatever the Orichinga
stream was doing during N Horizon times, there would have been ample op-
portunity for the Kagera to erode away these and earlier deposits right at the
mouth of the gorge where the Nsongezi site occurs. Further, there has been a
recent period of erosion; that which began with renewal of the tilting which
ultimately drained the lake from the area. This erosion, actively continuing
today, is responsible for the abandoned channel immediately beneath the site
(the swampy area in Fig. 1) and, in fact, the entire Lower Kagera Valley, i.e.
that which is cut into lacustrine sediments.
Whatever the stratigraphic situation of the Upper Rubble at Nsongezi, the
first strongly developed implement bearing horizon in post M-N times is O'Bri-
en's N Horizon, which has a well developed Sangoan industry. If it is true,
as has been suggested, that this Sangoan is later than that known from the
type area of the Sango Hills, this need not mean any more than that the environ-
ment of the immediate Nsongezi area at the time of the main Sangoan occupa-
tion of such areas as the hills around Sango Bay or the Gayaza area some 17 or
18 miles north of Nsongezi was such that no artifacts were left there (e.g. it
could have under been water). It should be mentioned that Sangoan tools are
sometimes found in the old hillside talus or rubbles behind the site and Way-
land, who did a lot of trenching of these deposits found them at Bumona, a
site of extensive excavations about a mile north of Nsongezi. Unfortunately
none of the artifacts have even been described. The few that are retained in the
Uganda Geological Survey collections are closely similar to Sangoan material
from the type area.
Sangoan artifacts characteristically turn up in superficial surface deposits.
They can often be recovered from borrow-pit workings in appropriate situations
and are often turned up by farmers who extract the undesirable rocks from their
garden plots. It was from piles of rocks thus accumulated that Wayland made
his first collections in the Sango Hills. The problem of trying to tie artifact
occurrences of this nature in with a stratified sequence is no simple one. Un-
fortunately then, one cannot know just where this Sangoan fits into the Nsongezi
area sequence. It almost certainly is later than the Acheulian of the M-N Hori-
zon. Typologically it seems less refined than that which occurs in O'Brien's
N Horizon and as it is unlikely that the Sango Hills would have been occupied
at the time that sediments were being put down over the M-N Horizon in the
Nsongezi area (the Sango Hills would have been islands well offshore under
such circumstances; see Bishop, in press, Plate VI), it can be suggested that the
type Sangoan belongs to the post-MN Horizon, pre-O'Brien N Horizon
period. But an industry which might be regarded as transitional between the
Sangoan and Acheulian seems not to be represented in the area. Bishop and
Posnansky have conducted excavations at Nyabusora, some 31 miles down-
stream from Nsongezi and have material there, found stratified above a rubble
which may equate with the M-N Horizon. This, material as yet not described,
may prove to represent such an industry. Certainly it contains elements more
reminiscent of a Late or Final Acheulian than anything that occurs with the
158 UGANDA JOURNAL
M-N Horizon at Nsongezi, and Posnansky (quoted by Bishop, 1963) has noted
Sangoan forms to be included.
In comparison with assemblages from other East African Acheulian sites,
the Nsongezi Acheulian is decidedly peculiar. A large majority of the imple-
ments are crudely worked in contrast to the common occurence of carefully to
elaborately trimmed tools from sites in Kenya and Tanzania. This can,
perhaps, be explained in part by the abundance of raw material.
It may, for instance, have been easier for a tool maker to discard a quickly and
crudely fashioned implement once its working edge was blunted and make a
new one than it would have been to resharpen a blunted tool, as Posnansky has
suggested (Bishop and Posnansky 1960). This is not to suggest that such
carefully shaped tools as many of the hand axes and cleavers, which are common
in certain of the Kenya and Tanzania Acheulian assemblages, attained their
well-finished character because of re-sharpening. On the other hand, the usual
minimally worked large flake tool of the Nsongezi Acheulian unquestionably
would take on a more finished look comparable to many other East African
artifacts if re-trimming for purposes of sharpening or re-shaping a working
edge or point had been resorted to. Hand axes and cleavers, which are major
assemblage components at such sites as Kariandusi, Olorgesailie, Isimila
(Kleindienst 1961) and Kalambo Falls (Clark 1964), are relatively scarce,
comprising less than 5 % of the total nonwaste portion of the artifact association.
It happens, however, that artifactual material excavated at Nsongezi displays
a wide range of wear. Acheulian type tools are rarely unworn and not often
heavily worn. Within the slightly worn grouping which contains most of the
Acheulian implements, then, the percentage of hand axes and cleavers is higher.
Such heavy duty implements as core scrapers and choppers are more abundant
and other implement types occurring as minor components in other East
African Acheulian assemblages are relatively abundant.10 Analysis of material
recovered from the M-N Horizon in situations where little or no post-Acheulian
enrichment should have occurred is in progress. It would seem, from the incom-
plete analysis, that the proportions of the implements which are characteristic of
the later part of the African Acheulian are, in fact, somewhat higher under
these circumstances but still are appreciably lower than they are in the well
known sites of Tanzania and Kenya. Even in the situations where post-Acheulian
enrichment has or could have occurred, only an occasional Sangoan-type artifact
is found. The characteristic material such as core scrapers and choppers are all
types which are known in East African Acheulian assemblages. They are
present in the Nsongezi Acheulian in markedly different proportions than
those known from other African Acheulian sites which have yielded large
collections of unselected material. Further, they are trimmed in the manner
usual with Acheulian rather than Sangoan material; e.g. step-flaking is not
characteristic. In situations where enrichment from post-Acheulian industries
can be ruled out, the Nsongezi M-N Horizon Acheulian does not appear to
contain elements which might be regarded as anticipating the overlying Sangoan
of the N Horizon of O'Brien nor that from the type area.
Until more is known about the Nyabusora material and pending further
discoveries of artifact bearing deposits pertaining to the relevant period in the
Nsongezi area, perhaps the best that can be done is to suggest that the Sangoan
developed in an area excluding southwestern Uganda. Portions of the Congo,
especially in Katanga and Kasai would seem possible. This is suggested by the
fact that the most elaborate manifestation of later phases of Sangoan, or Lu-
ARCHAEOLOGY IN SOUTH UGANDA 159
pemban, development, is found in this area. Curiously, however, a Congo-like
Sangoan or Lupemban industry seems to have flourished along the Yala River
and elsewhere around the Karivondo Gulf in western Kenya (Leakey and Owen,
1945). In the Congo the Sangoan may very well have developed out of a Late
Acheulian; such is O'Brien's opinion. Unfortunately, no very good Congo
Acheulian assemblages have been reported. Uganda, of course, belongs with
the eastern Congo geographically rather than the Kenya-Tanzania portion of
East Africa, so the Nsongezi Acheulian may be representative of the kind of
Acheulian which existed in the Congo. However, the evidence now available
from Nsongezi suggests that people making Sangoan tools moved into the area
rather than that the industry developed there from a Nsongezi-type Acheulian.
The fact that no very late (certainly not typologically Late or Final) Acheulian
seems to occur, obscures the question. It should be noted, however, that it is
exceedingly difficult to visualize an industry such as that from the Sangoan type
site developing from Late or Final Acheulian industries of the sort that occur
at Kalambo Falls or Isimila, characterized by rather small, delicately worked
hand axes, cleavers, and knives worked on flakes. Doubtless, one day there will
be found an industry transitional from some sort of an Acheulian to some
kind to Sangoan but in the situation as known at present in southern Uganda,
there seems to be an appreciable gap.
1. The work discussed in this paper was made possible by fellowship awards from the
National Science Foundation and the Wenner-Gren Foundation for Anthropological
Research. The work was done through affiliation with the Uganda Museum which
provided office space, library facilities and use of field equipment. The Uganda Geological
Survey provided camping, mapping and levelling equipment, and, through their Mbarara
field office, provided storage space and transportation to Kampala of the large quantities
of artifacts recovered. Makerere University College made available its library facilities
and certain laboratory equipment. The writer is most grateful to all of these institutions.
Of the many individuals who rendered assistance in one way or another, thanks are
particularly due to Drs. Posnansky and Bishop who helped in many ways, both in the
field and in Kampala.
2. Leakey (1960) has suggested rather that the Sangoan developed in South Africa.
3. Wayland used the term in debate with O'Brien (Wayland 1937a, b) but applied O'Brien's
term (1936) to a different feature than that named by O'Brien.
4. The sites mentioned by Wayland (1937a) where the separation was observed seem not to
have been at Nsongezi. One excavation about a mile north had two separate rubble zones
about three feet apart at about the appropriate depth which could have been regarded as
M and N by Wayland.
5. Posnansky, who kindly conducted the writer to this site, and Bishop have carried out
excavations there (Posnansky 1962, Bishop 1963).
6. The writer is obliged to O'Brien who kindly provided a copy of Van Riet Lowe's letter to
him including the section cited in the Man correspondence (1937b). At the time Van
Riet Lowe was at Nsongezi, only a single higher rubble was known. It was when Wayland
expanded the excavation in the 1953-54 excavations that it was found that the "N Hori-
zon" divided into two parts (Wayland n.d.). It should be noted that the term Middle
Stone Age as used by Van Riet Lowe in the 1930s involved quite a different concept than
does that used in recent years by some authors following a resolution of the 3rd Pan
African Congress (Clark and Cole, 1957: xxxiii, resolution 6). MSA as used by Van
Riet Lowe could not be confused with "Tumbian" or "Lupemban". In any event, this is
considerably at variance with the viewpoint which saw the N Horizon as containing
Acheulian 2, 3 and 4 artifacts (see Leakey, 1938, and Wayland's appended note).
7. More recently Wayland has used the term in a third way. The upper artifact occurrence
or "Horizon" at Nyabusora was "conclusively recognized as N, or the M-N Complex..."
160 UGANDA JOURNAL
8. There is no record that Wayland ever sank a pit or trench in a situation where the Upper
Rubble is known to occur (other than Tr. 16 which was not dug deep enough). No surface
indication of any other excavation was seen and Wayland's maps include none. It seems
unlikely that Wayland knew of this particular rubble for otherwise he would probably
have considered it to be his N Horizon and would not have had to cite the unconvincing
rubble stringer in the Paddock as its local manifeststion.
9. The sediments lying between the M-N Horizon and the Upper Rubble at Nsongezi are
rather coarse and poorly sorted. Interfaces dip away from the ridge relatively steeply
suggestive of sub-aereal deposition. These deposits may be remnants of a piedmont al-
luvial sheet involving reworked fluvio-lacustrine sediments, perhaps deposited by sheet-
wash. The presence of large amounts of Phyllite, sometimes more than 40% of the rubble
component, is significant. This is a soft, foliate rock which stands transport very poorly.
It is practically non-existent in those portions of the M-N Horizon that have been
caught up as stream channel gravels. Elsewhere in the stream deposits, including the
Orichinga channel sands, it tends to occur as about one percent of the rubble component.
The phyllite is derived from the ridge immediately behind the site (see Howell and Clark
1963: Fig. 12. This figure indicates that the ridge is entirely of quartzite. The writer,
however, found phyllite outcropping on its lower slopes). Artifacts are relatively scarce
in these sediments and are, for the most part, small flakes and chips falling in the same
size range as the phyllite fragments. These apparently could pertain either to Acheulian or
Sangoan. Two large Sangoan implements were found in a fine rubble immediately
underlying the Upper Rubble.
10. It should be stressed that these impressions are based on a consideration of the artifact
association as a whole which, at Nsongezi proper, at least, contains material representing
more than one assemblage.
The author acknowledges the kind permission of Dr. W. W. Bishop to base the map
(Fig. 1) upon that prepared by Dr. Bishop for the Uganda Geological Survey.
BISHOP, W. W.
1963 The later Tertiary and Pleistocene in eastern equatorial Africa. Viking Fund
Publications in Anthropology 36, pp. 246-275.
In press (Monograph on the Uganda Pleistocene to appear as a Memoir of the Uganda
BISHOP, W. W. and M. POSNANSKY
1960 Pleistocene environments and early man in Uganda. Uganda Journal 24,
CLARK, J. D.
1964 The influence of environment in inducing culture change at the Kalambo Falls
prehistoric site. South African Archaeological Bulletin. 19, No. 76, pp. 93-101
CLARK, J. D. and S. COLE (eds.)
1957 Proceedings of the Third Pan-African Congress on Prehistory, Livingstone, 1955.
1963 The Prehistory of East Africa, Macmillan, New York.
HOWELL, F. C. and J. D. CLARK
1963 Acheulian hunter-gatherers of sub-Saharan Africa. Viking Fund Publications in
Anthropology, 36, pp. 458-533.
KLEINDIENST, M. R.
1961 Vatriability within the late Acheulian assemblage in eastern Africa. South
African Archaeological Bulletin, 16, no. 62, pp. 35-52.
LEAKEY, L. S. B.
1938 The Stone Age cultures of Uganda. Man, no. 60, p. 63.
1953 Adam's Ancestors, 4th ed. London.
LEAKEY, L. S. B. and W. E. OWEN
1945 A contribution to the study of the Tumbian culture in East Africa. Coryndon
Museum Occasional Papers No. 1.
O'BRIEN, T. P.
1936 Notes on the Stone Age cultures of Uganda. Man, no. 53, pp. 41-44.
1937a Stone Age cultures of Uganda. Man, no. 129, pp. 103-104.
1937b The Stone-Age cultures of Uganda. Man, no. 237, p. 200.
1939 The Prehistory of Uganda Protectorate. Cambridge.
ARCHAEOLOGY IN SOUTH UGANDA 161
1962 Recent Palaeolithic discoveries in Uganda. Acte du IVe Congres Panafricaine
de Prehistoire et de L'etude du Quaternaire, Vol. II, pp. 207-215.
SOLOMON, J. D.
1939 The Pleistocene succession in Uganda. Chapter III in O'Brien 1939.
SPURR, A. M. M.
1955 The Pleistocene deposits of part of the Kagera Valley, Bukoba District.
Geological Survey of Tanganyika Report No. AMMS/27.
VAN RIET LOWE, C.
1952 The Pleistocene geology and prehistory of Uganda. Part II: Prehistory. Geological
Survey of Uganda Memoir No. VI.
WAY, H. J. R.
1937 Annual Report of the Geological Survey Department for the year ending 31st
December, 1936. 45, pp. 14-15.
WAYLAND, E. J.
1934 Rifts, rivers, rains and early man in Uganda Journal of the Royal Anthropolo-
gical Institute. 64, pp. 333-352.
1935 The M Horizon-A result of a climatic oscillation in the second pluvial
period. Geological Survey of Uganda Bulletin No. 2, pp. 67-76.
1937a The Stone Age cultures of Uganda. Man, no. 67, pp. 55-56.
1937b The Stone-Age cultures of Uganda. Man, no. 184, pp.149-150.
1954 A short report on a geo-archaeological investigation in Nyabusora, Tanganyika
Geological Survey of Tanganyika Report EJW/2.
n. d. Manuscript on file at the Uganda Geological Survey.
Uganda Journal, 29, 2 pp. (1965) 163-184
THE ROLE OF FORTS IN SAFEGUARDING THE UGANDA ROAD
By A. T. MATSON and J. E. G. SUTTON
Only the Banyoro in the west and the Nandi in the east of the tribes living in
the Uganda Protectorate of 1896 made any determined and sustained resist-
ance to the extension of British rule. The need to undertake the pacification of
these tribes had an important effect on the course of Buganda history. Re-
sources available to district officials in areas far distant from Protectorate
headquarters were too limited to allow anything but small, localised measures
to be undertaken against either the Banyoro or Nandi. For any major punitive
operations, reinforcements of officials, troops, auxiliaries and porters had to be
drafted from stations all over the Protectorate. In the case of the Nandi, there
was no question that their opposition to Brtish rule would have a direct influ-
ence on the internal politics of Uganda. For 12 years however they succeeded
in diverting men, money and materials from the task of developing those areas of
Uganda that had accepted British protection. From 1895 until 1906, when the
truculent south-eastern sections were moved to a reserve away from the railway,
five major and a number of minor punitive expeditions were needed to bring
the Nandi under control, and to maintain the security of the Uganda Road and,
from 1899 onwards, the telegraph line to Kampala and the Uganda Railway.
The maintenance of this tenuously held route from Mombasa to Kampala
was a major item in the policy of the Imperial British East Africa Company
and the protectorates which succeeded it. A natural corollary to this objective
was to provide a sufficiently secure route through British territory to encourage
the traders operating on the well-established route through German East
Africa to take up food and trade goods to the Uganda capital and road stations.
To begin with the Company and the government were averse from under-
taking the administrative control of the country through which the Uganda
Road passed. This policy was forcibly expressed by Johnston in October 1900,
when he despatched Jackson to the Eastern Province with plenipotentiary
powers: "This policy is consistent with the cardinal object of your work in
Eastern Province, which is to render the Nyando Valley rail and telegraph
route safer for transport than any other part of the Protectorate. The object is
so all important that no sacrifice is too great for its due accomplishment; and if
you consider Evatt has not got sufficient forces on the spot, you must apprise me
so that reinforcements may be raised elsewhere in the Protectorate."'
One of the means used to implement this policy was the formation of a
chain of fortified posts, with garrisons sufficiently large to keep the peace in
their immediate neighbourhood and to provide escorts for mail and other
caravans. These were essentially transport staging posts, where food could be
obtained by government caravans, sick porters left to recuperate and, in some
places, porters and animals supplied. They were sited primarily with this fun-
ction in mind rather than as strategically located centres for extending British
164 UGANDA JOURNAL
influence and jurisdiction. New forts were built to meet particular challenges,
or when the alignment of the Uganda Road was changed. Occasional the posts
were used as bases in punitive operations; a few of them remain to this day as
district headquarters. The majority, however, were abandoned when they ceased
to play a part in the Uganda transport system.
The first Uganda Road followed the route pioneered by coastal traders as
far as Mumias and was taken by Thomson in 1883. This, and the realignments
made by Lugard, Martin and others, circumvented the northern side of the
little known Nandi country with its reputedly hostile inhabitants. As a result,
caravans of porters and pack animals were faced with a long march through
foodless country from Fort Smith to Kavirondo. Mumias, which had been
established as a permanent Company post in May 1894, was the central station
on this route to Uganda. Intermediate posts were set up by the Uganda Govern-
ment at Ravine and Guasa Masa, the latter being serviced by a small food
buying depot in eastern Kabras. As a result of the Nandi expedition of 1895,
Sclater was able to shorten the length of his cart road by taking a line to the
south of this earlier track across the Uasin Gishu plateau. The road was
completed in 1896, when Kipture replaced Guasa Masa station and the Kabras
depot was abandoned for a similar post in Kakamega.
With the adoption of the Nyando Valley route for the Uganda Railway,
the telegraph and cart road in 1898, new dispositions had to be made to safe-
guard this area and, in particular, to prevent the Nandi from stealing railway
and telegraph materials and from menacing the working parties. This deviation
to the south first brought the Kipsigis into prominence, and prompted the
various methods designed to prevent an armed combination between the two
allied tribes. In the first instance, small posts were set up at the top and bottom
of the South Nandi escarpment, but as soon as the cart road was completed in
the valley, well garrisoned forts were built at Port Ugowe (Kisumu), Fort
Ternan, Kaptumo, Muhoroni, Kericho, Soba and Lumbwa. It is significant
that the Uganda administration had consistently refused to sanction a post in
Luo country, until the line of communication with Uganda passed through it,
and that requests for stations in Buret in central Kipsigis, on Elgon and to the
north of Ravine were unsuccessful or deferred. These posts could only be justi-
fied on administrative grounds; they had no connection with the Uganda
transport system, and therefore were not sanctioned.
In addition to the major posts protecting the Nyando Valley, there were
minor posts at Molo to the east and Kitotos near the Gulf. The Uganda Railway
stations, construction camps and isolated coolie landis were fortified, and
numerous temporary patrol posts, telegraph repair camps and so on were set
up throughout the Nyando Valley. A considerable proportion of the armed
forces of the Uganda Protectorate were employed as garrisons and on patrol
and escort duties in the area, but the targets were so numerous, dispersed and
vulnerable that these regular forces had to be supplemented by railway
and Protectorate police, friendly Africans, and Sudanese and Indian settlers at
Kibos. Moreover, Uganda's interest in the Eastern Province and the activities
of the Nandi and Kipsigis did not end when the region was transferred to the
East Africa Protectorate (later called Kenya) in April 1902 as detachments of
Uganda Rifles continued to be deployed in the Nyando Valley to safeguard
the railway. Apart from communication needs and military considerations,
Eastern Province stations were considered by the Uganda Government to be
useful for sanitary or recruiting purposes for Indian troops debilitated after
FORTS ON THE UGANDA ROAD 165
service in Kampala and other Protectorate posts. It was suggested that all
Indian troops should spend only six months at a stretch in Uganda and the
remainder of the year in Eastern Province.2 A study of the tremendous
efforts made by the Protectorate authorities to ensure the safety of its com-
munications with the coast underlines the fact that Uganda history cannot
be understood without a knowledge of the Nandi threat to the Uganda lifeline.
It is equally evident that the need to maintain the Uganda Road had a deep and
lasting influence on the pacification and development of western Kenya.
The following are only brief descriptions of some of the forts in and border-
ing Nandi country. Some sites have not been located exactly, but probably
could be through enquiries and fieldwork; more accurate and informative
plans of some others could doubtless be obtained by thorough clearing and
examination and certain points could be discovered and confirmed by ex-
cavations. Nevertheless, it seems fitting to record and describe now, before
further damage or destruction occurs, what is known of these sites. Sketch-
plans (figure 2) are included for those whose defences are preserved in outline,
or those for which the plans are easy to reconstruct. For two sites we are most
fortunate in having Meinertzhagen's original plans.3
These forts were designed to be prepared for intermittent attacks rather
than for continuous hostilities or siege. Though care was taken to select places
from which an unimpeded view of the approaches could be maintained, con-
siderations of supplies, communications and, sometimes, administrative con-
venience were taken into account as well as impregnability. The size, plan and
nature of the defences clearly depended in the first place on the intended purpose
and on local topographical, geological and other factors. However, a rectan-
gular plan with a bastion for a maxim gun at one corner, or at two opposite
corners, was quite common. The defences might consist of stone walls, ditching
and banking, and wire or thorn fencing. There must have been a considerable
amount of fencing on every site, but, except in the cases of those forts for which
the original plans are extant, the alignments can only be surmised. The same
applies to the lay-out of temporary buildings, tents and other features. It seems
that on many sites the main encampment area was normally behind wire or
thorn outside the essential fort.
The order in which the forts and establishments are described represents a
compromise between a geographical and chronological approach. All the sites
that have been precisely located have been given an East African grid map
reference. As on the newer editions of the Survey of Kenya maps this is to be
replaced by a Universal Transverse Mercator grid, the latitude and longitude
are also given, so that the sites can be located fairly closely. Height above sea
level is also given.
Eldama Ravine I (0 03' N, 35 43' E; HAB 015 055; c. 6,800')
Ravine, which is in Tugen (Kamasia) country and on the fringe of Nandi
territory, merits a lengthy study as a key point in the establishment and main-
tenance of British administration in Uganda. It is convenient to begin this
paper with a brief account of the two Ravine fort sites, since they were vital
transport, military and civil centres on the Uganda Road. They enabled the
administration to subdue the turbulent Tugen, and to extend its influence among
the Elgeyo, Njemps and tribes to the north. Until the establishment of the Kip-
166 UGANDA JOURNAL
ture and Fort Ternan posts, Ravine provided the only means of maintaining
contact with the Nandi and Kipsigis. The station was also used as a base for
some of the Nandi expeditions, and was to have been the assembly point for
the Macdonald expedition, which was postponed when the Sudanese mutiny
broke out near Ravine station in September 1897.
Lugard, Macdonald, Portal, Colvile and others camped by the Eldama
River during their journeys to and from Uganda from 1890 onwards. They all
urged that a station should be built there as an intermediate staging post in the
long foodless stretch between Fort Smith and Kavirondo. This view was
supported by the Departmental Committee which advised the Foreign Sectetary
on the administration of Uganda in April 1894. Apart from the need to streng-
then the tenous chain of communications to Uganda, the Committee was prob-
ably influenced in its decision by the fact that the projected route for the Uganda
Railway crossed the Ravine on the way to Port Victoria. Towards the end of
1894 a small temporary road camp guarded by a handful of Sudanese and
Swahili porters was set up.
Martin took command of this camp in January 1895 with orders from
Jackson to build a permanent station on a more easily defensible site higher
up the hill. The original site is half a mile west of the present township on
gently sloping ground beyond a small stream. It is marked by an open-ended
rectangle of gum trees, measuring about 100 by 60 yards. A pit inside this and
some depressions outside to the north and north-west may be original, but it is
not possible to reconstruct the camp plan from the ground.
Eldama Ravine II (0 03' N, 35 44 E; HAB 028 055; c. 7,000' Figure 2)
Men were set to work to build the new fort in January 1895 and on 5 May
that year M rtin evacuated the river site and moved into the new fort on the
top of a steep knoll above the present township. In July 1895 Berkeley decided
that Ravine was such an important link in the Uganda communications system
that he posted a senior officer, Jackson, to take over the station as headquarters
of the Mau district, a command which stretched from the Kedong River in the
east to the Kabras Hills in the west, and which included the most vulnerable
stretches of the Uganda Road. Jackson's chief responsibility was to ensure the
safe and speedy passage of mails, men and materials to Uganda. The supervision
of the transport service and the despatch of loads successively to Port Victoria,
along Sclater's Road, and to Port Ugowe (Kisumu) remained the chief reason
for Ravine's existence until 1900. By that date, the Uganda Railway had been re-
routed to bypass Ravine and to traverse the Nyando Valley, Sclater's Road was
falling into disrepair and Ravine ceased to have any major strategic Importance
and was relegated to the status of a district station. It was only after the over-
riding need to safeguard the Uganda lifeline was removed that any real progress
was made in administering the tribes of the Baringo district stretching north
from Ravine, as had been the case with the Nandi.
During its heyday as the key station on the Uganda Road between Fort
Smith and the Lake, considerable development took place in and around the
fort. Transport and general stores were operated by Smith, Mackenzie, Bous-
tead and Ridley, and an Indian bazaar was established with the ubiquitous
Allidina Visram as the principal merchant. The Uganda Transport Department
established a staging depot and a sanitary station was set up to prevent the
spread of smallpox to Uganda. Land grants were made in the neighbourhood of
Guasao Masa Land above
.. umi s "i ii 0 io 20
N o.k k e a ... ... ............. ........
p. u e ..... AP-
Koiparake Ketparak :.. : -:
:: : :: .... ... .. R vine
SAbush i ..... ..... ... .
U anda RiLwa ........... .... 0
3. :. ... . . . .. L. ... o '* ii
..virondo Ki to-os
168 UGANDA JOURNAL
the fort, mainly with a view to supplying the food requirements of caravans,
and settlements of Sudanese ex-soldiers and Uasin Gishu Masai were established
as a means of strengthening the regular garrison. Although small quantities of
food were procurable from time to time from the local tribes, Ravine was
almost entirely dependant on convoys from Fort Smith, Mumias, Kakamega
and Naivasha for its food supply. This factor limited the number of troops
stationed in the fort-the usual contingent was a quarter or half company only-
and made the station vulnerable to attacks by the Nandi and Tugen. A direct
frontal attack was never mounted against the fort, but stragglers were speared
and cattle looted a short distance outside the perimeter thorn bush enclosure.
Though it has long gone out of use as a military post, Ravine has continued
to opel ate to this day as a district or divisional headquarters. This has tended to
preserve the site, even though the early defence lines have undergone consider-
able damage with the erection of administrative buildings and residences. The
Sudanese colony remains, but the Uasin Gishu Masai were moved to Trans-
Mara in the 1920s and 1930s.
Two virtually concentric square enclosures are believed to be original.
The outer with sides 180 yards long is marked by a ditch up to 8 feet deep and
the slight remains of banks on either side, especially on the inner. A slight
irregulairty near the north-we tern corner may be a secondary feature. Which
of the present entrances are original is not certain. Two measuring 10 feet wide
probably are, while a broad gap on the southern side appears to have been
widened from a smaller earlier break. The inner enclosure measure 60 yards
square and has a thick stone wall. At its highest this stands to 8 feet with a
shallow ditch on its outer side. Only about 3 feet of wall at the most stands
free on the inner side, where the generally higher land has been levelled. A gap
5 feet wide on the southern side, leading onto a wooden causeway over the
ditch perhaps represents the only original entrance. The District Commissioner's
house built in the central enclosure in October 1899 was pulled down a few
years ago. The earliest surviving building is the stone gaol near the north-
western corner of the outer enclosure. It is now put to other purposes. Nearly
280 were voted for this building in December 1899, a large sum in those days,
and it was probably completed the following year. Various other walls and
grasscovered foundations are visible on the site; without detailed research it is
not possible to determine their purposes and relative dates over a period of 70
years. On the western side, immediately below the outer wall, the steep slope
appears to have been terraced. This is believed to have been a camp site for
Guasa Masa. (0 35' N, 35 01' E" approx; west of Land Registration 4287 and
4286; c. 5800')
Some confusion about the site of this fort has been caused by Vandeleur,
who located it in different positions on the map in Campaigning on the Upper
Nile and Niger (1898) and on the one he prepared for Cunningham's report on
the Nandi Expedition of 1895. From internal evidence it is clear that the fort
was constructed on the site marked on the expedition map. This was on a
ridge overolking the east bank of the Kipkarren River and to the south-east
of the confluence of the Kipkarren with its northern tributaries.
In March 1894, Colvile had proposed a subsidiary station on the Guasa
Masa as a link in the chain of posts he thought necessary to maintain his line of
FORTS ON THE UGANDA ROAD 169
communications from the coast to Uganda. Affairs in Bunyoro, however,
prevented him from implementing the proposal. The following May, Jackson
decided to set up the post as a subsidiary station under Hobley at Mumias,
but the project had to be shelved when De Winton, who had been earmarked
for the operation, failed to take up his appointment with the Uganda ad-
ministration. Two months later, on his way up from the coast, Berkeley con-
sidered that intermediate posts should be establiLhed between Ravine and
Kikuyu and between Ravine and Kavirondo to safeguard and improve the mail
and transport services to Uganda. By September, Nandi attacks on Andrew
Dicks's trading partner, Peters West and caravans on the Uganda Road forced
Berkeley's hand, and instructions were given to Sitwell to establish a post to
safeguard the Uganda lifeline. The latter, who had lately arrived at Mumias on
first appointment, set up a food-buying post in Kabras as a supply base for the
new fort, and marched over the Nandi hills to the Kipkarren Valley. The site
he selected was approved by Hobley and Grant as a suitable one for a military
and civil station, and on 25 September 1895 the Union Jack was run up for the
first time on Nandi soil.
With about a hundred Sudanese and a handful of porters Sitwell constructed
a rectangular wooden stockade 50 by 40 yards, a guard room and store house,
a protective outer thorn fence, a bridge over the Kipkarren and linked the fort
with the river and the Uganda Road. The troops were also engaged in es-
corting mails and small caravans and in carrying food from the Kabras depot
until this activity was taken over by 40 Lendu porters sent from Uganda.
Although Guasa Masa prevented the Nandi from continuing their attacks on
caravans in the neighbourhood of the fort, it failed to discourage them from
attacking parties on the roads from Ravine or from the Kabras villages. Cun-
ningham had originally intended using the fort as a base for the Nandi ex-
pedition, but h- changed his plan and Guasa Masa was only used as the starting-
off point for the second column under Sitwell and Foaker, and as an escort-
providing post for Uganda caravans and mails. In December at the close of the
first phase of the Nandi expedition, Foaker took over charge from Sitwell but
it is unlikely, with the few men under his command, that he was able to increase
the buildings in the fort, to undertake any work in stone or to strengthen the
fortifications by ditches. A published photograph shows what appears to be a
thick thorn-fence with bare rock outside.5
Guasa Masa had been a temporary expedient to meet a suddent challenge
to Uganda's communications, but it had no administrative value and had to be
provisioned from Mumias, Ravine, Kabras and Kakamega, and serviced from
Entebbe. When Cunningham's proposals for re-routing the Uganda Road
through Nandi and Kakamega were approved by Berkeley and Sclater, it was
only a matter of time before Guasa Masa was abandoned. On 27 May 1896
Foaker left the fort for Kipture. A native offer with 14 other ranks was left
in charge of the station goods and cattle, but shortly afterwards this token
garrison was withdrawn and the fort finally abandoned.
Kipture (0 11' N, 35 09' E; HAA 374 196; c. 6,400': Figure 2: photo. 1.)
This was the second fort and the first permanent post in Nandi country.
It is occasionally referred to as Kamsikak Fort but like its successor at Kap-
tumo, more often as Nandi or Nandi Fort.
( Numbers- see bext)
i ..I.. mA Earichwork
I -i Stonework
M Maxim gun stand
* NabouraL rocks
Figure 2 Sketch outlines of some Forts in and near Nandi Country.
FORTS ON THE UGANDA ROAD 171
more often as Nandi or Nandi Fort.
The site is on a ridge top, from which the land falls away to north and
south and very slightly to the west. The fort commands some long views,
besides covering its immediate surrounds. Sclater's Road passed close to the
south of the post, but its alignment seems to have been obscured by cultivation
over this stretch.
The fort itself remains quite well preserved. Though its southern side has
been used for shambas, most of it has been left to grass and a few trees, including
some gums planted during the time it was occupied as a station. It is rectangular
in shape, about 140 by 75 yards, with a projection 20 yards square at the north-
west corner. Huntingford in an unpublished report recorded that the sur-
rounding ditch was 5 to 10 feet deep in 1926; it nowhere exceeds 5 feet now.
On its inner side there are traces of a bank. There are five gaps, of which two-
one on the southern side near the south-west corner, the other near the middle
of the eastern side-appear to have been part of the original lay-out. Several
foundations for buildings, and possibly also gun emplacements, are visible as
small mounds inside. The largest reaches three feet in height. A broad mound
outside on the western side, 21 feet high, may possibly have been caused by
ants. The Nandi called these defences at Kipture irimet, after the fortified
villages they encountered when raiding the Abaluhya to their west.
Although the Kipture site had many advantages over Guasa Masa as an
administrative centre and an aid to pacification, and was, moreover, situated
in an occupied, food-producing area, it was established primarily as a base for
Sclater, while he pushed his cart road through from Ravine to Kakamega, and
thereafter as a garrison post to ensure the safety of caravans along the Uganda
Road. The new route decided upon by Sclater, as a result of the exploratory
work carried out by Cunningham and Vandeleur during the 1895 Nandi ex-
pedition, reduced the distance from Ravine to Mumias by several days, com-
pared with the old caravan track through Guasa Masa. In addition, only the
first two days' marches from Ravine were through foodless country, in contrast
to the long journey across the unoccupied Uasin Gishu plateau to Kabras.
Friendly relations were established with the Nandi near the post and the fort
was used on occasions as a base for minor punitive measures, but little attempt
was made to administer or control the tribes from Kipture. No traders or
transporters set up shops or stores and no land was leased for development.
The Uganda Transport Department ran a depot at the fort and a number of
Uasin Gishu Masai settled nearby. Kipture began and continued as a transit
staging post on the Uganda line rather than as an administrative centre for the
When Sclater and Foaker made a reconnaissance of the site in May 1896,
they made a temporary fort at Kapsigak (hence the Kamsigak Fort) until the
ine for the Sclater's Road had been settled. Berkeley and Jackson joined them
there in July and in the following month Bagge took over from Foaker the task
of building a fort on the site agreed upon by Berkeley and Sclater. No help was
forthcoming from the Nandi, so the construction was undertaken by the garrison
company of Sudanese, Swahili porters and Sclater's men. The garrison was
later reduced to half a company, but even so food had to be transported for the
station needs from Kakamega and Kitotos. During the Sudanese mutiny of
September 1897, Bagnall, the officer in charge, was unable to prevent the
mutineers from looting the station stores and ammunition. For several
172 UGANDA JOURNAL
months Kipture was left to a native headman with a token garrison. In February
1898, Jackson re-opened the fort.
The despatch of the first convoy from Kipture to the newly established post
at Ugowe Bay (Kisumu) in August 1899 led to the abandonment of the Sclater's
Road route to Kakamega and Mumias. For some time before this it was realized
that Kipture was sited on the fringe of Nandi country, and after the 1900
expedition it was decided to move the station to the south-west nearer the
Uganda Railway. In February 1901 the dilapidated buildings at Kipture were
abandoned and the station moved to Kaptumo. It was hoped that the new
station would discourage the Nandi from interfering with the new transport
route to Uganda down the Nyando Valley. The second Nandi station was
abandoned, like the first, when it had ceased to serve a useful purpose as a
staging post on the Uganda Road.
Abushiri (0 1' N, 35 1' E; HAA 230 010; c. 6,500')
This fortified camp, often referred to as Bushiri or Camp Ashirini na moja,
was established in June 1899 as the terminal station on the south Nandi escarp-
ment for the new cart route from Kipture to Port Ugowe (Kisumu) and thence
to Buganda, in preference to the longer route for goods and supplies along
Sclater's Road through Kakamega and Mumias. It was sited, some 7 miles
south-west of Camp Ashirin (Kaptumo), on a hill by the present Kapkolei
dispensary and school, at the head of a valley leading down to the Kitoto
(Kano) plains. There are no records of major works having been undertaken
on the site, and visible remains are very meagre and dubious. Carts from
Kipture turned at the post, and their loads were taken by porters down the
2,000 feet escarpment to a transit camp on the plains four miles from Abushiri.
This subsidiary camp, which received loads despatched in carts from Port
Ugowe, was called variously Bagamoyo, Mark's Boma or Sudi's, after the
transporter, Sudi C. Melimin, who operated the escarpment stretch under
contract with the Uganda Transport Department. The first convoys left Port
Ugowe and Kipture in August 1899, and this shuttle-service continued at
regular weekly intervals, until the escarpment route was abandoned in October
1900 for the road down the Nyando Valley.
Abushiri had been used as a food buying post for the Kipture garrison by
Bagnall as early as March 1897, and Cooper, when he set up the escarpment
station in June 1899, had found the populous local sections well disposed.
But it was soon evident that both the fort and the stretch of road between
Kaptumo and Abushiri were in areas over which Kipture had little control.
Though the formidable cart convoys passed without molestation, smaller
parties suffered severely at the hands of the Nandi warriors. Several of the
incidents which led to the 1900 expedition against the Nandi occurred in this
region, including the murder of Robert Livingstone6, successful attacks on a
caravan carrying a steam launch for Boustead, Ridley and Co., and on Uganda
Transport road repair parties. The fort, with its small garrison under a Sudanese
sergeant, failed to provide any protection out of gunshot range of its defences,
or to discourage the escarpment Nandi from raiding Kitoto's villages on the
plains. The fort itself was in jeopardy from time to time, and Colonel Evatt
was disquieted by its vulnerability, when he camped there in July 1900 at the
beginning of the 1900 expedition.
When Abushiri was evacuated and the cart convoys switched to the Nyando
Valley route, the fort had contributed little, if anything, to the extension of
FORTS ON THE UGANDA ROAD
-.. .: .: .: .,.;: -:W FW
Kip- e n h- ...'d-.-m,.e..
Kipture: northern ditch from outside 1964
Kaptumo II Northern ditch from outside, 1964
FORTS ON THE UGANDA ROAD 175
British control in Nandi, but had played a significant part in speeding the
movement of goods along the Uganda Road.
Kaptumo I (civil) (0 04' N, 35 04' E; HAA 292 077; c. 6,400' Figure 2.)
Johnston approved the move of the district headquarters from Kipture in
October 1900, and Hobley chose Kaptumo as the site for the new station.
This had previously been a staging post on the cart road between Kipture and
Abushiri, and had been known as Camp Ashirin (presumably so named as
the 20th camp on the road-from the Arabic Swahili asharini, 'twenty', now
commonly ishirini. According to porter caravan schedules, this enumeration
would make Machakos, Camp I. It is probable for carts along the Sclater's
Road, which by-passed Machakos, that some of the longer marches took
two days to complete, so that Kaptumo might be the 20th camp from Fort
Smith. The next camp to the west, Abushiri, is sometimes referred to as Ashirini
na moja, 'the twenty-first.7 In January 1901 Hobley measured out the station
"in accordance with the approved plan" and 600 were allocated from savings
for its construction. The fort was built by troops, station hands and Sudanese
settlers, supervised by specially recruited artisans. Hobley proposed transferring
the Uasin Gishu Masai colony at Kipture to Fort Ternan, but shortly after
the Kaptumo station had been occupied in February 1901, Uasin Gishu settlers
established themselves to the south-east of the fort.
The fort, which crowns a small rise, provides good all round views and a
clear field of fire. The two mounds acting as bastions on opposite corners of a
rectangle recall some of the other sites. Both reach some 10 feet in height and
the southern one still shows traces of buildings on it. They are 30 yards apart
and the ditches completing the rectangle between them remain steep and up to
10 feet in depth. The entrance gap on the southern side leads into an impressive
avenue of gum trees which Meinertzhagen planted. In the north-west corner
is a deep pit said by the Nandi to have been used as a goal. Inside the fort
perimeter, the foundations of two buildings are traceable, one of which is said
to have been the District Commissioner's house built in 1904. On the sloping
ground to the north is a larger square enclosure in which the askari lines were
built. This area is bounded by a ditch, now reaching 5 feet in depth, and a
narrow bank inside up to 7 feet high. The parade ground was located on
fairly level ground to the west of the fort.
The site and its surrounds is still used as an administrative sub-station for
southern Nandi and the site has undergone some modification. The mound
for the southern bastion has been used for holding barazas, while the northern
now supports a water tank. The interior of the askari enclosure and some ad-
joining land have been used as an experimental plot for cash crops.
The station was abandoned in November 1907, when district headquarters
was transferred to Kapsabet-the first station to be sited solely for adminis-
trative convenience rather than as a link in the chain of communication between
the coast and Uganda.
Kaptumo II (military) (0 05' N, 35 04' E; HAA 296 088; c. 6,400'. Figure 2;
This site lies two-thirds of a mile north of the civil boma (Kaptumo I).
It was constructed by Meinertzhagen in May 1905 around what had previously
been an unprotected house built under contract by Bhedwar in mid-1903.'
It commands good all rounds views.
176 UGANDA JOURNAL
Meinertzhagen's plan is most useful for interpreting other forts, notably
Fort Ternan I. All that now remain are the bastions with lowered but distinct
grass-covered parapets and ditches up to 4 feet deep. All trace of buildings has
been destroyed by later settlement on the site. Two forts were needed at Kap-
tumo to accommodate the civil and transport staff and the garrison, which was
normally of company strength.
Koiparak (0 07' N, 34 54' E; GAF 674 136 approx; c. 6,000'. Figure 2).
Known also as Kapwaren after this part of the South Nandi forest, this
fort was established by Meinertzhagen in June 1904 among the Koiparak
granite kopjess. Though it was rapidly erected and only used for a short period
following the murder of Wendte, an American visitor to the nearby Kaimosi
mission, Meinertzhagen's plan in his manuscript diary shows it to have been
a small but well devised establishment.
Kopjes block the longer view to the north and north-east, but the fort
commands adequate views over its immediate approaches. The rectangular
plan with a corner bastion was modified to utilise some rocks up to 24 feet high.
Earthworks reaching 10 feet in height and palisades or thorn fences around
the bastion and on part of the northern side completed the defences. All that
now remain of these are some stretches of ditch up to 3 feet deep. Sentry boxes
were placed in the bastion and by both entrances. Inside the roughly rectangular
enclosure 50 by 40 yards in extent were lines for askari and porters, an officer's
tent, store, office, cookhouse, sheep-pen and cattle-kraal.
In July 1904 Meinertzhagen handed over command of the fort to Wilson
and, after the Kapwaren Nandi had settled down, the half company of troops
was withdrawn and the fort abandoned. The withdrawal took place before
June 1905 as at that time Bagge was asking for a police post at Kaimosi to
guarantee the safety of the missionaries, who were evacuated to the C.M.S.
station in Maragoli during the Nandi expedition in the following October.
Fort Ternan I (0 12' S, 35 21' E; HZN 599 769; c. 5,100'. Figure 2.)
The military post, from which the railway station and small township of
Fort Ternan later took their name, was established in July 1899. After Blackett's
new route for the Uganda Railway down the Nyando Valley and been approved
by the Chief Engineer in November 1896, the Uganda Protectorate authorities
were forced to make fresh dispositions to protect the survey and construction
parties, and to safeguard road and telegraph communications between railhead
and Uganda. It was decided to press on with a cart road over Mau to Ugowe
Bay (Kavirondo Gulf) and to carry the loads of the lake-steamer, the Wiliam
Mackinnon, over this route. Ternan, who was Commandant of the Uganda
Rifles and also Acting Commissioner of the Uganda Protectorate at the time,
instructed Coles in May 1899 to reconnoitre a route through the valley and to
select sites for military posts along it. Coles was O.C. Road Military District,
a command which had been specifically instituted to secure the free passage
of men and materials to Uganda. After he had reported favourably on both
the route alignment and camp site, Coles ordered Pereira with details from
16 Company to march in September from the recently established station at
Ugowe Bay (Kisumu) to build a fortified post at a place Coles described as the
Cross Roads. The post continued to be known by this name, and also as Lum-
bwa Post, for some time, although the name Fort Teman was in current use
shortly after Pereira left Port Ugowe. It is assumed that Coles named the post
FORTS ON THE UGANDA ROAD 177
after his chief, and the designation was later accepted by the Uganda Railway.
Ternan first visited the fort in February 1900.
The site9 chosen by Coles is a third of a mile south of the railway station
on the southern bank of the River Namuting, which cuts at this place a small
gorge some 40 feet deep, above which the land slopes only very gently. The site
is reasonably defensible, and was probably deliberately chosen close to water.
The original cart track down the Nyando Valley appears, on the evidence of
cuttings and embankments, to have passed some 300 yards to the south of the
fort. From this and a disused Nandi and Kipsigis cattle track, which crosses the
Namuting at a salt-lick a short distance west of the fort, is derived the name
Cross Roads. Uganda officials and soldiers had noted previously that there
was a good deal of traffic on the cattle track between Nandi and Kipsigis,
and that this route was taken by Nandi when they wished to secrete their
cattle among the herds of their Kipsigis kinsmen in times of trouble. It was
hoped that a fort in this strategic position would discourage an alliance between
the two tribes, and would prevent the Nandi using the cattle track as an escape-
route for their cattle if punitive operations were mounted against them.
About May 1901, as the railway approached Fort Ternan, the fort was
abandoned for another to the north of the Namuting and the railway (see
Fort Ternan II below). During the short period of its existence the original
fort had provided escorts for caravans, mail and telegraph repair parties on
the Uganda Road. The garrison, which was normally of half to one company
strength, had also taken part in punitive operations against the Nandi. Early in
1900 the post acquired added importance as the headquarters of the Uganda
troops stationed in the Nyando Valley and adjoining areas. The fort had not
prevented the Nandi from attacking the Uganda lifeline, but it had contained
them and had probably been a decisive factor in discouraging an aggressive
alliance between the Nandi and Kipsigis.
The surviving or traceable features of the fort are set out in an appendix
and indicated by the same numbers as those used on the plan in figure 2.10
The various features of this fort site would seem to make better sense by
comparing them with Meinertzhagen's plan of Kaptumo II. The two main
enclosures at Fort Ternan would seem to represent bastions, in which the raised
platform in the northern (and perhaps also in the southern) would have been
for maxim guns. The use of stone walls at Fort Ternan, as opposed to the banks
and ditches at Kaptumo, merely reflects the local geology. The second entrance
in the northern bastion at Fort Ternan was presumably to give access to the
river. Probably the walls were surmounted by some form of fence, and a barbed
wire fence would have surrounded the area of the bastions and intervening space.
No fence wire has been noticed on the site. The sunken pits may have been
officers' quarters, sentry posts to guard the entrance to the southern enclosure,
The site of the damaged camp, and probably also that of the telegraph
line construction camp and the stone enclosures to the east, would have been
outside this central complex. From time to time considerable areas round about
the fortified post were used as tent lines for troops, porters, railway and tele-
graph workers and transport staff according to demands. Outer lines of fencing
and thorn bomas must be imagined, while the area of the bastions would
have been a citadel.
178 UGANDA JOURNAL
Fort Ternan Outpost (0 13' S, 35 20' E; HZN 597 759; c. 5,300')
Two-thirds of a mile south of Fort Ternan I, a small military site was iden-
tified on a ridge known as Kapkures and noted, archaeologically, for a group
of Late Stone Age cairns.
A triangular enclosure with sides measuring about 105, 90 and 60 feet
seemed apparent from some very rough and dilapidated lines of stones. It is
doubtful whether these were built walls, though the middle-sized of the three,
in which a cartridge was found, had been a substantial breastwork along the
edge of the ridge. This commanded an excellent view to the west and north,
including the site of Fort Ternan I and a large expanse of valley. An old track-
way ascending the hillside to the outpost seems discernible on the left. A pile
of wood ash and a heap of stones some 3 feet high and 20 feet broad existed
within the enclosure. On being excavated, the former was found to contain
food bones and nails; the latter a considerable amount of scrap metal and other
rubbish. Some of this is clearly later than the Nandi war, presumably it was
collected by the Kipsigis inhabiting Kapkures in recent years.
While this site may have been merely an outpost for guarding Fort Ternan
and its approaches, it could equally well have been a heliograph station to
communicate with Kaptumo or Ketparak on the Nandi escarpment, neither
of which is visible from the valley. Unfortunately it is not possible to establish
its date, but its position suggests that it was probably connected with Fort
Fort Ternan II
When the station at Fort Ternan was moved in May 1901, Gorges (O.C.
Eastern Military District, a military command which replaced the Road
Military District) noted that the new fort was situated "some distance off
the caravan road" and recommended that a civil officer should be detailed to
carry out the administrative functions undertaken by the military commander.
No administrative officer was sent and the station remained a military centre,
with the commanding officer keeping in touch with the neighboring Nandi
and Kipsigis, superintending the Indian bazaar, itinerant traders and the
Uganda Transport service and representing Uganda Protectorate in matters
concerning the Uganda Railway construction, defence and coolie camps.
When two companies of the Indian Contingent of the 5 K.A.R. occupied
tented encampments at the station in mid-1901, Fort Ternan became an im-
portant centre with a hospital supervised by a resident medical officer. Though
some food was bought from the Kipsigis, the bulk of the supply came from
Kitotos, where the railway had a permanent buying post under a European
storekeeper, and from Port Ugowe.
Freight traffic reached Fort Ternan in October 1901 but the railway station
buildings were not completed till the following June. After the rails reached
Kisumu in December 1901, over 8,000 railway workers were employed in the
Nyando Valley improving the line. Two Indian companies of the Uganda
Rifles, often supplemented by companies of local troops, were tied down in
the then Eastern Province of Uganda to protect the railway workers and to
safeguard the railway and telegraph installations. On occasions even this con-
centration of force and the measures taken to fortify railway camps and landis
did not prevent the Nandi from disrupting communications between the coast
and Uganda. The maintenance of these communications was so important to
Uganda that the Protectorate authorities agreed to maintain the two Indian
FORTS ON THE UGANDA ROAD 179
companies at Fort Ternan, after the Eastern Province had been transferred to
the sister protectorate in April 1902. The East Africa Protectorate paid 10,000
a year towards the cost of the Uganda garrison, which, apart from routine
escort, patrol and guard duties, also took part in military demonstrations and
in punitive operations against the Nandi.
The founding of Kericho post in June 1902 diminished Fort Ternan's value
as a point of contact with the Kipsigis. After an epidemic of malaria in April
1903, most of the non-government residents left Fort Ternan, the hospital
was evacuated to Londiani, and the Fort Ternan Times ceased publication.
The following month military headquarters were moved to Muhoroni and the
post office closed down. A detachment remained at the fort but, in April 1094
with the withdrawal of the Indian Contingent and their return to India, Fort
Ternan was abandoned as a military station. When Winston Churchill stayed
there in December 1907 on his way to Uganda, he described it as a placelesss
The site of Fort Ternan II cannot be definitely identified, though the local
belief that it is on the Fort Estate, north of the railway and west of the station,
seems fairly convincing. Two possible sites have been noted on the estate:
the first is indicated by some enigmatic archaeological remains, the second
by local tradition.
The first site (0 12' S, 35 20' E; HZN 580 775; c. 5,100') is a small peninsular
eminence about one acre in extent on the south-esatern side of the Murgot
river. The site certainly commends itself and has good command of its approa-
ches. Immediately upstream is a river crossing, which was possibly used by
the first cart road. Below the site, the river runs into a small gorge. There is a
vague suggestion of a stone bank or footing having cut off the peninsular.
On the eminence are a number of stone lines, representing not walls but fence
footings for a system of mostly rectangular enclosures. In one of these, the
foundations of a house, consisting of two rooms each 18 feet square, are visible.
These remains suggest an old farm site more than a military one, but there is no
memory of this locally. There is a rumour that a cartridge or cartridges were
picked up here.
The second possible site (0 121' S, 35 19' E; HZN 575 768; c. 5,200') is a
knoll, half a mile south-west of the first site, and one and a half miles west of
the railway station. It equally well commends itself with all-round views across
the Murgot to the west and north, and along a good stretch of the Namuting
valley and railway to the south. On the south-western side of the knoll is a
flattish area, which appears to have been cleared of stones, but reveals no
definite indications. However, there is a local belief (unfortunately not con-
firmed at first-hand) that this was the fort site; again, a cartridge was picked
up here. A house built later on the other side of the knoll is now in ruins.
Muhoroni (0 9' S, 35 12' E; HZN 440 830; c. 4,300')
Although small detachments of troops from Fort Ternan may have used
Muhoroni as a camping site earlier, it was not until late 1901 that a permanent
post was established south of the Muhoroni river and north-east of the railway
and cart road.12 After the opening of the railway station in January 1902,
a company of the Indian Contingent was posted there to protect the railway,
telegraph and the Uganda Road. No civil officer was stationed at Muhoroni,
and the role of the garrison was similar to that performed by the troops at
Fort Ternan. After the Kamelilo Expedition of April-May 1903, Muhoroni
replaced Fort Ternan as the military headquarters. In July that year, a contri-
180 UGANDA JOURNAL
butor to the East Africa and Uganda Mail (not always a reliable source of
information), reported that barracks were being built a mile and a half from
Muhoroni railway station. In April 1094 the Indian troops were withdrawn,
but in December a company of local troops was rushed to the station to safe-
guard the line from Muhoroni to Kibos fron Namdi attacks. During the Nandi
expedition of 1905, a column of the field force was based at Muhoroni, where
the Commissioner had peace talks with Nandi "chiefs" at the close of the
operations. After the Nandi had been moved from the Nyando Valley into a
reserve centred on Kabiyet, Muhoroni lost its value as a military station and
was abandoned as such.
The site of the camp or camps is unknown, but local enquiries and inspection
of the ground would probably be productive, as in its short hey-day as military
headquarters considerable construction work must have been undertaken.
It is said that, several years ago, some ditches were visible and cartridges
noticed a mile north-west of the township.
Soba (0 05' S, 35 16' E; HZN 525 910; c. 6,800')
This sub-station is sometimes referred to as Kamelilo, as it was established
in the territory of the Nandi pororiet of that name. The Kamelilo warriors had
terrorized the Uganda Railway staff and prevented the regular passage of trains
to and from Lake Victoria. After a punitive expedition had been mounted
against them in April 1903, the Foreign Office approved Eliot's proposal to
establish a station in Kamelilo country to ensure the safety of the rail connec-
tion to Uganda. 700 were allotted for the construction of the post, which was
sited by Coles about June 1903. Building began in August under Maclean and
on 24 October Soba was gazetted as an administrative station.
Local information asserts that the fort was built on an exposed saddle
which extends south-south-west from the summit and crater of Soba. This is
confirmed by Bacon's map drawn during the 1905-6 campaign. Two photogra-
phs in Sphere (6 January 1906, p. 35) show that the fort was square or rectan-
gular, defended by barbed wire staked fences and probably ditches about 40 to 60
yards long. It contained several buidlings with roofs of corrugated iron. All that
can be seen now are a few stones which might once have belonged to walls.
In June 1905 two hostile night demonstrations were made by the Nandi
against the garrison. The following September the fort was abandoned in favour
of a new post near Lumbwa railway station. Soba was not immediately written
off; for during the campaign that followed it served as a useful camp-site and
was worthy of an inspection by General Manning.13
Lumbwa (0 8' S, 35 28' E; HZN 737 780; c. 6,900')
This is the existing Lumbwa township on the Uganda Railway (not to be
confused with Fort Ternan which, as mentioned above, was sometimes known
by this name). In September 1905 Monckton, the civil officer in charge of Soba,
moved the post in Kamelilo country to Lumbwa. This move, shortly before the
punitive expedition, was said by some observers to have given encouragement
to the Nandi. Lumbwa was used as a base point for the expedition, but seems to
have been given up as an administrative station soon after the Nandi settlement,
probably on the formation of the Sotik Post in southern Kipsigis in 1906. A
company of troops remained there until they were despatched on the Kisii
expedition in January 1908, after which the post was abandoned as a military
FORTS ON THE UGANDA ROAD 181
There is no knowledge of any remains of the early civil or military post, but
again a search might prove rewarding.
Kericho (0 22' S, 35 17' E; HZN 535 595; c. 6,700')
The early railway surveyors reported unfavourably on a route to Uganda
through Kipsigis country, which was thus left almost untouched by the admini-
stration. However, as already seen, once the Nyando Valley became strategic-
ally important, one of the reasons for founding Fort Ternan in 1899 was to
prevent the closely related Nandi and Kipsigis (or Lumbwa, as they were
erroneously called) from combining in attacks or hiding each other's cattle
during punitive expeditions. Some success was achieved, but Fort Ternanr
could not stop the Kipsigis from raiding the Kisii and the friendly Luo. More-
over, as the railway pushed forward into the Nyando Valley, the Kipsigis
emulated the Nandi in their attacks on the line and staff. Hobley had pressed
for several years for a post in Kipsigis country, but it was not until some 500
warriors were provoked in April 1902 to raid a large.construction camp near
Fort Ternan that the Foreign Office agreed to set up a fort to control the tribe
and to attempt to isolate them from the Nandi. A board was appointed in May
1902, as a result of which its chairman, Hobley, authorised Gorges to build a
station at Kericho, cautiously placed only 12 miles as the crow flies from Fort
Ternan. 600 were voted for the new station and for connecting roads with
Fort Ternan and Muhoroni.
Gorges began building in June and quickly got into touch with the Kipsigis
leaders. As a result of his sympathetic administration and that of his successor,
Partington, Kipsigis raids on the Uganda Railway and road almost ceased, and
the neutrality of the tribe was assured in both the 1903 and 1905-6 Nandi
expediditions. It was not long before Hoble.y was pressing, unsuccessfully, for a
post further south in Buret, and at the conclusion of the 1905 expedition it was
found necessary to establish a station in Sotik.
Kericho began as a joint civil and military station and has continued as a
district headquarters. Troops remained there for some time after the abandon-
ment of the Nyando Valley stations, but were later transferred to Nairobi and
only returned to Kericho as occasion demanded. Little seems to have survived of
its earlier period.14
Kisumu (0 6' S, 34 35' E; GZU 500 890; c. 3,750')
When Whitehouse decided in November 1898 that the terminus of the
Uganda Railways would be on the Kavirondo Gulf (then called Ugowe Bay)
rather than at the more northerly Port Victoria site selected by the railway
survey in 1892, the Uganda administration was forced to establish a civil
station at Port Ugowe. For many years Hobley had advocated a station at
Kitotos, in what was then known as South Kavirondo, as he was unable to
control this district, or to make use of its food supplies, from the Munias
headquarters. Although it was some time before the railway reached Ugowe
Bay, the service and cart roads pushed through the Nyando Valley ahead of
the railway construction works shortened the land journey to the lake via
Mumias by several days. From the port on Ugowe Bay, loads could be shipped
to entebbe (also called Port Alice) and Manyunyu (the port for Kampala)
by the government steel boat, dhows and Basoga canoes. Furthermore, the
William Mackinnon steamer was to be assembled near the new terminus by
the railway authorities, a long awaited event which it was hoped would speed
182 UGANDA JOURNAL
the transport of government freight. The post at Port Ugowe was, therefore,
established as part of the Uganda transport system rather than for administrative
reasons. It was soon to eclipse Mumias as the leading centre in Kavirondo,
as Mumias ceased tp be a key station on the Uganda Road and became merely
an administrative centre for the North Kavirondo district.
On 13 June 1899, Coles established a military post at Port Ugowe, shortly
before its eastern counterpart was set up at Fort Ternan. The following month,
Gait established a civil post at Port Ugowe by transferring men and stores
from Port Victoria, which was then abandoned as a district station. At the same
time Nandi, which was more easily supervised from Port Ugowe than from
Ravine, was transferred from Mau to Kavirondo district. By the time Hobley
moved his headquarters from Mumias in May 1900, the Uganda Railway had
occupied the gulf head area, which the Chief Engineer had designated as Port
Florence. Towards the end of 1901, the civil station was moved to the south of
the gulf head to the position it occupies today. The garrison at Port Ugowe
was never a large one, but the station was used as a base in some of the Nandi
expeditions. From his new headquarters, Hobley was able to get in touch with
the Luo sections and to extend the Protectorate's influence to the tribes to the
south. But in its early years, Kisumu (as it came to be called from its Luo name)
was essentially a railway town and a transshipment port for Uganda and German
Other Eastern Province forts and posts not described in detail in this paper
are Fort Purkiss (Berkeley Bay), Mumias, Port Victoria, Kabras, Naivasha,
Kakamega (Fort Maxted), Kitotos, and the Baringo posts. Also E.A.P. posts
existed at Karungu and Sotik.
Description of features and finds at Fort Ternan I
1. A pentagonal stone wall enclosure, covering about a ninth of an acre, immediately
above the gorge. It has two entrances, both probably 4 feet wide originally, one on the south-
ern side, the other next to a platform 9 or 10 feet square and raised about l feet in the north-
ern apex. From here to the southern entrance is exactly 100 feet, and on this axis the enclosure
seems to have been planned symmetrically with walls 50, 70 and 30 feet long, though in fact
the measurements are not exact nor the alignments true. In the south-east it appears that the
wall collapsed and was rebuilt on a crooked line.
The walls are dry-built of shapeless lava boulders, with little attempt at coursing, and a
width of 3 feet. With allowance for fallen stones the walls would have stood some 4 feet
high, though now they nowhere exceed 3j feet. The side overlooking the river forms more of a
breastwork on the crest of the gorge than a free-standing wall.
Inside the enclosure is a line of stones suggesting a wall foundation, and some half dozen
piles of stones reaching 2 feet in height and 4 feet in width. These piles, which are not arranged
in any pattern, are probably of no military significance. Nor are they likely to have resulted
from agricultural activity as there is no indication or local memory of the enclosure having
been used for this purpose. They could conceivably be graves, dating after the abandonment
of the fort. The site may have been used as a railway workers' camp, and the enclosure may
have recommended itself as a graveyard.
2. A second stone-wall enclosure 100 yards to the south. It measures about a fourteenth
of an acre, and is sub-rectangular in plan with a broad entrance facing the northern enclosure.
Half of the entrance is blocked by a low curving wall, which looks like secondary work. A
tumble of stones in the south-eastern corner may represent a raised platform like that in the
The walls are similar to those forming the first enclosure, but rather more collapsed,
probably because the interior was recently used as a shamba the cultivator does not remember
any features or finds of interest.
FORTS ON THE UGANDA ROAD 183
3. Two pairs of stone-lined pits covering the entrance to enclosure no. 2 The western
pair is now filled in. Tent ferrules were found in the centres of both pits. The eastern pair is
smaller, with its walls raised at front and back to provide a curtain to the entrance enclosure,
thus giving a slit-trench appearance. These pits may have been designed for tents sunken to
protect them from fire, or their inhabitants from arrows.
4. Stone enclosure to the east, virtually destroyed and not possible to plan in detail. It is
understood that one of them measured about 30 feet by 10 feet, with an entrance on a longer
side facing the river, while beside it were nine smaller open-ended rectangles arranged in
lines of three. A porter's metal tag found here may point to their purpose.
5. A telegraph line construction camp or temporary telegraph post, to the west of the large
enclosures. There is nothing visible now, but some telegraph line nails were found in a tree,
and close to it a morse key, a small metal plate, inscribed "U.R." (Uganda Railway) and a
fair amount of domestic rubbish. A temporary telegraph line was hastily constructed from
railhead by railway engineers at a cost to Uganda of 14,000. It was operating to Fort Ternan
in October 1899 and to Kisumu by the end of the year. The permanent line reached Fort
Ternan in November 1901. Nandi attacks on the line began in October 1899. The temporary
line followed the cart road; the permanent line the railway.
6. The site of what appears to have been an overrun or burnt-out camp, somewhat apart
to the south-west. A large variety of objects were found within a small area in the camp in-
cluding parts of machine guns and rifles, cartridges; camp equipment such as tent grommets
and ferrules; military brass buttons, cap badges (including 3 of Uganda Rifles, 1 of Sudanese
crescent and star) buckles etc; domestic equipment including tin-openers, soda-syphon bulbs
and bottle glass; certain trade goods such as copper wire coils, cowrie shells and beads mostly
blue and white though including two carnelian and several native iron arrow and spearheads
and some almost mint coins including an Indian rupee of 1898, 9 IBEAC 1888 copper and 8
EAP piece of 1897 and 1898.
These finds suggest, either that some tents housing both British officers and local troops
was surprised at night by Nandi or possibly Kipsigis and set on fire, or that a fire broke out in
the encampment, with the result that considerable quantities of possessions and clothing were
destroyed or damaged. This event clearly took place not long after a mint consignment of 1898
coins had been paid out. Although camps, landis and parties of men were attacked in the
neighbourhood of Fort Ternan, there is no record of any assault on or close to the fort itself.
During the 1900 Nandi Expedition the fort was beleaguered until relieved from Ravine, but
no attack was made. In November 1901 the government herd had been raided and the herds-
man speared, but there is no mention of an assault on any encampment, which could mean
that the fire was accidental. Because of the number of men, and quantities of stores and am-
munition packed into the confined spaces of the camps, the fire hazard was great, and many
cases are recorded from other places in the Uganda Protectorate of damage caused by fire. On
the other hand the arrows and fired, as opposed to blown, cartridges suggest that fighting did
take place. They could, however, have been lying about the camp before, or have been dumped
along with some of the other rubbish afterwards. A burnt-out camp-site might have recom-
mended itself as a place for subsequent incineration.
1. Entebbe Secretariat Archives A/9/1; 7-10-00.
2. FO 2/599. For the transfer see Ingham, Uganda J., 21 (1957) and Matson ibid 22 (1958).
3. For an account of the expeditions against the Nandi told from Memoirs, military and
political records see H. Moyse-Bartlett, The King's African Rifles (1956) and R. Meinertz-
hagen Kenya Diary 1902-1906 (1957).
4. Thanks are due to Major J. S. Ross for information on both of the Ravine sites. See
his articles "Eldama Ravine and military history", Askari 1(964).
5. Geographical J., 9, (1987), p. 385.
6. A manumitted slave with an interesting career. See H. B. Thomas, Uganda J. 23 (1959)
7. Meinertzhagen, op. cit., pp. 193-194.
8. Ibid. p. 167.
9. See Lt. Wortham's map in FO 2/804, Uganda Intelligence Reports II, May, 1901,
184 UGANDA JOURNAL
10. Especial thanks are due to Mr. F. D. P. Wicker for pointing out these features, and for
information on finds.
11. W. S. Churchill, My African journey, (1908), p. 82.
12. See sketch-map in FO 2/804 IX, February 1902.
13. Meinertzhagan, op. cit., pp. 252 & 265.
14. For an account of Kericho, see Matson, "The founding of Kericho", Kenya Weekly
News, 31st October 1958.
The authors gratefully acknowledge the permission granted by Col. R. Meinertzhagen
and Messrs Oliver and Boyd to use the plan of Kaptumo II from page 193 of Kenya
Diary 1902-1906; and by Col. Meinertzhagen for the plan of the Koiparak fort re-
drawn from his manuscript diary.
Uganda Journal, 29, 2 (1965) pp. 185-192
H. M. STANLEY'S JOURNEY THROUGH ANKOLE IN 1889
By NTARE SCHOOL HISTORY SOCIETY*
The authors, T. Kabwegyere, E. Rwakasisi, F. Byakunaga, G.
Alibaruho, S. Baitwababo and P. Ruhinda, have used F. Lukyn
Williams' interpretation of Stanley's route through Ankole, Early
Explorers in Ankole, Uganda Journal, Vol. 2, no. 3. January 1935
pp. 196-208. The original account appears in H. M. Stanley In Darkest
Africa Vol. 2, chapter 32. In addition, the authors have traversed
much of the route and consulted many of the local residents. Amongst
these are two local historians P. Kanyamunyu and L. Kamugungunu.
The standard history of Ankole in Runyankore, Abagabe B'Ankole by
A. G. Katate and L. Kamugungunu, is not referred to but the account
of Stanley's visit in the second volume of this work would seem to
be drawn from Lukyn Williams. The account given here must be
regarded as a supplementation of Lukyn Williams' work and in-
terested scholars should refer back to his article of 1935. The essay
printed here is an attempt to trace the route again in 1964 and adds
a few interesting points of detail.
The coming of the nineteenth century marked a milestone in the march of
African history, for it was a time when Europeans took a keen interest in ex-
ploring what was then known as the 'Dark Continent'. In connection with this
heroic enterprise, let it suffice to mention only a few prominent names: David
Livingstone for his work in central Africa; Mungo Park, Clapperton and the
Landers for their connection with the river Niger; Burton, Speke and Grant
with the source of the Nile; and H. M. Stanley for his work on the Congo basin.
Stanley's last journey which led him through Ankole on his way to the coast
is the concern of this essay.
Between 1870 and 1891, Stanley made four journeys in Africa the last of
which was undertaken to rescue Emin Pasha, the Governor of Equatoria
Province of the Sudan. Emin Pasha was thought to be in danger of being over-
whelmed by the forces of the Mahdi who had recently defeated the British
General Gordon. Just before entering Ankole, the expedition was faced with
a choice of one of the three possible routes:
(1) Through Buganda, Kavirondo and on eastwards to the coast. This
route was dangerous owing to the hostility of Mwanga, the Kabaka
of Buganda, towards the Europeans. It was the shortest of them all.
(2) Through Ankole, via Karagwe, to the coast. Across this route lay
another obstacle; Stanley feared Ntare's forces, who were backed by
*This essay was submitted for the Group Research Project essay competition of the Royal
Commonweath Society in 1964 for teams of senior pupils from Commonwealth schools.
It received special mention.
186 UGANDA JOURNAL
(3) The third route was through Rwanda, and via Lake Tanganyika or
Lake Nyasa to the coast. This route was the longest and the least known.
As the officers were unable to make up their minds, Stanley says the matter
was referred to him to solve. He chose the route through Ankole.
The Kingdom of Ankole at that time was ruled by the Mugabe Ntare,
referred to by Stanley as Antari. His official residence was at Nyakakoni in
the present Kashari county. The Kingdom of Ankole at this time included
Kabula, now in Buganda, and stretched westwards to Mpororo, with no well-
defined boundaries. For the purpose of replenishing livestock, raids were con-
stantly made into Buddu and Koki; the Baganda in their turn would retaliate.
The kingdom was a centralised monarchy with all the power stemming from
the king. He appointed representatives (Bakungu) who ruled on his behalf.
Certain adjoining areas like Buhweju, Buzimba and Igara were semi-independent
having their own Bakama, (petty kings) who paid tribute to the Mugabe.
The Queen Mother, Kiboga, had her residence at Byanamira, a spot situated
not far from the Kashongi turning on the Ibanda road. She had her own
court and wielded considerable influence in the kingdom. She had authority to
declare war in the absence of the king. This is not the case today. With the
advent of modern constitutional government, the centre of political gravity has
shifted and the Queen Mother has disappeared from the political arena. The
King's sister Magwende, officially known as Rubuga, lived in Buhweju. She
had an elevated position and her own court like the Queen Mother. Even today
Rubugais still highly respected and a king without a sister must adopt one
on his accession.
Ntare was one of the fiercest warrior kings (abagabe). During his reign,
his influence extended as far as Buhweju, Kitagwenda, Buzimba, Igara and
Mpororo. Stanley mentions in his account that Ntare paid tribute to Mwanga,
but this is not true. Ntare was as independent and supreme a king as Mwanga.
This is proved further by the 1901 Ankole agreement with the British, which
Ntare entered into without the influence of Buganda.
The route taken by Stanley through Ankole is roughly divided into three
geographical regions. These are: The Kitagwenda-Buhweju region, a hilly
area; the Kashari lowland between Buhweju and Rwampara hills; and the
Rwampara range and the descending lowlands to the Kagera River (Alexandra
Nile). This division is based mainly on relief features. These features have
directly influenced the occupation of the people. In each region, especially
the first two, cattle-grazing was the predominant occupation. The Bahima (a
Hamitic tribe in Ankole) were exclusively herdsmen whereas the Bairu (the
indigenous population) did the cultivation, producing bananas, millet, beans,
potatoes, peas and pumpkins, using locally-made farm implements. The Bairu
settled on the hillsides while the Bahima were a nomadic tribe. They had no
permanent settlements but roamed about in search of fresh pastures for their
herds. Since that time many changes have occurred. The majority of the people
work on the land. Cattle-rearing has shifted to different parts of Ankole where
in Nyabushozi, Isingiro and Kashari cattle are kept mainly by the Bahima, but
cattle-keeping is no longer so predominantly the concern of the Bahima alone.
(a) Kitagwenda-Buhweju highlands. This region is largely an area of
rolling hills which adjoin the eastern end of the western rift valley. From the
floor of the rift valley, the hills rise steeply to over 5,000 feet, being dominated
by the Ntsinda and Kinyamugara. The hills stretch along the route into Buhweju
STANLEY IN ANKOLE 187
where they shoot up to over 6,000 feet. It is a region of relatively heavy rainfall
The vegetation on the hillsides is dense (tall grass with scattered acacia) but
it thins out higher up. The hill mass is dissected by deep valleys filled with
(b) Kashari lowland. This is more or less a plain, the general level being
about 4,500 feet, with low undulating hills rising to 5,000 feet. Large anthills
are a common feature of the area and at the time of Stanley's passing through
the Kingdom, kraals would be scattered everywhere. In the dry season the old
bush is burnt so that new, rich grass can grow. As the area has a relatively
small rainfall, the vegetation is of short grass with scattered, stunted trees.
The Ruizi River with its many tributaries drains this plain, meandering slowly
through it in great loops. Kashari was, and still is, a cattle rearing area. Milk
was the staple food.
(c) Rwampara-Isingiro area. This area rises to over 6,000 feet. It is inter-
rupted by wide, bowl-like valleys such as Mamiyanja which drops about 400
feet below the general level and Mabona Valley which widens out into what
Stanley calls the Alexandra Valley (i.e. the Kagera Basin). The grass is short
and acacia trees predominate. The area near the Kagera would be ideal for
cattle were it not for the tsetse fly which harasses them. The population was
sparse and cattle-keeping was their chief occupation.'
At the beginning of July in 1889, Stanley's expedition first entered Ankole.
On the 4th of that month the expedition left the shores of Lake George2 and
followed the track of the salt-traders. The salt was extracted from Lake Katwe
in Toro and as there was no means of quick transport, the traders carried it
on their heads. The gradual ascent from the floor of the rift valley led the party
to Kitete,3 a place about 1,000 feet above lake-level (Lake George being at
2,997 feet above sea level). This was the first settlement Stanley arrived at.
At the sight of these strange people, the settlers were frightened and quickly
ran to inform the local chief, (Mukungu). When Stanley saw the frightened
people making a hullaballoo and dashing out of their houses, he thought they
were leaving their establishments to give him accommodation. He misinter-
preted the 'reception', for the settlers were scared by his presence. That is why
he quotes-"Room, for the guests of Antari! Room, for the friends of Masa-
juma! Ha! villains, don't you hear? Out with you, bag and baggage." Lukyn
Williams in his 'Early Explorers in Ankole', claims that Stanley sent messengers
to Ntare on his arrival in Ankole. From what we have found out, the fleeing
villagers spread the news of his arrival until it reached the Queen Mother,
As the expedition moved through the deserted village, an unusual incident
occurred. It was about to rain. Stanley and his men entered a hut to take shelter.
The occupants had run away leaving a 5- or 6-year-old boy. This is the only
man we have seen who saw Stanley. His name is Yakobo Kakugasha who now
lives in Kihani village in Mitoma county. When Stanley entered the evacuated
house, the boy was asleep. He instantly awoke and Stanley held him on his lap.
Finding that the boy had been left behind, the father, Kajogi, returned brand-
ishing his spear in the air. But his fury subsided as he caught sight of the Euro-
pean. Then a dialogue followed:
"Is this your house?" Stanley questioned Kajogi.
"Yes, it is mine," was the reply.
ANKOLE DISTRICT, UGANDA
Stn uoral. a NMobf
STANLEY IN ANKOLE 189
"Is this child yours ?" Stanley inquired further.
"Yes" Kajogi answered in a monosyllable.
"Have you any other children?"
"Yes," came the reply.
"Go and bring them and your wife, then I shall give you this child. I don't
see why you are running away from me. I am harmless."
This information we got from Kamugasha, a child then, but now an old
man in his eighties. When Kajogi brought back his family, Stanley tore two
yards of cloth from a roll and gave it to the child seated on his lap, as a sign
of friendship. Kamugasha says that besides Stanley there was another white
man called 'Grace'. Stanley's retinue included Baganda, and Nubians also.4
They brought with them brown sheep two of which were slaughtered that even-
ing. Stanley sent Kamugasha a piece of the meat.
The expedition left the following day, but some days later the party appeared
again. The reasons for their return are obscure. They now camped at a place
nearby, Kahenda,5 where a chapel has been built. Kamugasha has been a
catechist in this area.
The next day, the party took their last leave of the village. They began a
steady ascent from Kitete to Kibwiga, then continued in4 a valley between
Kinyamugara and the Denny Range (Marangara). Eventually the expedition
reached the summit of Kinyamugara, over 6,000 feet above sea-level. From there
it proceeded to the house of the Mukungu, a local chief, whom Stanley called
Masakuma. Stanley refers to him as the governor of the Lake Province. The
chief was eager to know how Stanley had overcome the Banyoro. In his own
account, Stanley says he saw women wearing "bead-worked caps, and beaded
tassles." The beads referred to, had been brought into Ankole by Arab traders
who gave them to local chiefs in exchange for slaves. It is said that the first
Arab, Sri Bin Amin, nicknamed "Kiyengo", came to Ankole during Muta-
mbuka's reign in 1852.6 At the chief's residence Stanley was regarded as super-
human (Omuchwezi)7 because he had defeated Kabarega, the King of Bunyoro-
Messengers were sent to the Queen Mother Kiboga, and the king, Ntare.
News was brought back from the Queen Mother that she was indisposed and
that the king was away on a raiding expedition. Under these circumstances,
there was nobody to welcome them at the court. Nevertheless, food would be
provided. Masakuma was ordered to provide them with a guide to lead them
through the kingdom. If they came back the king would be glad to receive them.
From what we have found out, this was merely a manoeuvre to keep Stanley
away from the capital because he was suspected of evil designs. The king was
present and the Queen Mother was in perfect health. There was also a super-
stition that if two kings met, one of them would die or suffer some serious
disaster. Ntare might have regarded Stanley as a man of his own rank; this
was one of the reasons why he did not want to meet him in person.
The expedition then proceeded to Kitara, a valley in the present Buhweju
county which is still known by that name. The village of Kitega, mentioned by
Stanley, is still to be found at the foot of the path leading into the valley. Here
the expedition camped. It was also here that deputations of Baganda Christians,
who had taken refuge in Ankole, met Stanley and told him about the religious
strife in Buganda which arose from the fighting between Moslems and Christians.
The expedition marched on, descending into the present Kishari county
and camped at Wamaganga (Rwamaganga forest). This is what Stanley calls
190 UGANDA JOURNAL
the Ruizi Basin, but is really the upper course of the River Koga, a tributary
of the present-day Ruizi.8 Stanley found the area inhabited by the Bahima
herdsmen whom he called Watusi, and Bairu agriculturists whom he called
Wanyankori. On the next day, Stanley and his party marched along the Koga
which spreads out into a papyrus swamp about a mile wide. Although it was a
dry season, the crossing must have been difficult since twenty-four head of
cattle were lost. The next camping station was the present village of Kashaka,
a place which Stanley calls Kasari, in the plain of Kashari. During his stay here,
he received presents from the king and the Queen Mother. These included
seven head of cattle and a tusk of ivory. The idea of blood-brotherhood,
concluded later, was initiated here. It strikes one that Stanley does not mention
crossing the Ruizi proper, which is a river about four yards wide. .At this time
it was presumably narrower, and, as it was a dry season, the crossing presented
no serious difficulty. The river was crossed at Kyempene, a mile south-west of
Mbarara Stock Farm.
The distance to the next camp at Nyamatoso (Nyamatojo) is about eleven
miles. Not much was said about the lowland by Stanley because the area is
flat land; it presented little or no difficulty. The area on the other side of the
river had a populous settlement of cultivators. The predominant crop was the
banana. Even today bananas are the staple food of the inhabitants. The swamp
referred to by one member of the expedition in this area is Ruchenche. By this
time the party members had grown so much used to the people of Ankole
that they could go individually to ask for their help. Despite the warm receptions
given them, they plodded on, following a path between two fairly high hills,
Ibare9 on the right and Rwakwezi on the left. A branch of the Ruchenche
swamp is between these two hills. Today a road from Mbarara to Mwizi, in
the Rwampara range, traces the route at this point. Exhausted as they were,
the expedition halted at Rusussu (Kasusu) which is identified by Lukyn Williams
as Kashozwa, and camped nearby at Nyamabare in the Mamiyanja Valley.10
They spent three days here to recuperate from fever attacks. It is to be noted
that from Nyamatojo to this camp, Stanley invariably kept travelling on
mountain tops. Feeling strong enough to resume the journey, the expedition
set out again from Nyamabare. Before they reached Viaruha (Byaruha) Stanley
says that they met "herd after herd of the finest and fattest cattle". It appears
that the cattle at this time were much fatter than they are today. There were
larger pastures and in addition the tsetse fly was then unknown. Moreover,
subsequent invasions of rinderpest weakened the breed. The party was waylaid
by a group of spearmen. Fortunately these were dispersed by a few well-directed
shots. Close behind the attackers was Prince Buchunku (Uchunku), a young
man about fourteen years old, the cousin of Ntare whom Stanley, mistakenly,
called his son. He had been sent to effect the blood-brotherhood with Stanley.
Ntare had delayed this ceremony until Stanley was nearly on the border because
he wanted to make sure that the European was harmless to the Omugabe
himself, and his subjects. The area where the ceremony took place is called
Byaruha but the actual village is called Nyabugando". Among the witnesses
of the performance were Byeziga from Kahwenda (Mitoma county) and the
Rwamuniogo, who died in the 1930s. The procedure differed from the traditional
pattern in that blood was not swallowed. Normally there were two ways of
effecting this blood-brotherhood.12 One consisted in making an incision near
the umbilicus of the two people concerned, each licking the other's blood
and swallowing it. The second method was that blood from each person was
STANLEY IN ANKOLE 191
smeared on two seeds of coffee from the same berry, each of them swallowing
the other's blood. The ceremony normally took place in the morning before
meals. In Stanley's case an incision was made on both people's left arms;
the blood mixed with a portion of butter and the mixture smeared on thire
foreheads. The customary method was deliberately disregarded by Buchunku
to avoid swallowing a stranger's blood. It was also rare for a Muhinda (prince)
to mix his blood with that of a commoner. The ceremony established perpetual
friendship and as a result of this blood-brotherhood, therefore, Ntare could
now be sure that the white man and his friends would not harm him and his
people. An exchange of presents crowned the ceremony. In this area we were
told a story that Stanley produced fire from his shin. What actually happened
was simple. The people had never seen a box of matches and Stanley struck
a match near his leg to prevent the wind blowing it out. We were told that
Stanley gave the Union Jack to Buchunku to take to Ntare.
On the following day, the expedition made a steady descent from the hills
into the fertile Mavona (Mabona) plaint3, which joins the Kagera basin. Byeziga,
a courtier who had witnessed the ceremony, led the expedition to the Alexandra
Nile (Kagera). We were not able to locate the actual place where the party
crossed the Kagera. The river has widened and the banks have steepened in
the course of time. No people live in the area, who could supply information.
The distance from Mabona to the point where they met the river is about twelve
miles. From Stanley's account, it appears that they walked along the river,
downstream, in search of a suitable ford. They finally crossed at a point about
one and a half miles downstream from Nsongezi. We could not ascertain whether
there was a ferry service. It is probable that the boats they crossed in may have
been used by the local people to cross to Karagwe.
Although originally it was not Stanley's intention to explore Ankole, his
journey increased knowledge of the area. It preceded visits of many other
Europeans, culminating in the 1901 Agreement which incorporated Ankole
into the Uganda Protectorate.
1. The population of parts of Ankole in earlier times is open to question; quite probably
the area was more populated at the time that Stanley passed through than it became later.
On the tops of now deserted hills pieces of broken pottery are commonly found in earth
dug up by animals, and ploughings at new ranching schemes, in Nyabushozi are also rich in
broken pottery. (EDs. from information supplied by Mr. H. Plummer, Geologist, Ankole.)
2. Stanley did not recognize the Kazinga Channel and thought that Lake George and
Lake Edward were one lake which he named Albert Edward Nyanza after the Prince of Wales,
afterwards King Edward VII.
3. In this account names recorded by Stanley are given in italics in their original spelling
4. Stanley had five Europeans with him apart from Emin Pasha and his companions.
People who gave us information like Yakobo Kamugasha, (the only person now living who
saw Stanley) says that Stanley had only one European in his retinue but it is probable that
he did not see part of the expedition. Stanley does not mention having Baganda with him,
but we think he must have had one or two who acted as interpreters.
5. The name Kahenda does not appear in Stanley or Lukyn Williams. It is marked
on the 1:50,000 Map (76/II Ibanda, Sheet Series Y 732) as a small settlement, without a chapel,
on the northwestern edge of the North Ankole Forest in Mitoma county. The hill named
Kinyamugara is on the north side of the Mitoma-Kabale border six miles east of Ibanda.
Kitete and Kibwiga do not appear to be in current use. Kitega is still mapped as a small
village 4 miles northeast of Nsika (Sheet 76/IV Nsika, Sheet Series Y 732). [EDs]
6. Sri bin Amin (who would be the Arab Snay bin Amir well-known to both Burton and
Speke as having reached Buganda in 1852) and Kiyengo are probably two distinct persons.
Both Speke and Grant have numerous references to "Dr." K'yengo, a much-travelled Munya-
192 UGANDA JOURNAL
mwezi trader and medicine man. He had visited the Coast and when Speke and Grant reached
Karagwe was acting as Rumanika's 'head magician'. He accompanied Grant from Karagwe
to Mutesa's in April-May 1862 carrying with him gifts from Rumanika to Mutesa. Speke
and Grant found a party of his men installed at Kamurasi's when they reached Bunyoro.
He was a talkative old man when Stanley met him at Kafuro in August 1889. It may well
be that either, or both, Snay and Kiyengo had contacts with Ankole from Karagwe as early
as 1852. [EDS]
7. The Bachwezi were a legendary race who ruled Ankole before the Bahinda dynasty.
They were believed to have supernatural powers.
8. Ruizi is a local name for a river. It means nothing more than water.
9. The place is called Ibare because there are many rocks and stones exposed. Ibare
means "stone" in the vernacular.
10. The names Kashaka, Kyempene, Nyamatojo, (- -- zho) can easily be found on the
current 1:50,000 maps but Wamaganga can not be traced though it must have been in use in
1935. Kashozwa, Nyamabare and Namiyanja are all still in use but are less easy to locate.
The first two survive as tiny settlements on the Rwampara-Isingiro boundary to the west of
the Rugaga River (1:50,000 Map 86/III Chezho Valley, Sheet Series Y 732) [EDs]
11. A monument commemorating this event stands at mile 15 on the Mbarara-Kikagati
road. This reference to Nyabugando is new since neither Lukyn Williams nor Katete and
Kamugungunu refer to it. Byaruha is marked in the extreme northwest corner of map 86/III
as a tiny settlement. Nyabugando is on the main road Mbarara to Kikagati. Byaruha is
usually regarded as the location of the ceremony. [EDS]
12. For a fuller account see F. Lukyn Williams, Blood-brotherhood in Ankole, Uganda J.,
2, 1934, pp. 33-41 and a recorded account by Y. K. Bamunoba, Uganda J., 28, 1964, pp. 217-218
13. This is the present Oruchinga Valley. The Batutsi refugees, who recently fled from
their country, Rwanda, following the political upheavals, have been settled here temporarily
by the Uganda Government.
Uganda Journal, 29, 2, (1965) pp. 193-199
THE EDUCATION OF AFRICAN MUSLIMS IN UGANDA
by FELICE CARTER
Formal 'Western' education in Uganda began as an intrinisic part of
missionary evangelism. It was necessary that potential converts, who became
known as 'Readers' should be able to read the Bible and Prayer Book, and their
instruction was extended automatically by the European missionaries to include
writing and arithmetic. The expansion of the Christian missions led to the
creation of a network of catechistical centres or "bush schools." The need for
an educated African clergy and the need to supply Government with clerks
and interpreters led to the opening of more advanced schools where English
and a little history and geography were included in the curriculum. Even after
the establishment of a Government Education Department in Uganda in 1925,
education was still in the hands of the Missions. Though aided by Government
and meeting a secular need, the missionaries continued to assert that their
educational activities were essential to their spiritual work, for they gained
many converts from among their pupils and ensured that the African leaders
of the future would have received a Christian education.
Although the Christian community in Uganda remained a fairly small
proportion of the total population their schools were also open to pagans, for
whose conversion they had originally been intended. The avowedly evangelistic
aims of the mission schools deterred the majority of Muslims from exposing
their children to mission proseletisation. From the last two decades of the
nineteenth century the Muslims were a small but recognized minority and
throughout the twentieth century their numbers have increased. Arab traders
first reached Buganda in 1844, and although expelled for a brief period thirty
years later, they wielded, and continued to wield some influence at the Kabaka's
court. Mutesa's public debates as to the relative value of Christianity and Islam
meant that the latter became both known and politically respectable. Some of
the Baganda chiefs became Muslims, and Arab influence had led to the adoption
of Swahili and Arab dress by many of the chiefly class. More converts were
gained when Sudanese soldiers were introduced into Uganda by the Imperial
British East Africa Company, and were later employed by the Protectorate
Government. A further influence favouring conversion came from the Coast
Swahili employed by the Government and as private servants. These were all
men enjoying a special relationship with Europeans and their social prestige
gave status to the religion. Moreover Islam was a less demanding religion than
Christianity. The convert might retain his wives and many of this traditional
beliefs without fear of excommunication and without undergoing the spiritual
and cultural crisis which must have accompanied conversion to Christianity.
The numbers of adherents grew. By 1911 there were 58,000 Muslims and twenty
years later there were 122,000 as compared with 800,000 Christians out of a
total population of 3,535,000.1
194 UGANDA JOURNAL
Muslim political influence reached its peak in 1888 when the Muslim faction
established their protegee Kiwewa as Kabaka; shortly to be succeeded by the
more malleable Kalema. But the Muslim party was defeated by a combination
of the Roman Catholic and Protestants in 1890 and never regained political
control as the rising of 1893 and 1897 both proved abortive. However, under
the leadership of Mbogo, an uncle of Mutesa, the Muslims enjoyed a certain
status. Mbogo was party to the agreement which guaranteed him an annual
salary in return for his renunciation of all claims to the Kabakaship. In 1892
when Lugard imposed a land settlement on the Baganda the Muslims were
granted the three counties of Gomba, Busujju and Butambala, but in 1900
they were left only with Butambala. Thus of the chieftainships allotted in 1900
the Muslims gained one as against eleven which went to the Protestants and eight
to the Roman Catholics.
During the life of Mgobo the Muslims represented a united and by no means
negligible minority, but unlike the Christians they made no attempt to establish
schools and so prepare their children for significant role in the new social
order. In 1918 there were only traditional Koranic schools where Arabic and
Islamic Law were taught and by 1939 there was only one full Primary School
of a recognized standard. By 1944 of the 134,000 spent on Education only
213 represented grants in aid to Muslim schools.2 The educational backward-
ness of the community cannot be explained in terms of their number in relation
to the total population of the Protectorate, for although they represented only
2% in 1911 and 3.6% in 1931, nevertheless they were concentrated in certain
areas of Uganda. Nearly half of all the Uganda Muslims lived in Buganda
where they represented 15% of the Baganda population in 1911; and though
there were Muslims in every county, the majority lived in Butambala. There
are other concentrations of Muslims in Busoga and in West Nile where in Aringa
county 80% of the population were Muslims by 1962.3
Whereas for the Christians, European education was a means of assimilation
into another culture of which their religion was a part, for the Muslims Western
education represented an alien influence. Conservatives urged that a Koranic
education was necessary if Islam were to survive and claimed that the traditional
system represented the supreme intellectual activity. The Waalim, the religious
preachers, gave no encouragement towards the support of secular schools.4
Muslim apathy towards education may also have been caused by a realisa-
tion that the power structure in Buganda, and to a lesser extent in the other
provinces, was in the hands of the Christians. In Buganda the Kabaka and all
but one of the saza chiefs were Christians, predonimantly Protestants, and
they owned the largest areas of land. Thus they enjoyed both wealth and political
power and were able to appoint their supporters to office within the Native
Administrations and to ensure that by educating their children at mission schools
their off-spring would be the most eligible candidates for Government positions.
In Ankole, Toro and Bunyoro the kings and leading chiefs were Christians
and land owners. In the other areas of the Protectorate the Government had
originally appointed Baganda as chiefs and continued to appoint where possible
educated men, and educated inevitably meant ex-mission boys. The Muslims
on the other hand remained in ignorance of the advantages which education
provided. Musoke, himself a Ugandan Muslim teaching at Tabora in Tanga-
nyika, thought in 1939 that the Mohammedans were "ignorant of the benefits
education could bring".5
MUSLIM EDUCATION IN UGANDA 195
The Muslims lacked both organisation and unity after the death of Mbogo
in 1921. The community split into two factions, ostensibly because of a religious
dispute concerning the noon prayer, but the division probably represented
a struggle for power between rival leaders.6 The Kibuli sect recognized the
leadership of Prince Badrui, Mbogo's son and appointed successor, and the
Butambala sect were led by Sheik Sekimwanyi. Inspite of the Government's
attempts to find a man of sufficient religious authority to arbitrate on the
dispute, the two factions were not united until 1945, and even later minorities
from each group retained their separate identity. Antagonism was such that the
sects attempted to close each others' mosques, and each provided their own
Koranic schools.7 Their mutual hostility obviated the possibility of a united
effort to raise the standard of their schools. Also there was no administrative
organisation through which the Muslims could administer an educational
system and what secretarial work was necessary was provided by the Education
The development of Muslim education was further hampered by the hostility
of the Christian missions. Islam represented a rival faith which it was feared
might eventually predominate in East Africa. Writing in the local Church
Missionary Society journal in 1904, Rev. Willis, who was later to become the
Anglican Bishop, thought it more than probable that in a few years Uganda
will be Christian or Mohammedan"8. Three years later it was urged that there
should be more mission work in the Eastern Province to counteract the spread
of Islam9. In 1930 mission opposition to the introduction of Swahili as lingua
franca and as a medium of instruction in the primary schools in the mixed
language areas was based partly on the fear that an increase in the use of
Swahili would result in an extension of Muslim influence.to The Apostolic
Delegate to the Roman Catholic Missions in Africa was quoted as saying that,
"the opportunity for the Church must be seized at once or others will seize it,
materialism, Moslemism or other wretched isms."" The missions strongly
opposed any attempts made by the Government to provide for the Muslims
the schools which they could or would not provide for themselves, for when
in 1905 George Wilson, then Acting Governor suggested that a school might
be established by the Government for the Muslims, Bishop Tucker immediately
protested to both the Uganda and British Governments, arguing that such a
measure would commit the Government to a system of state schools and that
any Government contribution to education in the Protectorate should be spent
on mission schools already in existence. In 1924 Tucker's successor, Willis,
similarly opposed the education plan drawn up by the recently appointed
Director of Education, Eric Hussey, which advocated the creation of six ele-
mentary and one Intermediate Government school. He thought a dual system
of Government and mission schools unnecessary, that it was unreasonable
that the former should be established for the benefit of the Mohammedans as
they had done nothing for education as a community and should not therefore
receive preferential treatment. Moreover the provision by Government of
schools for Muslims would be a breach of Government religious neutrality.
Protesting against the establishment of a Government Middle School, Canon
Grace asked whether there were no Mohammedan missionaries who could
build schools and pointed out that the Director of Education had anyway
agreed that the Muslims were a very unimportant sector of the community2
As the sole providers of any form of education before 1921 and as virtual
monopolists of the field (though admittedly with government aid) until after
196 UGANDA JOURNAL
the second World War, the Christian missions wielded much influence. Their
combined opposition on the Advisory Council for African Education led to an
alteration in Hussey's plan, from a proposed six Government elementary
schools to one. Moreover, neither the Government elementary nor secondary
schools provided in the inter-war period lasted for more than three years.
Fears of Muslim expansion were not confined to mission circles. The
Assistant Secretary for Native Affairs minuted in 1923 that the spread of
Christianity should be encouraged "to develop this Protectorate as a bulwark
against Mohammedan influence coming from the North." He was opposed
to the provision of European education for Mohammedans and felt that it
was by no means undesirable that Mohammedans generally should be looked
down upon owing to their lack of education.14 The Chief Secretary agreed
that since it was the policy of Government to aid mission schools it would be
better not to subsidise Mohammedan efforts.15
However, this view was never recognized Government policy. Wilson in
1905 had doubted the wisdom of leaving education in the hands of the missions
and attacked the application of the theory of self-help in relation to the Muslim
Community. But there was then no money for the establishment of Government
school, and Ellis an official of the Colonian Office suggested that such schemes
migh wait until Uganda were self-supporting." The Governmet's policy
towards the establishment of mission churches and schools was laid down in a
circular of 1917. It stated that although Government wanted to encourage the
spread of Christian principles, equal toleration should be extended to all reli-
gions provided their teaching was not seditious, and Government would give
reasoble encouragement to all genuine ir s tutions of an educational charac-
ter. Mohammedanism was undoubtedly entitled to the same toleration as
Christianity but since it was "peculiarly liable to fanatic development it should
be judiciosly wathed".'7
In 1923 the Government's attention was directed towards Mohammedan
education when Colonel Hill, the Commanding Officer of the troops in Uganda,
asked for aid in the provision of a school at Bombo for the sons of serving and
discharged soldiers most of whom were Sudanese. The soldiers wanted their
children educated beyond the existing Koranic school and he urged that the
fact that they were Mohammedans should not be a bar to their advancement.18
His request, which was finally met by the establishment at Bombo of a Govern-
ment school, raised the question of Government policy in relation to Moham-
medan education. The Governor, Sir Geoffrey Archer, inspite of dissentient
voices from his administrative staff, re-affirmed the official principle of religious
neutrality. He thought that the Mohammedan population should not be debar-
red from educational facilities because they were Muslims, and that since there
was a considerable Muslim element living in Uganda the Government was not
justified in leaving education in the hands of the missions.19 Two years later
hi successor, Gowers recorded his complete agreement with Archer's opinion.
Although as has been seen, Hussey modified his original plan for Govern-
ment schools in the face of mission opposition his concern for the Mohammedan
community remained. He hoped that the Kadi who was to be appointed as
arbitrator between the two Mohammedan communities could also act as
instructor in the Government Normal School, as he wanted trained Moham-
medans on the staff of the Government elementary and intermediate schools.
When after two years a man had not been found the Assistant Secretary for
Native Affairs pointed out "as far as the Director of Education is concerned
MUSLIM EDUCATION IN UGANDA 197
the need for this man is as great as ever," for it was hoped in that year, 1927,
that the Government Intermediate School might open and if the Kadi could
teach religion, more Mohammedans might enrol. Apart from the Intermediate
School, which eventually opened in 1930 and closed again within twelve months,
there were other government institutions from which the Mohammedans
might have benefitted. A Normal School was opened in 1928 in the hope that a
supply of trained teachers would raise the standard of the Mohammedan
schools. In addition a centre was opened at Kasawo to train teachers, for the
sub-grade schools in 1935 and its standard was raised to train for the full
primary course in 1939; but this was taken over by the Government in 1944.
The apathy of the Muslim population is illustrated by their failure to utilise
these facilities. The Normal School accommodated sixty students each year,
of whom, on average, only three had been Muslims, and enrolment of Muslims
at the elementary and intermediate schools had not justified their continuance.
The Muslims were slow in raising the standard of their elementary schools and
indeed even in founding them. Elementary education lay within the jurisdiction
of the District Education Boards and the Boards aided schools which had
attained a recognized standard. It was their function to help existing schools
rather than to establish new ones, but the Muslims lacked interest in their
elementary schools. Often they were closed within a few years of opening owing
to lack of support. This happened in Buganda in the late 1920's after the
impetus given to Mohammedan education by a Lukiko grant.21 In 1935 four
of eighteen registered schools closed in the one year, and in 1936 Arua and Gulu
also lost their schools.22 In a report on Mohammedan education, Ousley, a
member of the Education Department commented, "This community is so
apathetic regarding the educational needs of its children that its members
consistently fail to demonstrate by even moderate self help that they have an
obligation towards them."23 Earlier Lt.-Col. Hill had urged support for his
school at Bombo on the grounds that Mohammedan interest in education was
so rare that it should be encouraged.24
What development there was, was due largely to stimulus from outside the
community. Much depended on the interest of the local administrative officer.
In 1936, the District Commissioner at Fort Portal persuaded the Mohammedans
through their local leader, Sheikh Aziyi Abdullah to raise 300/- for their school,
which was rewarded by a grant of a further 300/- from the local Education
Board. Similarly the District Commissioner at Masindi did much to encourage
local effort.25 The Masindi school was taken over by the Government in 1939
to prevent its deteriorating through bad administration.26 The Buganda Govern-
ment, dominated in the inter-war period by Protestant chiefs, was also concerned
at the lack of Muslim education. Kagwa had urged, from the early days of the
Protectiorate, that the Muslims should send their children to Mission schools
and Kn 1922 the Lukiko erected two schools in Buganda, one at Kibuli and on
at Gabasanda, from Lukiko funds. The Buganda Government also financed
the C Crade Teacher Training Course at Kasawo whilst the Education Depart-
ment continued to provide the administrative machinery which the Muslims
After 1940 the situation improved. There was already one full primard
school of six classes for Mohammedans, (one of four classes) and seven selecteds
sub-grade schools" receiving grants from the local boards. Most of the staff
were Mission educated men, and the general standard of the Muslim schools wae
still very low. However the passing of the Colonial Development and Welfarr
198 UGANDA JOURNAL
Act by the British Parliament, which provided for the allocation of funds for
welfare projects in the depndencies, enabled an expansion of educational effort
in Uganda. Government Normal Schools were established at which Muslim
teachers could be trained and a few students were sent for training at the inter-
territorial centre in Zanzibar. The report of the Thomas Committee, set up in
1940 to consider educational development in Uganda provided for Moham-
medan representation on the reconstructed local education committees in
areas where they were sufficiently numerous to warrant such representation.
At the end of the war there were twenty six aided Mohammedan schools, two
of them offering a six year course, and nine a four year course. 1945 also, sow
the inauguration of the first year of secondary education at Kibuli which was
partly financed by the Mohammedan Welfare Association of the Ismailia
community.27 The Ismailia community had been encouraged by the Aga Khan
to develop social welfare services, and their African co-religionists also benefit-
The Mohammedans of Uganda have remained an unprogressive and educa-
tionally backward community. A report on Muslim education in East Africa
made in 1958 by Dr. Sergeant, Professor of Arabic at the University of London,
and by an ex-Sudan Civil Servant, found that the Mohammedan community
was so backward, economically and educationally, that it needed special help.
A conference was held a year later in Dar es Salaam to consider the problem,
but no action has been taken of its findings.28 The Uganda Department of
Education Annual Report of 1957 pointed to the "regretably low standard of
Mohammedan education". In 1960, only two years before Uganda attained
her Independence, there were only 905 children in Mohammedan junior sec-
ondary schools, and none in Mohammedan senior secondary.29 Christians
have continued to dominate political scene.30
1. Uganda, Report on the census of the population of Uganda Protectorate, 1931.
2. Uganda, Legislative Council Debates, 18th December, 1944. (Jaffer).
3. J. S. Trimingham, Islam in East Africa, Oxford, Clarendon Press, 1963, p. 47.
4. M. Musoke, Mohammedan Education, Uganda Teachers Journal, 1, no. 4, October, 1939.
5. M. Musoke, op. cit.
6. Entebbe Archives, S.M.P., 6900, Gowers to Secretary of State for the Colonies, 5th
7. T. W. Gee, A century of Mohammedan influence in Buganda, 1852-1951, Uganda J.,
22, no. 2, September, 1958.
8. J. Willis, Uganda Notes, February, 1905.
9. Rev. Crabtree, Uganda Notes, February, 1907.
10. Text of the bishop's memorandum on language, reported in Uganda Herald 22nd April,
11. Extract from The Tablet, 18th January 1930 in Entebbe Archives, S.M.P. third series, 6875.
12. The protests of Bishop Willis were conveyed to the British Government by way of
Joseph Oldham, the representative of the Protestant missions on the Colonial Office
Advisory Committee for Education in Tropical Africa; see Oldham Papers, Edinburgh
House, Willis to Oldham, 21st April 1924.
13. Oldham Papers, Edinburgh House, Grace to Oldham, February, 1930.
14. Entebbe Archives, S.M.P. second series 7898, E. L. Scott to Chief Secretary, 27th July,
15. Entebbe Archives, S.M.P., second series 7898, Jarvis to the Governor, 31st July, 1923.
16. P.R.O. 536/6 Correspondence, November, 1095.
17. Entebbe Archives, S.M.P. second series, 6900.
18. Entebbe Archives, S.M.P., second series, 7898, Lt.-Col. Hill to Chief Secretary, 6th
MUSLIM EDUCATION IN UGANDA 199
19. Entebbe Archives, S.M.P. second series, 7898, Governor to Chief Secretary, 30th August,
20. Uganda, Department of Education, Annual Report 1931.
21. M. Musoke, op.cit.
22. Uganda, Department of Education, Annual Reports, 1935 and 1936.
23. Uganda, Department of Education, Annual Report, 1935.
24. Entebbe Archives, S.M.P. second series 7898, Lt.-Col. Hill to Chief Secretary, 6th
25. Uganda, Department of Education, Annual Report, 1935.
26. Entebbe Archives, S.M.P. S 94, Correspondence.
27. There had been little contact between the Indian and African Muslims before 1945.
Twenty years earlier the Katambala group had made contact with the Indian com-
munity in the hope that they might procure a teacher from India. The Government
had been alarmed by the possible results of such a move and was equally alarmed
when, after no teacher had arrived, African Muslims began to attend an Indian School.
Entebbe Archives, S.M.P. second series, Governor to Secretary of State for the Colonies
5th October, 1925.
28. Ahmed Abdullah, Ambivalence of African Muslim Education, East African Journal,
29. Though there were a few Muslims in Mission and Government senior secondary schools.
30. L. A. Fallers in The King's Men, O.U.P., 1964, p. 8 writes. "The leading Buganda Govern-
ment officials, the leaders of the most important political parties, are all overwhelmingly
'old-boys' of the two elite boarding schools, one Anglican and the other Roman
The author wishes to acknowledge the kind permission of the authorities of the Uganda
Government to consult official archives.
Uganda Journal 29, 2 (1965) 201-214
THE DIARIES OF EMIN PASHA-EXTRACTS X
Edited by SIR JOHN GRAY
(The extracts from Die Tagebiicher von Dr. Emin Pascha, edited by Dr.
Franz Stuhlmann, vols. i, ii, iii, iv and vi, (Braunschweig: Westermann, 1961-
27), have been translated and provided with introductory notes and comments
by Sir John Gray. They have been planned to appear in The Uganda Journal as
a series covering Emin's first visit to Buganda in 1876, his visit to Bunyoro in
1877 and his second visit to Buganda in the same year, followed by such portions
of his later diaries as are relevant to Emin's contacts with the Uganda region
during the years spent as Governor of Equatoria until his withdrawal in 1889.
Extracts I to IX have appeared in successive issues of The Uganda Journal
commencing with Extracts I in Uganda J., 25 (1961), 1.-EDS.)
X FIRST MEETING OF EMIN WITH STANLEY
25 April-26 May 1888
In the period covered by these present extracts Emin first made contact with
the members of the Emin Pasha Relief Expedition. Reference to Donald H.
Simpson's Bibliography of Emin Pasha Uganda J. 24, (1960) pp. 138-64), shows
how copious the literature on the subject has become. It would be impossible to
reproduce even a tithe of the works cited therein for purposes of comparison
with these Extracts and no attempt will be made to do so. All that can be done is
to imitate Franz Stuhlmann in his editorship of the Tagebiicher and to set out
cross references to the more important of these works.
(a) Stanley H. M. In Darkest Africa (1890), vol. i, chapters XIV to XVI.
(b) Schweitzer, G. (translated by Felkin), Emin Pasha (1898) vol. i
(c) Mounteney Jephson, A. J. Emin Pasha and the Rebellion at the Equator
(d) Parke, T. H. My Personal Experiences in Equatorial Africa (1891).
(e) Casati, G. Ten Years in Equatoria (1891).
(f) Vita Hassan, Die Wharheit uber Emin Pascha, (1895).
With regard to (a) Stanley kept a diary during the relevant. period-very
probably in shortand-and has reproduced in In Darkest Africa what are said
to be copious extracts therefrom. How far these entries were in fact made at the
time and how far embellished two years later for the purposes of Une pi&ce
justicative is it impossible to say. Even if one assumes that there has been an
extensive embellishment, Stanley's version of the incidents described by Emin
in his diary requires to be carefully studied.
202 UGANDA JOURNAL
(b) is a very disappointing piece of work. It purports to be a translation of
Georg Schweitzer's Emin Pascha, which is very sketchy as regards this period
and has been written with an undisguisedly anti-Stanley and anglophobian bias.
Moreover, Felkin's translation is at times faulty and in places far too free.
As to (c) Jephson kept a diary during this particular period, but it has to be
borne in mind that his book was compiled, as the title page shows, "with the
Revision and Co-operation of Henry M. Stanley", which must account in some
degree for its anti-Emin bias.
Parke, the author of (d) also kept a diary and quotes numerous extracts
in his book, which does not appear to have been submitted to Stanley for
revision or co-operation.
Casati's references to this period are meagre and it has to be remembered
that he and Emin were not on good terms.
Vita Hassans's references in (e) are also very meagre.
Much of the correspondence in Appendix A to these extracts has never
previously been in print except in Stuhlmann's Vol. iv of the Tagebucher.
Parts of it show that there was, as H. R. F. Bourne alleges in the title to his
book, The Other Side of the Emin Pasha Relief Expedition (1891).
Extracts from Emin's Diaries
April 23, 1888, Monday (Emin was at Tunguru). A non-commissioned
officer from Mswa presented himself with a letter bearing the inscription "good
news". The non-commissioned officer had the difficult and rocky way from
Mswa in the astonishingly short time of ten hours. the letter read as follows:
"Just now one of Mogi's men has reached here with the news that Stanley
has reached Mugango in Viarifu and will be coming today by water to Mswa.
Please therefore come as quickly as you can. The messenger has seen the stranger
with his own eyes."
I have at once sent an express messenger to Wadelai to order the steamer.
There is no guide here. Otherwise I would willingly go overland.
At eight o'clock a messenger has come from Mswa with another letter.
April 24, Tuesday. Nobody has slept, and today is full of rejoicing.
I have sent officer by land to Mawfi (with a letter for Mr. Jephson).
((The letter is set out in Appendix A) i.
April 25, Wednesday. Hoped that the steamer would arrive today at noon.
I will try to inform Mr. (E. C.) Gordon (of the C.M.S.) in Uganda of Stanley's
April 26, Thursday. 'El intisar eshidd al en-nar', as the Arabs say (Waiting
is worse than fire), and I have never realized that so clearly as now, when I am
waiting for the steamer and it does not come.
Towards evening I had the post from Mswa. Suleiman Effendi reports
his arrival, and that Mr. Jephson is most impatient to see me-not more im-
patient than I, surely. A Sudanese soldier formerly at Lado is in Mswa with
Dr. Junker's man, Binza. Jephson has written to Stanley to inform him of his
arrival, and that we are all alive, also that I would soon come.
(Here follows a letter from Jephson to Emin printed in Appendix Ai):
And still that steamer does not come.
April 27, Friday. At last the steamer arrived, at 12.30 a.m. As usual, no
one will own to any responsibility for the delay. I had the steamer promptly
unloaded, and then sent the boat to the forest to take in wood. Twenty men
have gone along to help. The grain and sesame from Wadelai we shall take
DIARIES OF EMIN PASHA-EXTRACTS X 203
south with us. At 11 a.m. the steamer was back, and by 11.55 we were under way.
Our progress was slow, our wood being very damp; a good deal of time was
lost, and we did not get to Mswa until 6.32 p.m. (Casati accompanied Emin).
It was dark, and I could scarcely distinguish a man, still very young, who
was waiting for me on the shore. Assembled behind him were his Zanzibaris,
keeping up a crackling rifle-fire. Going to the station, we remained chatting
together till nine.' The expedition must have undergone awful sufferings.
Stanley is on the mountains of Nyamsassi, awaiting me.
He has written me a long letter which I insert here in the original. I have
not many hopes, but my resolution is fixed. Go, I will not. Tomorrow I shall
prepare everything for the voyage to the south (A note by Stuhlmann shows
that Stanley's original letter is no longer in the diary. Stanley's version is printed
at full length at In Darkest Africa, i, 369-70. Extracts therefrom are set out in
April 28, Saturday. Renewed conversation with Jephson; then Mogo came.
I hear Butros Sarkis, my Khartum agent, is in Egypt. I think I shall decide
for Monbettu. It seems that at Cairo they are very anxious for me to leave
April 29, Sunday. At 8.17 a.m. we steamed away from Mswa. I had taken
on board cows, goats, fowls, meal and grain. Moreover, there were a good
many people on board, so that our vessel was heavily freighted, especially
as we were also towing Jephson's boat, a beautiful light ten-oared craft, Still,
we made good headway, and by one o'clock had passed our former night
quarters at Kayanza.
At 6.32 p.m. we anchored before Nyamsassi, and were not a little surprised
at hearing ourselves welcome by gunshots. It then appeared that Stanley had
taken up his quarters about half an hour farther south, to wait for us. I there-
fore had the vessel moored, and, in company of Casati and Jephson, went
southward in the latter's boat. We traversed a kind of swamp; on the bank,
the almost frenzied Zanzibaris greeted us with their rifles, and at last I was in
the presence of Stanley. It was a moment I shall never forget.
With Stanley was Dr. Parke2, who went through the Sudan campaign.
The other Europeans were left behind with the baggage. Stanley alone had
hurried on by forced marches, passing through horrible trials on the way,
and losing numbers of people by famine. We remained a good while in lively
conversation-my bad English must have been troublesome to Stanley. Of news
there were volumes.
A big pack was delivered to me containing some clothes kindly sent by the
Government, a packet of Petermann's Mittheilungen and maps from Gotha,
and a bundle of letters-all in a sorry condition owing to the continued damp-
ness. Stanley was good enough to entertain me with champagne, and the hours
sped like a dream till past ten.
April 30, Monday. (Steamer proceeded to Nsabe, a village of Mpigwa,
nearer to Stanley's encampment and began to unload provisions, etc.) I keep
silent about Stanley's story and it is still hardly clear to me what on earth I
have done to deserve such sympathy and sacrifice. Today I have read the Vice-
roy and Nubar Pasha's instructions (see Appendix C). I told Stanley the ins-
truction was not entirely clear to me. He told me Dr. Junker had so often spoken
of my love for the country that it was feared that I would decline a direct
invitation. At the same time Stanley tried to ascertain what personally I wished
to do, whether or not I refused to accompany him, but I promised to give
204 UGANDA JOURNAL
time to reflect thereon so that I might prepare the soldiers to undertake the
journey, which they would have to make. I briefly explained to him our position
and the motives which had occurred to lead my people astray and to make
them flinch from it.
May I Tuesday. Today I had a personal interview with Katonza, son of
Magala and chief of Nyamvio-if that is the name of the land, and not Magala.
The old chief, who visited me at Nyamvio, was one of his chiefs and I have been
"completely had" (reingefallen). Katonza is a youngish man with a good appear-
ance. He complains that after my journey Kabarega's people had attacked him
and robbed him of his cattle. His people are scattered all over the mountains
and he asks me now to attack Rwabudongo who is with the cattle on a nearby
Towards evening I went to see Stanley, and the question again came up as
to whether I intended to leave or was determined to say. He said he had two
proposals to make to me, but could not put them forward until I had acquainted
him with my final resolution. I declined this, because I do not want to come to
any conclusion until I have heard what my people have to say. If they will
not go, I presumably shall stay as well, provided that they follow me to some
place whence communications with the world are practicable, which is not the
case here, for, as soon as Stanley's expedition has retired, every connection
with the outer world naturally ceases. In the course of conversation, Stanley
asked what would I say if somebody found me a sufficient salary, and gave
me enough yearly to pay and provide for my troops-would that be an induce-
ment to stay? I said no; in the present state of affairs, victualling is impossible
here, and to accept a subsidy of the kind under the circumstances would amount
to robbing those who paid it.
Stanley rejoined by pointing out the possibility of occupying the north-
eastern corner of the Victoria Nyanza, from which communications could
be established immediately, and where the country was healthy; he thought
a project of this nature would at once find support in England. The project
appeared so extraordinarily practical and easy of execution, that I entered into
it heat and soul, and I was very pleased to see Stanley, who was a rule is fairly
reticent, take a great interest in it too3
May 2, Wednesday. Stanley has been to me, and brought Chief Kavalli,
who by his services to the expedition has deserved well of us all. I listened
to the long story of his complaints against his brother Kadongo (or Jajenkongo)
and promised him relief; but I believe we ought to abstain from all direct
attacks. Should the Khedive's order to withdraw be executed, incalculable
misery will be entailed upon the populations of these districts, as Kabarega
with his hordes will promptly scour the country, and of course such people
as the Mswa and others, who kept faith with us, will be most terribly afflicted.
It is a dreadful business, and the idea thereof haunts me increasingly.
Having given Chief Kavalli presents, fairly costly ones considering our
circumstances, which however he would not accept until with my own hands
I had given him nine cowries, he withdrew. Stanley, who was in capital spirits,
stayed, chatting about all manner of things.
Today again we reverted to the proposals that were to be made to me;
but I once more declined to decide, promising, however, that as soon as my
people had made up their minds he should not have to wait for me. But I
declared myself perfectly ready to put in execution the Khedive's order to
retire to Egypt, provided my people wanted to go; in the contrary event, it
DIARIES OF EMIN PASHA-EXTRACTS X 205
would of course be my duty to think first of the Egyptians here and their dis-
patch thither. So far as I can see at present, all the gentlemen of the expedition
are particularly bent upon getting me to withdraw to Egypt or to England,
and it sounded almost comical to hear Stanley say: 'If Major Barttelot hears
you are not coming, he will, I fear, shoot himself. He is a very impetuous man'.
Stanley had asked me repeatedly what Casati intended to do, and I always
avowed my ignorance of his intentions. When the question was renewed today,
I asked Stanley to put it to Casati himself, and on his pleading his limited
knowledge of French, I offered to act as interpreter. In the evening, as I was
going to Stanley, I invited Casati to accompany me-strange to say, he always
kept aloof. In answer to Stanley's interrogation, he said he would stick to me,
whether it meant staying or going. Such had been his promise to me, and besides,
Camperio had asked him to do so. Stanley, who is never surprised at anything,
said nothing; I myself must own that I am not altogether satisfied, because
latterly Casati has been so sensitive, and that does not augur well for the future.
It may well be that nothing will come of (this the place is mosquito riden
and Parke is feverish).
May 3, Thursday. (The Khedive left at 8 a.m. for Mswa to bring back things
Emin wanted from Wadelai). Just about mid-day Jephson came to me and we
stayed together, chatting. He also adverted to the Victoria Nyanza project.
Stanley had mentioned it to him, and according to Jephson was very enthusiastic
about it. He referred to a characteristic project of the English merchants and
English politicians. I was told about the possibility of a railway and steamers
being constructed. But as a matter of first importance Jephson constantly
said that it would be better if I came to Egypt and London.
Towards evening I went to Stanley, who was suffering from a bad cold.
We talked of many things, but were very disagreeably interrupted by the bring-
ing in one of Stanley's people, badly hurt whilst buffalo hunting.
May 4, Friday. I had given up all idea of an interview with Stanley and was
busily employed with my notes and diaries, when to my surprise towards mid-
day Stanley himself came.
We chatted a good while about many things, chiefly concerning the equip-
ment of the expedition, and eventually the conversation turned, as usual,
upon our situation. I endeavoured to convince Stanely that if I had found
support and encouragement, it would never have entered my mind to go away
from here, He replied, somewhat piqued as I first thought from his tone (I
was mistaken, however): 'In coming here I believed you to be in need of help,
being in distress, you and your men'. I felt rather hurt at the doubt implied
in these words, and I may have retorted with some warmth, for he at once
urgently requested me to give him a clear and definite answer to his question
as to whether I would go or not, and without waiting for the decision of the
Thereupon I told him practically the following:
'The Khedive leaves it to me and to my people, to come or to stay. Doubt
of our loyalty is thus implied, at which we cannot help feeling hurt; since up
to now we have been loyal. But something more is implied-responsibilities
that I cannot assume. It is perfectly clear to me, and will be equally clear to
my people, that after the departure of the expedition we cannot remain here,
remote from all intercourse, and cut off from every means of communication.
But I doubt whether my people are either anxious or even willing to go to
Egypt excepting the Egyptians and clerks, whom I said I was at once willing
206 UGANDA JOURNAL
to hand over to Stanley to take to Egypt. Had the Khedive, who surely has not
given up the hope of re-occupying the Sudan, commanded me to collect my
people at a point nearer the sea, and consequently within easier reach of com-
munications, there to await his orders, I should have done so, and the people
would have followed me: of that I was as firmly convinced as of their reluctance
to go to Egypt-with few exceptions. As regards myself, I said the matter
was simple. I myself had no desire whatever to go to Egypt, but should refrain
from interfering in any way. He, Stanley, had promised that he would leave
Jephson and the three Sudanese soldiers who accompanied him here while he
went to fetch the remaining members of the expedition. He was to give them a
kind of proclamation for my people, embodying the wishes of the Government.
Jephson could then satisfy himself as to what the people really wanted. If
they proved willing to go, I would be the first to lead them. But if only the
the Egyptians and an exceedingly small minority of Sudanese would go, I
would hand them over to him, but would remain myself, as I would not leave
my people; they had my promise to that effect. He should not be angry with
me on that account. That I should remain here I could hardly promise, as I
should have to find some spot admitting of our communicating with the world.
Whether this would be from Monbuttu by an advance to the Nepoko-Aruwimi,
or from the Victoria Nyanza or whether I could make my way to Lake Tanga-
nyika, were matters for time and reflection.'
Stanley listened attentively. He remained silent a few moments, then told
me substantially what follows: 'From what I have just heard I gather you have
no desire to return to Egypt, and would remain here if you could meet with
support. I consider the course you propose to adopt as regards the troops,
and the questions to to be correct. If the soldiers decide on going
to Egypt, it will naturally be your duty as well as mine to take them there.
But if the soldiers, or at least the great majority of them, refuse to leave, and
prefer to remain under your leadership and to follow you whenever you go,
your connection with the Egyptian Government ceases, as a matter of course;
and in view of that contingency I have two proposals to make to you. Trusting
to your honour, I will anticipate and make them now; premising, of course,
that you will always do your best to fall in with the wishes of Egypt, and further,
that you keep my secret.' That I promised, and intend to do. 'To begin with,
the King of the Belgians proposes that you should stay where you are, as
administrator of the country for him. You would be made Governor-General,
with the rank of General. He leaves it to you to name your salary; he will
place at your disposal eight to twelve thousand pounds every year to defray
expenses of the administration and to pay the troops until the country is self-
supporting. All further arrangements can easily be made, and supplies are
ready to be sent to you. The second proposal is this: Bring together the people
ready to follow you, and establish yourself at the north-eastern corner of the
Victoria Nyanza. Build stations, and at once advise Mr. Mackinnon of your
arrival. A group of English merchants is simply awaiting your arrival, to form
a company, like the former East India Company. 400,000 are ready. They ask
you to have confidence: all the rest will be seen to forthwith. A first caravan
of supplies for you would immediately leave the coast.'
To my query as to what would become of my officers in the matter of rank
and pay, I was told that they would be confirmed by the new company.
Stanley asked me to think it over and give him my decision later on. About
sundown he left, requesting that I should come to him the same evening, as
DIARIES OF EMIN PASHA-EXTRACTS X 207
he wished to show me some papers.
I did so. He handed me a map of the Congo region, and a copy of the Deli-
mitation Treaty between France and Belgium, or rather the Congo State.
He showed me the sheet on which immediately after his interview with King
Leopold he had noted his proposals. It is evident from this that the King is
most anxious to secure the Nile route.
I did not return home till ten: I think this is perhaps the most memorable
day of my life.
May 5, Saturday. (Farewell visit from Kavalli to whom Emin gave an extra
special copper bracelet. Also a visit from Mpigwa, who complains that Kabarega
raids him two or three times each year for cattle, women and children.) I first
went to Stanley at 5 p.m. because my people and Casati had gone there. That
is why I stayed behind. We had a long conversation, but let yesterday's matter
rest Stanley is an excellent story-teller, full of dry humour and has seen
and travelled throughout the whole world.4
May 6, Sunday. (A visit and a long conversation with Parke and Jephson
about (inter alia) thought reading. Then went to see Stanley in which he refer-
red to the Livingstone Relief Expedition and King Leopold of the Belgians.)
Regarding the project of the day before yesterday, as I had noted that it had
not been so well explained, regarding recruiting most of my soldiers for the new
Company. After a hesitant "yes" he began forthwith an account of a fine future,
steamers on Lake Victoria etc. I must incidentally do Stanley justice (by saying)
that he in no way tried to win me over to either project, but left it to my free
May 7, Monday. I began to obtain geographical information regarding
the local countries. (Kato is a good two days' march from here. His people
are Baganda living amongst strangers.) Dined with Stanley. The journey to
Zanzibar and the settlement on Victoria were discussed. Stanley told me Dr.
Kirk was the first to make this proposal and that he, Stanley, was empowered
contrary to my wishes, to decide that all my people should follow us. I told
him that, as Casati had previously suggested, he would like to concentrate
on Mangbettu and to establish there a trade with the Congo. A march with
not be too much stressed. Stanley appeared to be somewhat surprised or taken
aback. But he changed the conversation and I had no opportunity for returning
to the subject. I must certainly not neglect in the days coming a clear agreement
May 8, Tuesday. (Visit from Mpigwa. He and Katonza, Kavalli and his
brother Kadongo are Bahima from Banyoro. Kavalli and Melindwa's people
came from Bugoma. Mpigwa's people come from Kitara. These last are called
Bakwonga and are descended from Nkovajagali and Kavalli's people call
May 9, Wednesday. (An interpreter speaking the Mwenge dialect of Runyoro
arrived. On behalf of Stanley Emin interpreted into Arabic and the interpreter
interpreted this into Runyoro. Interpreter was originally in Bunyampaka and
later went to Busongora because of cattle raids by Kabarega.) Busongora was
conquered by a brother of Kamurasi named Kabojo, who had fled from his
brother. Kabojo's son Nyaika succeeded him. When he died he left children
who were minors. Kabarega then invaded the land and carried off the cattle.
Some of Nyaika's children were brought to Kabarega, and others fled to Bugan-
da. Of these last the eldest is Kakende and has returned home and taken over
the chieftainship. He is supported by the Baganda and they have promised
208 UGANDA JOURNAL
him help against Kabarega.
Stanley brought me a paper in his handwriting, which contained the pro-
clamation he wishes to be made to my people and asked me to translate it into
Arabic.. I set out the words of Stanley's proclamation (set out in Appendix B).
May 10, Thursday. I have just made the Arabic translation of the foregoing
message. In it are two passages which will presumably give offence and make
the people refuse to undertake the journey. The first passage is "If you stay here,
you are no longer his soldiers". The second passage is "Your pay continues
until you arrive in Egypt". I do not know what the instructions of the Govern-
ment are and cannot make any assumptions that the intention is to discharge
all the people here. Moreover, this may be the result of a badly-drawn-up
May 12, Saturday. Rwabudongo is on an island called Bukokwa with an
Arab and in possession of a supply of Kabarega's gunpowder. Elephant tusks
etc. are underneath his chief hut. Ireta is two days march from him and on the
mainland. Both of them have very few weapons because at the invitation of
chief Kakande of Busongora the Baganda have undertaken a fresh raid against
Kabarega and these weapons would probably be seized by them. Kabarega
has ordered Ireta and Rwabudongo to send northwards to him any herds or
supplies because of the danger that they will fall into the hands of the Baganda.
I naturally cannot determine how much of this story is true or false, but Katonzi
appears to be in earnest ..
With Katonzi came another chief with a strikingly light beard with a
pronounced Bahima physiognomy from Uale or Katualo which is on the mount-
ain. He is by origin from Njakkasadie, wither his ancestors, Babito like Kaba-
rega himself, wandered before these times of oppression. After his father's
death Kabarega laid claim to his children and cattle, but he is going back and
this plundering will cease. He, Nsagara, has made an alliance with Kabarega
and has now come to me to free himself from it. His father Kafumo took care
to agree with Kabarega and formerly warned him to do so. Katonzi, whose
name before he became a chief was Kakongo, is, as he boasts, one of the Babito.
Stanley, to whom I gave this note, says. Katonzi is a liar and is perhaps
luring me into a trap.
May 13, Sunday. (Return of Parke from a foraging expedition in the Lendu
May 14, Monday. (The steamer arrived at 4 p.m. bringing from Wadelai
corn, some cattle, and other things for which Emin had written: no wood was
brought. The native boat was sent to Waledali to fetch Emin's people who were,
as usual not ready).
May 15, Tuesday. Katonzi and his people are Babito pure and no Bahima
dwell near them. Of the chiefs in the south the most powerful are Kakende,
son of Kiboja in Busongora and Nsagara, who was recently here.
May 16, Wednesday. (Blood-brotherhood concluded between Katonzi and
Emin's officer, Shukri Aga. Katonzi and Mpigwa sailed on the steamer to the
north. Also Casati. Only Shukri Aga and twenty of Emin's man remain with
May 17-20. (Emin borrowed Stanley's sextant to make various observations.
The boat returned on 20 May from Mswa with a letter from Casati of 17 May.
He reached Tunguru on 17 May and found Vita Hassan was ill).
May 22, Tuesday. At eleven in the morning the two steamers arrived with
130 porters for Stanley, 80 soldiers, Hawash Effendi, Selim Aga, Kodi Aga and
DIARIES OF EMIN PASHA-EXTRACTS X 209
Rehan Aga. Grain too, and some fruit and other things for my guests, one of
whom, Jephson, is unfortunately ill again. I at once took the officers to Stanley.
He was delighted by the arrival of the 130 porters, and informed me he would
be leaving the day after tomorrow. After conversing with the officers for some
time, Stanley promising them that he would explain the orders of the Govern-
ment in the evening at my house, we went home.
I am in receipt of letters from Wadelai: all is well; but the leopards enter
the station every night. One of them, the other night, carried off from a hut a
boy twelve years old, the son of Ali Aga Kurdi. Nothing has been found of him
but his head. Hawash Effendi reports another accident. As the Nyanza, with
Osman Effendi on board, was going from Wadelai to Dufile, the boat in tow
capsized, and eleven persons were drowned in the river; this was due to a
storm. In Dufile they are having a lot of rain, and the crops everywhere are
In the evening Stanley came as promised, and the officers had an opportunity
of hearing from his own mouth the Khedive's view and wishes. (The Khedive's
letter will be found in Appendix C)
Hawash Effendi, as usual, having the most say, of course everybody was
ready to obey implicitly. But will this welling up of noble sentiment endure?
Kedi Aga alone said, 'Where my chief goes, there I go'. Not until long after
sunset did the company break up, and I had an opportunity of chatting with
Stanley and making divers arrangements.
Tomorrow we will prepare everything for his departure.
May 23, Wednesday. I am expecting the neighboring chiefs today, having
summoned them yesterday.
Katonzi came at noon. He tells me that the messenger whom we recently
sent to the people on the plateau is on the way, and may be expected tonight
or tomorrow morning; he was bringing no people, but doubtless had news for us.
The Baganda had invaded Bunyoro, driven Kabarega out of Mruli (he was
now in Mparo), and stollen his flocks. Rwabudongo with a few armed men was
still in Bukokwa; I ought to drive him out. In answer to my question, he told
me that no steamer could go so far; we should have to leave it at Ndigi, proceed
six hours in boats, and the rest on the way, ten hours, overland. That make
the project totally impracticable, at all events for the people at my disposal;
so I shall have to arrange the matter in another way.8
I made use of the opportunity to ask Katonzi to what tribe a girl belonged
who, as I heard, was brought from Kabarega's country to the Acholi and could
not make herself understood there. When I mentioned that she had four of
her incisors filed down, and had a reddish skin, he said at once that she must
belong to the Bairu, who had come down to me with Chief Kavalli. He stated
that these Bairu, whom he called Bahima (shepherds), came from a great
river in the west, named by the Btoli or Uortali (is it Stnaley's Ituri?). They
were come for cattle, and bartered their produce for meat, etc. Their villages
are Bkaranga, Ssiri, Undjoca, Vlangue, Bumbula, and many others. They spoke
the language of the various countries, but their own language among themselves.
In the evening Stanley came, bringing a chief only arrived today, from the
regions he had traversed-Chief Mpinga, from Usansa, a Muiru chief. He con-
firmed all the foregoing names and news, and told us that from his house
alone Kabarega's people had carried off twenty-five persons, of whom five
had got back, but twenty were still missing. It would be singular if the girl
I spoke of turned out to belong to him.
210 UGANDA JOURNAL
I conferred at some length with Stanley about his departure tomorrow, and
we again discussed all the measures to be adopted. In the evening I entertained
Jephson and Parke, as guests.
May 24, Thursday. At five in the morning I marched out with fifty men
to shake hands with Stanley and Parke two hours from here. They followed
me soon after sunrise, and a short leave-taking parted us-for six or seven
months at least. I retraced my steps with a somewhat heavy heart. The man
left behind by Stanley, owing to his being hurt in a buffalo hunt, I had conveyed
to the camp, also several women and children committed to my care by Stanley's
people. Then Jephson came over.
A little later Mpigwa came. He again begs me to attack the Balega (Lendu)
and to take the present opportunity of doing so. He also spoke about the in-
vasion of Bunyoro by the Baganda and asked me to agree to an attack on
Katonzi and Kavalli.
At noon six men came from Dr. Parke. Twenty one Madi porters had
deserted on the way. I at once sent to search for them.
In the evening after sunset Dr. Parke suddenly appeared with about fifteen
men and delivered to me the following letter from Stanley.
(The letter is set out in Appendix A5).
I at once sent out a punitive party to Kanamo to seize the deserters so that
Dr. Parke could go back with them early in the morning. At 8.30 p.m. were
having supper when a boat came from Mswa with news tnat 120 porters from
Dufile had arrived there I desired to request Stanley to wait only one day more
and I sent the Khedive at midnight with Vita to Mswa to bring the people here.
May 25, Friday. At 6.30 a.m Dr. Parke, to whom I gave an second of ten
men and my letter to Stanley, marched off. He promised to my men back
this evening with Stanley's answer.
(Emin's letter is set out in Appendix A.6)
I will, if I possibly can, go to look at the new snow mountain. Katonzi,
after being summoned by me is returning hom. I have promised him that in
four days' time I will attack Kabarega's people.
... Kavalli met me today with large parties of his people. He has established
himself in Stanleys abandoned camp. He came wiht about five hundred men
to go on some raids. I told him that I had caught spy at Katonzi's place and
that it would be better if I did not get mixed up in Katonzi's affairs. In order
to avoid Rwabudongo. I shall proceed on the east side of the lake and would
set out the day after tomorrow. I asked him to go back to Stanley and to help
him in every way.
At 5 p.m. the soldiers sent by me with Dr. Parke came back also ten men
of Stanley's to take over the porters we were expecting. They brought a letter
from Stanley and said they had not met a single deserter on the way. I hear
that the deserters have found their way to Njangobia and were held by the
inhabitants, but on their saying that they came from me they had been set
free and have got away. Without doubt they will go to Dufile and cause much
(Here follows Stanley's letter to Emin which is set out in Appendix A.7)
May 26, Saturday. At 2.30 a.m. the steamer came with 82 porters whom
I at once had shackled. Their weapons were collected in bundles and, will be
restored to them when they can no longer run away. I sent twenty-five men and
an officer with a letter from myself (Appendix A8) and Kavalli's people. It is
to be hoped that the porters will soon return Stanley had 21 in addition to
DIARIES OF EMIN PASHA-EXTRACTS X 211
these 82. The soldiers have come back. So I will this evening get rid of Mpigwa
and the Lendu and hope to go tomorrow morning to Namsassi to load wood.
Today Jephson has gone with a new net, which he made yesterday, to hunt
At 11.15 a.m. the officer came back with the people, whom I had sent with
the porters. He has promptly delivered the porters and brings the enclosed
letter (Appendix A9)-the last from him for full many a month. May God
give him the good fortune to come back to us!
At 9 p.m. I sent fifty men and two officers under the guidance of Mpigwa's
people against the rebel Balega.
Correspondence between Emin and members of the Stanley Expendition. (Except where other
wise stated, the original letters were incorporated by Emin in his diary.)
Jephson to Emin 24 April, 1888, Mswa.
Mr. Stanley sent me in our boat to bring you a letter and to tell you where we are. I
started from Nyamsassi the day before yesterday and reached here at 10 p.m. Not finding
you, I am sending you this note by one of your men, who tells me you will reach this by 12
Emin to Japhson 24 April, 1888, Tunguru. (Mounteney-Jephson,
Emin Pasha and the Rebellion at the Equator, p. 20)
Your letter of yesterday was delivered to me this night. Be heartily welcome among us;
we have waited for you many a long day. I propose to start at once to rejoin you at Mswa;
the steamer having gone, however, to fetch some corn, and the people being busy with the
fields, I must necessarily delay until the steamer returns. I have sent for and expect her tomir-
row. It goes without saying that at her arrival I start.
I have given orders to my men to provide for all your needs, and those of your men;
please therefore to acquaint Shukri Aga, the officer in charge of the station, with all your
wants. Suleiman Effendi, the bearer of this, has to stay with you until my arrival.
Hoping to see you very soon.
Jephson to Emin 25 April, 1888, Mswa.
Your letter has just reached; Suleiman Effendi tells me he is just sending you word of
his arrival here. I therefore take the opportunity of sending you a line to say how glad I shall
be to see you tomorrow. We have travelled through interminable forests for so many months.
and have gone through so many difficulties to reach you, that it was a great disappointment
to me, arriving here, to find you absent. I trust we shall meet tomorrow.
With sincere thanks for your expressions of welcome.
Stanley to Emin. 18 April 1888, Kavalli's. (Extracts from letter as printed In Darkest Africa,
Your letter was put into my hands by Chief Mbiassi of Kavalli (on the plateau), the day
before yesterday, and it gave us all great pleasure.
The instructions of the Egyptian Government were to guide you out of Africa, if you were
willing to leave Africa; if not, then I was to leave such ammunition as we had brought with us
for you, and your people were then to consider yourselves as out of the service of Egypt,
and your pay was to cease upon such notification being given by you. If you were willing
to leave Africa, then the pay of yourself, officers and men was to continue until you landed
in Egypt .
Now I await your decision at Nyamsassi .
If this letter reaches you before you leave your place, I should advise you to bring in your
steamer and boats, rations sufficient to subsist us while we await your removal, say about
12,000 or 15,000 lbs. of grain, millet or Indian corn, etc., which, if your steamer is of any
capacity, you can easily bring.
If you are already resolved on leaving Africa, I would suggest you should bring with you
all your cattle, and every native willing to follow you ...
The letters from the Ministry of War, and from Nubar Pasha, which I bring, will inform
you fully of the intention of the Egyptian Government, and perhaps you had better wait to
see them before taking any action .
212 UGANDA JOURNAL
I hear you have abundance of cattle with you. Three or four milk cows would be very
grateful to us if you can bring them in your steamer and boats ...
We shall have to forage far and near for food while we await your attendance at Nyamsassi,
but you may depend upon it we shall endeavour to stay here until we see you.
Stanley to Emin. 24 May 1888, Badwza Village.
I regret to state that only 19 of your carriers have arrived. At the first river 21 disappeared
and four men were despatched with a note to you I believe. Within a mile of this village 89
deserted firing their arrows at the rearguard. Several shots were fired in return but more with a
view of terrifying them then with intent to harm. The Doctor however, seeing matters serious,
shot one of the natives. The remaining 19 were bound and conducted here.
The chief (Osman) of the escort says that he thinks the natives will attend (sic) to reach
your Camp tonight as there is no other way for them. He suggests that cowhides should be cut
up and made into cords, and that the natives be tied by the neck in gangs of six and that they
should be deprived of their weapons. He thinks it will be easy then for each of our men to
drive here a gang of six.
Should they (natives) appear please act as you think necessary. Our men will abide by
your orders, and I will wait for the here tomorrow. Now that we understand their nature we
will do our best to keep what we have. Anyhow we shall not trouble you a second time about
Also I beg to suggest that if these carriers are of the same kind as you propose that you
take on your journey to the Nyanza (Victoria) that you take advantage of the period of our
absence to make some light chains for a set of 10 carriers with neck rings. Your steamer
anchor chains, funnel chains, boat anchor chains will take a few, but you will require about
a hundred chains, and a hundred men would easily guard the entire number. On arriving at
Ituri we could set them free as there would be little fear of their desertion there. It is an unples-
ant task. but the lives of your heroes demand a little sternness towards such people as the
carriers we have had with us today.
I beg to tell you a bit of news. For the first time I have seen a snow mountain in Africa.
Its altitude is immense. The condition of the atmosphere rendered the mountain singularly
distinct for 2 hours.
I can get no name for it from Gavira's natives who are with me, but I suspect it is Ru-
wenzori and that the "white metal" is snow.
I took bearings of it and found the centre of it to bear 210; Mag from a spot 4 miles from
This village is 10 miles from your Camp.
Please say to Mr. Jepshon that I missed him this morning, when starting on the road, and
that I regret losing my chance of wishing him heartily the best of good health, and happy
restful days until our return and to you my dear Pasha let me thank you once more for your
abundant hospitality and to wish good health and a tranquil mind until we shall meet again.
Stanley's Proclamation to Emin's soldiers, 9 May 1888.
(Jephson, op. cit. 48-50.)
Soldiers of Emin Pasha! After a long journey from Zanzibar I have at last reached your
Nyanza and seen your Pasha. I have come expressly at the command of the Khedive Tewfik,
to lead you out of this country and show you the way to Egypt. For you must know that the
river el-Abiad is closed, that Khartoum is in the hand of the followers of Mohammed Ahmed,
that the great Pasha Gordon and all his people were killed three years ago and that all the
country and river between Wadi Halfa and the Bahr el-Ghazal is occupied by your enemies
and by the rebels. Four times have the Khedive and your friends made attempts to help you.
First Gordon Pasha was sent to Khartoum to bring you all home, but before he could safely
leave Khartoum, that city was taken and he himself killed. Next the English soldiers came
near to Khartoum to try and help Gordon Pasha, but they were four days too late, for Gordon
was dead, and Khartoum was lost, Next came Dr. Fischer by way of the Nyanza of Uganda,
but he found too many enemies in the path, and returned home and died. Next came Dr.
Lenz, by way of the Congo, but he could not find men enough to carry his goods, and he also
went home. I tell you these things to prove to you that you have not been forgotten in Egypt.
No, the Khedive and his vizier Nubar Pasha have always kept you in mind though they
could not reach you. They have heard from your Pasha, by way of Uganda, how bravely you
have held to your posts, and how staunch you have been to your duties as soldiers. Therefore
they sent me to tell you this, and to say to you that you are well remembered and that your
reward is awaiting you. At the same time the Khedive says that if you think the road is too
DIARIES OF EMIN PASHA-EXTRACTS X 213
long, or are afraid of the journey, you may stay here, but if you do so you are no longer his
soldiers, and that your pay stops at once, and that if any trouble befall you hereafter you are
not to expect any help from him. Should you decide to obey him and follow me to Egypt I am
to show you the way to Zanzibar, and there put you on board a steamer and take you to Suez,
and thence to Cairo, and that your pay continues until you arrive in Egypt, and that all pro-
motions made here will be secured to you, and all rewards promised to you here will be paid
in full. I send one of my officers Mr. Jephson to read to you this message, and that you may
know he comes from me I lend him my sword. I now go back a little way to collect all my people
and goods and bring them here. After a few months I shall return to hear what
you have to say. If you say, 'Let us go to Egypt', I will then show you a safe road, and will
accompany you and not leave you until you stand before the Khedive. If you say, 'We shall not
leave this country', then I will bid you my farewell and return to Egypt with my own people
and give the Khedive your answer. May God have you in his safe keeping! This is from your
good friend, STANLEY.
Instructions of Tewfik Pasha to Emin.
(Schweitzer (Felkin) op. cit. i, 277-8.)
To His Excellency Mehmed Emin,
Mudir of Hatt-el-Estiva:--
We have already expressed to you and your offices our thanks for the valorous and vi-
ctorious defence of the Equatorial Province confined to your administration, and for the firm-
ness displayed by yourself and the officers under your command.
We have therefore rewarded you by raising your rank to that of Leva Pasha (Brigadier-
General). We have also approved of the promotions which you deemed necessary for the
officers under your orders, as I wrote you before, on the 29th of November, 1886. That com-
munication (No. 30), together with other documents sent you by His Excellency Nubar Pasha,
President of the Council, must have reached you.
I being our heartfelt desire to deliver you, together with your officers and men, from your
present difficult position, our Government has arrived at a conclusion as to the ways and
means of rescuing you, your officers and men.
As a Relief Expedition has been organised under the command of Mr. Stanley, the cele-
brated and experienced explorer, whose fame is spread over all the world, and as he intends to
enter upon his mission with all the supplies you may require in order that he may bring you to
Cairo with your officers and men by the route he considers advisable, we have issued this order
to you. The same will be remitted to you by the hands of Mr. Stanley that you may know what
is to be done, and I command you, as soon as this order is in your hands, to convey my greet-
ing to the officers and men. You are absolutely free either to withdraw to Cairo, or to stay
where you are with the officers and men.
Our Government has decided to pay your salary, and the pay of the officers and men.
Such of the officers and men as wish to stay may do so on their own responsibility, but
must not expect any help from the Government in future.
Endeavour to grasp the real purport of this letter, and make the officers and men familiar
with it, so that they may know what they have to do.
8 Gamad Aual, 1304. (2 February, 1887)
(Signed) Mehemet Tewfik
(The communication of the 29th of November, 1886, to which the Khedive refers, appears
never to have come into Emin's hands.)
(Nuber Pasha's letter is dated one day after that of Tewfik Pasha. When received by Emin,
parts of it were illegible owing to damp. It would appear to have been to the same purport
and effect as that of Tewfik Pasha and is accordingly here.)
1. Mounteney-Jepbson describes Emin as being a small, wiry, neat but most unmilitary-
looking man, with unmistakable German politeness of manner. He spoke English with much
ease and fluency," Casati did not understand English but had a limited knowledge of French
(Emin Pasha and the Rebellion at the Equator, p. 4)
2. Parke describes Emin in his diary as being "extremely short sighted. He is very slightly
built and rather short of state (about five feet seven in height). He had a bronze skin and black
hair. He looked cheerful and was excessively polite." (My Personal Experiences in Equatorial
Africa, p. 223.)
214 UGANDA JOURNAL
3. "Emin Pasha is still very doubtful about leaving his province. We have brought him
many packages of letters, which we are hoping may influence him. He seems to look upon
himself as the slave of his people and his services are entirely theirs, to be used as they think
proper." (Parke, op. cit. p. 226) Mounteney Jephson was accordingly detailed for that purpose.
4. "Emin Pasha requests Mr. Stanley to allow one of his officers to visit the province
so that his people would have secular evidence that our expedition had actually come
for their relief." (Parke, op. cit. p. 226). Mounteney-Jephson was accordingly detailed for
5. "Emin Pasha still seems very undecided about coming out with us. He avers he would
be quite happy in Africa, if he could receive letters, and papers from Europe. He says he will
come away if his people allow him. He does not like the idea of leaving his people and does
not think that they will come out with him, as they are all aboriginals and hate the Egyptians."
(Parke, op. cit. p. 227)
6. Parke describes this incident in some detail. "On May 10th seven of our Zanzibaris
took their rifles yesterday and rambled away to perform some private looting among the
natives on the hill side." The natives resisted and one of the raiders was wounded by an arrow
and he and his rifle were captured. Parke was sent with a party to recover the man and the gun.
The natives of a nearly village attacked them. Parke halted for the night having "told off a
small detachment to burn about 30 huts and give us a clear space of about fifty yards round
our position." He returned to Stanley's camp on 12 May (op. cit. pp. 228-9).
7. Parke alleges that at this meeting "the Pasha showed his weakness by asking Hawash
Effendi in our presence, if he could promise there before Mr. Stanley to use his influence to get
his people out. This lowered his status at once in our eyes as it clearly proved the Pasha's
want of authority. He evidently has no power to lead off his people in any direction. He has
a voice among them, perhaps a little more than the average" (op. cit. p. 228). One may ask
whether this was really a show of weakness on Emin's part. One recalls the incident described
by Shakespeare-Henry IV, Part 2, Act. 3, when Glendower asserted that "I can call spirits
from the vasty deep". and Hotspur retorted "Why, so can I, or so can any man. But will they
come when you do call for them?"
8. This paragraph does not appear in Stuhlmann's edition of the Tegebucher. As said in
his preface to Volume IV of that work, he had with a view to economy in printing cut out
about one quarter of the passages appearing in the original diary. This would appear to be one
of those thus excised. It appears in Georg Schweitzer's Emin Pascha and, as a later passage in
Emin's entry in the diary of this date shows, it evidently did appear in the original MS.
9. Stanley, approaching Lake Albert for the second time, had at last made contact with
Emin at Kavalli's village on 18 April 1888, when he was handed a letter from Emin dated 25
March, 1888 (printed at In Darkest Africa, i, 267-8). On 20th Stanley despatched Jephson and
Parke to the Lake shore with the Expedition's steel boat. Parke records that whole on the march
he and Jephson distinctly saw snow on the top of a huge mountain and reported this to Stanley.
(My Personal Experiencew, p. 217).
Mrs. Dorothy Middleton, who is editing Jephson's diaries for the Hakluyt Society has
very kindly informed me that, when Emin acquainted Jephson with the contents of this letter
from Stanley he (Jephson) wrote in his diary (26 May 1888) that-when he had informed
Stanley of his and Parke's discovery Stanley had "poopoohed the idea" and had "laughed at
the idea and said it was only mist." Clearly to Jephson and Parke and not to Stanley must be
ascribed the credit of being the first European to see the snow-peaks of Ruwenzori (cf. Uganda
J. 2 (1934-5), 249).
Uganda Journal, 29, 2, (1965) pp. 215-221
MAIZE IN UGANDA
by B. W. LANGLANDS
In the most recent edition of the Journal of African History there is an
article on the history of maize in Africa by Marvin Miracle' in which he com-
pares the observation of Grant in 1863 that maize was "very rare as the equator
is approached; and quite unknown beyond it to 5ON",2 with a remark by
Johnston in 1901 that "wheat, oats and barley grow well in some districts,
Indian corn everywhere "3 Miracle states that "after Grant's note there is
no further reference to maize in Uganda until 1901", and thus deduces that
"maize appears to have been introduced into Uganda between 1863 and 1900".
He goes on to discuss the possibility of maize having been brought into Uganda
by Arab traders from the south, but recognizes that "it may have been intro-
duced to one part of Uganda by one means, and to other areas by another."
The discussion would probably not be worth taking further were it not that
these statements have been written by a leading scholar on maize in Africa
and published in the leading journal of African history. The following note
and references are therefore submitted by way of clarification and amplification.
The validity of the two references is not disputed; but it will appear that they
have been put to an odd use and the omission of many other references to maize
in Uganda can only reveal that the commentator can have made little search
for further evidence.
To take the Johnston reference first, it must be noted that the Uganda of that
time extended to the Kenya rift valley and included therefore the west Kenya
highlands which undoubtedly was the area thaL Johnston had in mind as being
suitable for wheat and barley. It is unfortunate that the quotation as presented
by Miracle gives the impression that by 1900 maize was already being grown
extensively in Uganda, rather than, as Johnston probably intended that it was
grown in some areas and that conditions were favourable for its growth every-
where.4 In point of fact even by the 1930's maize only occupied a relatively
small cultivated area, averaging 68,000 acres between 1932 and 1935, with a
peak of 142,000 acres in 1930. It was mentioned in this period as an important
crop in the diet of the people in the Lake Albert lowlands and in Bugisu;
though it was grown in some other districts as a minor crop in the agricultural
rotation. In Buganda it was grown fairly frequently in conjunction with the
cotton crop, especially around the borders of the cotton plots; but it is only
more recently that attempts have been made to include it in rotation systems
with cotton. In addition in Buganda it was commonly grown by the 1930's
at any rate, in the banana gardens and in the sweet potato plots.5 The cultivation
of maize has since increased and reached a maximum of 500,000 acres in 1953,
but much of this increase has been as a result of its cultivation as a cash crop,
especially in parts of Eastern Province where its growth has been influenced
216 UGANDA JOURNAL
by the fact that neighboring areas of Kenya have become maize growing areas,
and also by the fact that migration took place from western Kenya into this
area. By 1958 small quantities of maize at least were grown in every district
of Uganda, though little came from the drier areas of the northeast.6
This brief review of maize distribution in the twentieth century may assist
in the further appreciation of maize in the preceding decades. In the brief
reference to maize in Tothill's Agriculture in Uganda, already referred to,
C. G. Hansford writes "Nothing is known of the introduction of maize to
Uganda, but it has been cultivated in most parts of the country for a long peroid
and is mentioned by several of the early explorers."7 It is this aspect of the
maize problem that must now be developed since it is somewhat at variance
with the statements that this survey opened with.
One can begin with a closer enquiry into the writings of Johnston and Grant.
In the first place it may be noted that although Miracle gives an obscure reference
to a mention of maize by Johnston in a Government report; Johnston also
makes a reference to maize in Uganda in his book on Uganda which
was published only a year later. In this he mentions that "a little maize and
still less sorghum is grown for food" in Buganda and Busoga.8 Maize is also
listed as a food crop in Acholi, together with many other crops.9 but other than
stating that Indian corn flourished well in Uganda, Johnston makes no further
reference to maize in the area of the present-day Uganda.10 This then gives a
clear indication of maize in central Uganda and Acholi, and recognizes the
fact that it was not grown universally in 1900. To turn to the other key reference
of the paper under discussion one can look more widely at the writing of Grant.
In his own account of his travels through Uganda in 1862 he mentions the pre-
sence of maize at least twice. For central Mengo, near what was then the border
between Buganda and Bunyoro at Karee he recorded in July 1862 "plantain
groves were more abundant than fields of sessamum and Indian corn"."
Then again in central Acholi, in what was then called Gani, Grant wrote In
December 1862 "we had entirely lost the plantain trees of Uganda, and reariy
came upon Indian corn or groundnuts, which up to Unyoro had been always
procurable."12 From these quotations, then, it is evident that even in 1862
maize was to be found in central Buganda and in Bunyoro but that it was scarce,
(though not necessarily absent) in Acholi. In this respect it would appear
that the statements by Grant differ in his own book from the extremely brief
summary he had compiled for Speke's book. It is unfortunate that Miracle
has been led to assume from the latter that maize was entirely absent, whereas
a closer reading of Grant himself would have revealed that this was not neces-
sarily the case. Thus it would seem certain that maize did already exist in Uganda
even by 1862 but that it may not have been universal. Looked at in this way
the remarks of Johnston in 1902 do not seem so very different from those
those of Grant in 1862.
The next stage in this discussion is to refute the statement of Miracle that
nothing was written about maize in Uganda between 1862 and 1900. The
following quotations cannot claim to be a complete record of all references
to maize during this period but at least they will present sufficient evidence to
reconstruct the picture of the distribution of maize during this period. The
references have been arranged simply in time sequence for Buganda, Bunyoro,
Toro and Ankole, West Lake Albert and Acholi, for these seem to be the
areas where maize was definitely being grown in the latter half of the nineteeth
The next reference to maize in Buganda after that of Grant that I have come
across is in the account of their visit to Buganda of Wilson and Felkins; in a list
of crops grown in Buganda made by Stanley in October 1875 no mention is
made of maize. Stanley writes that in the gardens of the peasant "grow large
sweet potatoes, yams, green peas, kidney beans, some crawling over the ground,
others clinging to supporters, field beans, vetches, and tomatoes. The garden
is bordered by castor-oil, manioc, coffee and tobacco plants. On either side
are small patches of millets, sesamum and sugar cane. Behind the house and
courts and enfolding them are the more extensive banana and plantain plan-
tations and grain crops, which furnish his principal food, and from one of
which he manufactures a wine and from the other his potent pombe."13 It
seems odd from this that maize should not have been listed if it was in fact
being grown at the time, unless it was that Stanley had intended to include it
amongst the unspecified "grain crops". However, in a similar review less than
two years later it is recorded by the missionaries Wilson and Felkins, "the
sweet potato forms, next to the banana, the main article of food and is the chief
vegetable cultivated by the natives. In addition to these plants, the Waganda
also grow Colocassia antiquorum, Helmia bulbifera, yams, several species of
beans, two or three kinds of pumpkins, a species of solanum, sugar cane,
a kind of spinach, cassava, maize, millet, tullaboon and sesamum."14 In Dec-
ember of the same year, 1877, Emin Pasha came through to Rubaga and men-
tions the presence of maize on a number of occasions on the route from Mruli.
Thus at what must now be part of central or north Mengo he wrote "the further
we proceeded the more beautiful and highly cultivated grew the land: luxuriant
fields of sweet potatoes, juicy green tobacco, yellow-flowering groundnuts,
fields of maize and sorghum, all divided from one another by clearly kept
paths; whilst in the background there were groves of bananas."15 Briefer
mentions of maize are made on this same expedition for neighboring portions
of Buganda north of Rubaga.16 It would seem possible from these references
to maize in Buganda from Grant, Wilson and Emin Pasha, and even by the
omission from Stanley that maize was more likely to be met in the drier zone
away from the lake than it was in the lake zone of the country. This situation
is confirmed for a slightly later date by Lugard who on reaching the Kanya-
ngoro river on the borders then between Buganda and Bunyoro in Singo in
April 1891 observed "the soil is extremely rich, and the cultivation consists
more of grain (millet, maize and wimbi) and beans, than of interminable
bananas and roots."17 Another interesting list of crops grown in Buganda in the
1890's which includes maize (Indian corn) is that of Decle for 1893-1894 which
states that "the plants cultivated are very numerous comprising bananas,
sweet potatoes, Indian corn, cassava, peas, beans, a root called mayuni, monkey
nuts, millet, wheat, another small kind of red millet called wimbi, rice, tobacco,
gourds and vegetable marrows of various sorts."18 It would be rash to argue
that the higher place of maize on this list than on previous ones indicates an
increase of maize cultivation between 1877 and 1893, nevertheless the change
is worth attention.
As already seen from the quotation from Grant maize was already growing
in Bunyoro by 1862; but since the area of Bunyoro that he passed through is
mostly incorporated in Buganda this statement may be taken as indicating the
218 UGANDA JOURNAL
presence of maize in what is now north Mengo. Nevertheless, Emin Pasha
confirms the presence of maize in the present-day area of Bunyoro as early
as 1877, for in May of that year on his way across north Bunyoro to Mruli
he "arrived at clearings, where bananas, sweet potatoes and lubias intermingled,
and here and there the green stalks of maize were seen, or the broad leaves of
Virginian tobacco."19 Likewise in August of that year at Kiroto in north
Bunyoro he reported "any number of cultivated plants grow wild round this
place tomatoes (which grow wild in masses) solanum, melongena, maize,
lubia "20 At Foweira, on the bend of the Victoria Nile he commented in
1878 upon the fertility of the soil "a white sand mixed with a yellowish clay, in
which sweet potatoes, durrah and especially maize thrive splendidly."21 Later
in 1885-1886 he records maize growing at the northwestern corner of Bunyoro
at Fanyumori, south of Magungo, "fields planted with groundnuts and maize".22
Reporting on agricultural practices in Bunyoro after a fairly long stay and after
a journey across central Bunyoro Emin's companion Casati wrote in 1887
that "maize, Eleusine coracana, Batata edulis, Helmia bulbifera, manioc, and
various species of vegetables thrive there, and bananas, the staple food con-
sumed by the people, yields abundance of excellent fruit."28 During the course
of the Bunyoro wars of the next decade various military commentators passing
through central Bunyoro mentioned the presence of maize. Thus Thruston
after crossing the Kafu river in December, 1893 records "our next two marches
lay through a very fertile, picturesque and hilly country with broad valleys,
in which were the pleasant plantations of the Wunyoro; smiling with banana
groves, patches of sweet potatoes, Indian corn and sesame,"24 and two years
later Vandeleur made a similar observation for the area around Mt. Fumbi in
Toro and Ankole
There are also references to maize in other parts of western Uganda in the
latter part of the nineteenth century. The earliest of these is from Stanley who
on his first journey across Ankole to Lake George in January 1876 noted that
an "abundance of Indian corn, millet, sweet potatoes, bananas and sugar cane
had also been seen on the lake shore."26 In his subsequent visit to the Ankole-
Toro borderland Stanley again reported maize growing at Katari in Unyampaka
in July 1889.27 Certainly by the turn of the century maize was established in
south Ankole where "the natives in Buddu, Ngarama, Ruampara and west-
wards cultivate sweet potatoes, groundnuts, beans, peas, manioc, and a little
maize and tobacco;"28 on the Semliki29, 30 and in central Toro.31, 32.
West Lake Albert
Another area at least adjacent to these areas of Bunyoro and west Uganda
already considered in which maize was growing in the latter part of the nine-
teenth century can be recognized on the coastlands and hills facing the western
side of Lake Albert. Maize was noted in the Alur area of the north of this region
near Mahagi by Emin Pasha in November 1879 where "this year's second crop
of durrah was just ripe. Maize, red and white durrah, eleusine, sesame, a little
tobacco, and a kind of cucumber... and groundnuts have been planted here."33
Further soutf maize was met with in 1888 and 1889 when Stanley and his com-
panions of the Emin relief expedition arrived to find the "huts were stored
with Indian corn" and "no lack of Indian corn",34 and "large fields cultivated
for growing Indian corn, durrah and sweet potatoes".35 At Mswa south of
Mahagi where Emin had first noticed it Mounteaey-Jephson reported "large
fields of m'tama, Indian corn, sweet potatoes and groundnuts, mingled with
groves of bananas surround the villages,."36
It is also clear from the accounts of Emin Pasha that maize was being grown
in Acholi by 1880, and though his troops may have been responsible for the
introduction of new types and for the spread of it to new areas by planting
it in the vicinity of their forts, it is also clear that maize had preceded Emin into
this area. In November 1880 Emin reported that it was growing at Fachora,
southeast of Emin's Fatiko, "tobacco, Helmia (of which only the bulbs are
eaten), lubia, three kinds of gourd, sesame, maize and Vigna were cultivated."37
For Equatoria in general, but including Acholi, Emin noted in 1885 that
"maize grows excellently in every part of the country and its cultivation is
extending"38 The cultivation of maize about Emin's own station at Dufile in
1888 is recorded by Mounteney-Jephson.39 The fact that Emin himself was
responsible for bringing new types of maize to Beden north of the present
Acholi boundary is reported by Emin himself to Dr. Schweinfurth in 1881
the "horse-tooth maize which His Excellency Stone Pasha sent me also thrives
well; each crop, however, requires to be sown on new ground."40 By the turn
of the century maize seems to have been grown fairly generally throughout
Acholi. Thus Bishop Kitching noted that "the Gani also sow the largest variety
of crops perhaps, millets of many kinds, semsem in large quantities, ground-
nuts, a little maize, sweet potatoes and a good variety of minor herbs, beans,
etc,"41 and Delm6-Radcliffe "throughout the Nile Province red durrah is culti-
vated, telebum, duku, Indian corn, semsem, tobacco, groundnuts and sweet
potatoes are largely grown."42
Other parts of Uganda.
Outside these parts of Uganda already examined very few references to
maize before 1900 have been traced; but for the most part the remaining areas
of Uganda were not on the main routes of the explorers. The extreme eastern
and northeastern borderlands were not opened up until the late 1890's, but
even for this area there is at least one reference to maize already existing.
Austin who accompanied Macdonald's expedition to the north records that
in November 1897 the Save Plateau of north Elgon was "thickly cultivated with
bananas, Indian corn, mtama, beans, pumpkins, yams and various other
foodstuffs."43 But for Karamoja, Cunningham in a book published only jnst,
after the turn of the century records that there are also good crops of potatoes
mtama, wimbi and beans. Indian corn does not thrive."44 Little can be said about
the cultivation of maize in the remaining districts of Uganda. Cunningham
listed it amongst the crops grown in Busoga4s as also does Scott-Elliot.46 Roscoe
writing of the crops cultivated by the Gisu, Sebei, Kiga and Konjo includes
maize in each case but this refers to 1920.47 In the West Nile Stigand mentions
maize on what is now the Sudan border in 1914.48 Likewise Driberg mentions
it as one of the crops grown by the Lango but this again refers to 192049. Signi-
ficantly, though, the standard book on the Teso, even for a much later date,
does not mention maize,50 though the Iteso do have a word for it in their language.
It is not intended here to speculate upon the mode of introduction into
Uganda. This has already been the subject of lengthy debate in the pages of
the Uganda Journal. This was initiated by an article by A. C. A. Wright who
220 UGANDA JOURNAL
hypothesised on the basis of linguistic evidence that "maize reached Bunyoro
and the Lake Victoria region from Portuguese and Turkish sources through
Ethiopia via the Amhara, the Macha Galla, the Anuak and the Acholi."51
Though this is disputed by Father Muratori, who both in a letter quoted by
Wright and in a separate note, dated the arrival of maize in the southern Sudan
at 1870 and attributed its introduction to Arab influence.52 Wright incidentally
quoted the view of Ham Mukasa that maize had been known in Uganda for
over a century,53 and Wright also argues that maize may have reached Acholi
before it reached certain parts of south Sudan. This paper of Wright in 1949
set in train a lengthy discussion until 1956 in the pages of this journal and
covered the introduction of maize in the whole of eastern Africa. The object
of this present note is merely to survey the literature concerning the presence
of maize in Uganda in the nineteenth century as a basis upon which others
1. M. P. Miracle. The introduction and spread of maize in Africa. Journal of African
History, 6, no. 1, 1965, pp. 39-55.
2. J. A. Grant in the appendix to J. H. Speke, Journal of the discovery of the source of
the Nile, 1864, p. 651.
3. The source for this quotation is not given precisely by Miracle but is acknowledged as a
reference by somebody else and said to be made in a letter to the Foreign Office.
The reference is in fact a published Government Report, Great Britain, Africa No. 7
(1901) Cd 671, Report by His Majesty's Special Commissioner on the Protectorate
of Uganda. p. 10.
4. That this is the interpretation to be given to this reference is enforced by the fact that
the report of Johnston made in the previous year which contains a long account of
what was in fact grown does not contain any reference at all to maize. H. H. Johnston,
Great Britain, Africa No. 6 (1900) Cd 256, Preliminary Report by Her Majesty's
Special Commissioner on the Protectorate of Uganda.
5. J. D. Tothill (Editor). Agriculture in Uganda, 1940, especially pp. 149-150 by C. G.
6. D. N. McMaster. A subsistence crop geography of Uganda, 1962, pp. 60-63.
7. Tothill, op. cit., p. 149.
8. H. H. Johnston. The Uganda Protectorate, Vol. II, 1902, p. 673.
9. op. cit., II p. 776.
10. op. cit., I p. 290.
11. J. A. Grant. A walk across Africa, 1874, p. 237.
12. op. cit. p. 321.
13. H. M. Stanley. Through the dark continent, Vol. I, 1879, p. 383.
14. C. T. Wilson and R. W. Felkin. Uganda and the Egyptian Sudan, Vol. I, 1882, p. 159
in the portion written by C. T. Wilson.
15. G. Schweinfurth (Editor). Emin Pasha in Central Africa, 1888, p. 34.
16. op. cit. pp. 30, 38, 44 and 45.
17. F. D. Lugard. The rise of our East African empire, Vol. II, 1893, pp. 131-132.
18. L. Decle. Three years in savage Africa, 1898, p. 450.
19. G. Schweinfurth, op. cit. p. 55.
20. op. cit. p. 20.
21. op. cit. p. 137 and again on p. 281 for 1880 and for nearby Deang on p. 282.
22. op. cit. p. 164.
23. G. Casati. Ten years in Equatoria, Vol. II, 1891, p. 38-39.
24. A. B. Thruston, African incidents, 1900, p. 136.
25. C. F. S. Vandeleur. Two years travel in Uganda, Unyoro. and on the upper Nile, Geo-
graphical Journal, 9, April, 1897, p. 372.
26. H. M. Stanley, op. cit., Vol. I, p. 439.
27. H. M. Stanley. In darkest Africa, Vol. II, 1890, p. 326 and at Katwe p. 314.
28. C. Delm6-Radcliffe. Surveys and studies in Uganda. Geographical Journal, 26, no. 6,
December, 1905., p. 626. "Westwards" includes Kigezi, p. 623.
29. E. S. Grogan. From the Cape to Cairo, 1900, p. 221.
30. G. F. Scott Elliot. A naturalist in mid-Africa, 1896, p. 152.
31. A. B. Lloyd. In dwarfland and cannibal country, 1899, p. 198.
32. R. Fisher. On the borders ofpigmyland, pp. 72, 115 and 168 with reference to 1900-1904.
33. G. Schweinfurth, op. cit., p. 149.
34. H. M. Stanley. In darkest Africa, Vol. II, 1890, p. 247, and also p. 203.
35. T. H. Parke. My personal experiences in equatorial Africa, 1891, p. 371 and a further
reference to maize on p. 422.
36. A. J. Mounteney-Jephson. Emin Pasha, the rebellion at the equator, 1890, p. 22, with at
least one other reference to maize in this area on p. 8.
37. G. Schweinfurth, op. cit. p. 288 for Fachora and Modo immediately north of the
present Atura in Lango District. There is also an earlier reference to maize in
Acholi in S. W. Baker, Ismailia, Vol. II, 1874, p. 128.
38. G. Schweitzer. The life of Emin Pasha, Vol. I, 1898, p. 86.
39. A. J. Mounteney-Jephson, op. cit. p. 84.
40. G. Schweinfurth, op. cit., p. 418.
41. A. L. Kitching. On the Backwaters of the Nile, 1912, p. 110 on the basis of his visit to
Acholi in 1904.
42. C. Delm6-Radcliffe. Surveys and studies in Uganda, Geographical Journal, 26. no. 5,
November 1905, p. 489.
43. H. H. Austin. With Macdonald in Uganda, 1903, p. 74
44. J. F. Cunningham. Uganda and Its peoples, 1905, p. 345.
45. J. F. Cunningham, op. cit., p. 110.
46. G. F. Scott Elliot, op. cit. p. 397.
47. J. Roscoe, The Bagisu, 1924, pp. 7 and 15 (Gisu), 57 (Sebei), 169 (Kiga) and 139 (Konjo).
48. C. H. Stigand, Equatoria, the Lado enclave, 1923, p. 73.
49. J. H. Driberg, The Lango, 1923, p. 97.
50. J. C. D. Lawrence, The Iteso, 1957.
51. A. C. A. Wright, Maize names as indicators of economic contacts, Uganda Journal, 17,
no. 1, March, 1949, p. 69.
52. Father C. Muratori quoted in A.C.A. Wright op. cit., p. 75-78 and Maize names and
history: a further discussion, Uganda Journal, 16, no. 1, p. 81.
53. A. C. A. Wright, op. cit., p. 76.
MAP OF THE TRENCHES.
SCALE IOOYDS TO I INCH. I 3,600
(same as Bisset& map)
-3 A' g
A4L.BIZIATREE */ V
rS9ALAMITES TREE a
-- =PATH 0
---- =CAUSEWAY ACROSS TRENCH
SECTION OF THE TRENCH &LOCATION OF SAMPLING PITS-
feet 60 so 40 30 20 10 9
S---dislal growd surfact exposed in pitr.
PIT1 .-... Proloblc oriqinal profile of tranctn-
Uganda Journal, 29, 2, (1965) pp 223-224
ENSA ZA KATEBOHA
Ancient earthworks and waterholes at Karwata, Bunyoro.
By H. A. OSMASTON.
There are several sets of ancient earthworks in western Uganda, tradition-
ally associated with the Abacwezi, of which those at Bigo have been described
The earthworks at Karwata in Bunyoro, known locally as Ensa za Kateboha
(Kateboha's Trenches) have been mentioned by several writers but the only
descriptions and maps have had a very limited circulation; this is an attempt to
collate existing work and to provide a basis for more detailed investigation.
The first record of their existence seems to have been by Wayland (1934),2
reporting information supplied by D. N. Stafford of Hoima. They were sur-
veyed by Bisset in 19383 and again roughly by Lanning in 1952 who also pre-
pared a brief typescript report.4 In 1959 they were visited by the writer who
included a map and brief description in the working plan for the Bugoma Forest
The trenches lie at latitude 10 14': N, Long 31002': E, close to the road from
Kabwoya to Kahuka and 3.3 miles from the road junction at the former place.
After passing through a narrow belt of scrub on the north side of the road, the
way crosses some grassland, a narrow strip of forest, and a smaller patch of
grassland until the trenches are found in the next stretch of forest. They consist
of one main trench about 1,600 feet long, forming two sides of a rough rect-
angle, with a short detached trench at one end. At points 7, 8 & 10 on the
map the trench is interrupted by narrow causeways apparently part of the
original plan, but at a few other points it is crossed now by paths apparently
made later by cattle or wild animals. The trenches are at present 15-20 feet
wide and about 6 feet deep, with banks of spoil on each side, but pits dug
across one sample section showed that the bottom of the ditch originally reach-
ed 8 feet below ground level, and that the banks may have stood 4 feet above it.
Pottery shards and small fragments of charcoal occurred in these pits. There is
no obvious sign of any trench completing the perimeter. There is a possible
indication on the air photographs of an old track leading southwards, but the
trenches themselves cannot be distinguished.
The forest that now covers the area includes large trees of over 6 ft. girth
(probably 50-150 years old) growing in and near the trenches and clearly young-
er than them. Besides providing a minimum age for the earthworks, this shows
that since their abandonment conditions of climate, or of reduced interference
by man and cattle, have favoured the extension of the forest, which has been
mapped and described by Cahusac and Osmaston.6 The trenches actually lie in
an area of colonising forest in its second stage comprising various trees and
shrubs characteristic of young forest such as Sapium ellipticum an Alchornea
sp. Enclosed within the trenches is a patch of small-crowned mixed forest in
224 UGANDA JOURNAL
which Albizia spp., Celtis spp. and Balanites wilsoniana are common. Nearby
are two loug arms of forest of Cynometra alexandri (Uganda ironwood).
Although the status of the colonising forest is clear, the relative status of the
other two communities is not. Eggeling (1947) considered that in the neighbour-
ing Budongo forest the areas of Cynometra forest were the oldest, and that
mixed forest was a stage in its development.7 If this is so it is possible that when
the trenches were dug they enclosed an area of grassland between them and
what was then the forest edge, marked now by the edge of the Cynometra
forest. Unfortunately the detailed measurements collected in the Bugoma
forest throw some doubt on this view of the relationship between the types of
forest and suggest that the vegetational differences may be determined by
differences in the soil or by other causes. Archaeological study here may be
able to throw light on this problem.
Another feature of the neighbourhood which may be connected with the
origin of the trenches is the number of waterholes which occur on broad ridges
in the grasslands, there being one about three quarters of a mile west of the
trenches. They occupy hollows up to 500 feet across and 15 feet deep, and in
the bottoms of which pools of water collect during the rains and frequently
last long into the dry season. Although the formation of similar pans in the
drier and flatter parts of Africa has been ascribed to a combination of wind
erosion and the removal of mud on the legs of animals which drink there8 it is
difficult in view of the topography and present climate of this area to ascribe
their initial formation to these causes. It is possible that they were originally
dug by a cattle-owning tribe and have been enlarged by the subsequent acti-
vities of cattle, or of elephant and buffalo which are abundant there. Although
the country appears well watered, there are in fact few permanent streams, and
in the dry season drinking water can be a serious problem. Bisset (1937) has
discussed similar hollows in Masaka district.9
Mr. B. Kirwan has suggested to me that the name Kateboha might be sym-
bolic rather than that of a real man; being derived from the Lunyoro verb
'okuboha', (to fasten or hold back) with addition of a negative prefix. Ensa is
the plural of orusa, a trench or embankment. The reference might therefore be
to a failure to complete or to defend the trenches.10
1. M. Posnansky. Pottery types from archaeological sites in East Africa Jl. of African
History, 2, no. 2, 1961, pp. 177-198.
2. E. J. Wayland, Biggo bya Mugenyi. Uganda J. 2, no. 1, 1934, p. 25.
3. C. B. Bisset, Geological Survey Department, Map no. 415, 1940.
4. E. C. Lanning, Ancient earthworks in western Uganda. Uganda J., 13, no. 1, 1953,
5. H. A. Osmaston, Working plan for the Bugoma forest. Entebbe, Uganda Forest Depart-
6. Ibid, plate 1, pp. 12-16.
7. W. J. Eggeling, Observations on the ecology of the Budongo rain forest. Jl., of Ecology,
74, no. 1, 1947, pp. 20-87.
8. G. E. Hutchinson, A treatise on limnology, Vol. 1, New Yorkd, Wiley, p. 1015.
9. C. B. Bisset, Hilltop hollows in Masaka district. Uganda J., 5, no. 2, 1937,
10. The name Kateboha is discussed to greater length by E. C. Lanning, The daath of chief-
tain Kateboha. Uganda J., 27, no, 2, 1959, pp. 186-188.
Uganda Journal, 29, 2, (1965) pp. 225-228
HILL-TOP HOLLOWS-FURTHER INVESTIGATIONS
By M. J. PATZ
As part of a larger study of the landforms of Buganda, a detailed study was
made of one laterite capped hill near Lugazi, and in particular of an enclosed
hollow on its summit. These hollows pose a problem in their mode of origin.
Wayland' has described man-made hollows on laterite hill-tops, and the pur-
poses for which they were made. Later, Bisset2 further described these features
in Maska district, and stressed the cultural significance of these hill-tops in the
not too distant past. Pottery fragments, cowrie shells, hearth stones and the
like may often be found in association with them. The purposes for which
certain oval holes had been dug are by no means clear; but even more puzzling
are the larger hollows and depressions which appear to be natural features of
these flat-topped hills. These natural, entirely enclosed hollows, are of relatively
frequent occurrence, and, as would be indicated by this recent work, were also
of cultural significance, though of an earlier period, than that of the Iron Age.
The hollow in question is situated on the highest part of Nakawungu hill,
south of Lugazi. The hill is over 4,400 feet high and is capped by a lateritic
ironstone which slopes gently northeastwards. The depression is oval in shape,
with a length of 230 yards and a breadth of 160 yards. The sides dip most steeply
where the oval is narrowest and here the rim is over 18 feet above the base of the
hollow. At its lowest, the rim is only 5 feet above this base. It is nevertheless
entirely enclosed and as there is up to five feet of infill, must have been even
more pronounced as a hollow when it was occupied by Stone Age Man.
A trench, 12 feet by 6 feet was dug at the base of the depression. This passed
through 21 feet of unindurated material, (fig.1) consisting of a few inches of
dark, humus-rich top-soil, overlying a soft, red soil, in which manganese pisoli-
ths appear at about 12 inches and increase in frequency with depth. The lower
portion is predominantly a pisolith sheet. Below this, and to a depth of 5 feet,
the infill is indurated, by iron and managanese cementation; and this rests on an
older, relatively impervious laterite. the indurated zone can be divided into two:
an upper section, from 21 to 3 feet, and a lower from 3 feet to 5 feet. These
respectively become increasingly rich in manganese with depth and attain a
Two artifact horizons were revealed. The upper is in the unindurated piso-
lith sheet below the red soil, and the lower is contained in the indurated infill
from 21 to 3 feet. Below this there were also pebble inclusions in a sufficiently
significant quantity to invite speculation as to their origin; but there was
little of diagnostic value and consequently only two artifact assemblages as
such are recognized. In order to obtain a sufficiently large collection for dia-
gnostic purposes, the area of the trench was doubled and excavated through
the two artifact horizons.
The earlier of the two assemblages contains a higher proportion of larger
flakes and implements. The material used was predominantly medium grained
226 UGANDA JOURNAL
quartzite. The tools were fresh, in that they had not been rolled, although they
may have moved a short distance from the periphery towards the centre of the
hollow. They have been considerably affected by the Iron-rich solutions which
drain into the hollow. Much of the coarser grained material had become 'sug-
ary', the grains having separated. The finer grained material was in a better
state of preservation, but all had acquired a heavy 'giraffe' patination of iron
and manganese, where pisoliths had formed adjacent to them. Despite this a
sufficiently large assemblage was retrieved to be able to attribute a date to the
The assemblage includes two well made hand-axes of fine grained quartzite.
The smaller of the two measures 11 cms. in length and 7J in width. It has an
extremely bulbous profile (fig. 2. 1.) with a maximum thickness of 51 cms.,
and tapers very sharply to the point. The larger hand-axe, of which unfortunately
the point is missing, is a much more slender tool (fig. 2.2.) and has been fashion-
ed from a flake. Its present length is 11 cms., its width 8 cms. and its maximum
thickness only 31 cms. Both have been step-flaked and so diagnostic of late
Acheulian or Sangoan artifacts (40,000 B.P.). Except for the assemblage from
near Muganga on the Buganda shore of the Nile near Jinja and apart from the
single specimen found on Kikaya Hill by Mr. Collison in 1950, these are the
first well authenticated hand-axes to be found in Buganda. Also among the
collection is a 'limace-like' piece (fig. 2.3) similarly made of a fine grained
quartzite. It is 91 cms. long, 41 wide and 21 thick. This tool appears to be
particularly well represented in collections in Uganda, and has also been
found by the recent expedition to Lolui Island,3 and in the Nsongezi erosion
gullies. The only other diagnostic implement is what appears to be a broken
lance. As a whole, the material bears marked affinities to the collection found at
the Nsongezi erosion gullies in the Orichinga valley.4 It is characteristically
early Middle Stone Age and assignable to the Sangoan, a culture which takes it
name from the Sango Hills where it was first found, though it does not include
the picks and heavy core scrapers of the classic Nsongezi assemblages of the
The second and younger artifact assemblage comes from the unindurated
pisolith sheet. It is characterized by smaller tools, predominantly quartz. (fig. 2:
4 and 5). There is no patination. Unlike the older assemblage these are un-
weathered and extremely fresh. It includes quadrilateral flakes which are
typical of the Middle Stone Age, possibly Stillbay-Magosian industries
(20000-15000 B.P.). With these tools were found charcoal fragments for which
a possible radio-carbon date may be available later. The inclusion of dark
humus soil nodules within this unindurated profile at various depths suggests
that there has been considerable re-working by termite activity and the present
position of the charcoal with these artifacts may not be original.
Thus, for more than one period during the Middle to Late Stone Age, this
enclosed hollow provided a suitable environment for early man. Further in-
vestigation of such similar sites may prove fruitful.
1. E. J. Wayland, Notes on the Biggo bya Mugenyi, Uganda Journal, 2, 1934 p. 21.
2. C. B. Bisset Hill-top hollows in Masaka District, Uganda Journal, 5. 1937, p. 130
3. M. Posnansky. Interim report of the expedition to Lolui Island. (unpublished 1965)
4. G. H. Cole, Recent archaeological work in southern Uganda. Uganda Journal, 29,
1965, pp. 149-161 of this issue.
1)4O LL .O
Imv IV (fL9.1)
Figure 2 Artifacts from Nakawungo Hill Nos. 1, 2 and 3 Acheuvlian. No. 4 and 5 Still Ba