Anderson, Ross (2001) World War I in East Africa. PhD thesis.

http://theses.gla.ac.uk/5195/

Copyright and moral rights for this thesis are retained by the author
A copy can be downloaded for personal non-commercial research or
study, without prior permission or charge
This thesis cannot be reproduced or quoted extensively from without first
obtaining permission in writing from the Author
The content must not be changed in any way or sold commercially in any
format or medium without the formal permission of the Author
When referring to this work, full bibliographic details including the
author, title, awarding institution and date of the thesis must be given

Glasgow Theses Service

http://theses.gla.ac.uk/
theses@gla.ac.uk

WORLD WAR I IN EAST AFRICA

1916-1918

RossAnderson
Octobcr 2001

Submittedto the Departmentof Modem History, University of Glasgow in partial

fulfilment for the degreeof Doctor of Philosophy

C RossAnderson October 2001

ABSTRACT
The campaignin East Africa is one of the lesser-knownand understoodtheatresof the
First World War. Strategically unimportant to the main war effort, East Africa has
remained in the shadows of the much better known Western, Mesopotamian and
PalestineFronts. Despite this, the campaign lasted from August 1914 to November
1918and covereda large portion of East Africa as well as drawing in the majority of
its population. Scholarly works have been greatly outnumberedby popular accounts
and the final two years, 1916 to 1918,remain vague and contradictory. Nevertheless,
a greatdeal of valuable primary material exists in various archives and it is the aim of
this dissertationto describeand analysethe military operationsof this period.
At the outbreak of war, the imperial powers in East Africa were unprepared for a
major campaign. Although the colonies possessedlittle strategicvalue in themselves,
the dynamicsof imperial rivalry quickly generatedarmed conflict. The East African
campaign evolved haphazardly from neutralising German wireless communications
andnaval facilities to a wildly over-ambitiousplan to conquerthe whole of the colony
with scantforces. The British wanted to keep any potential spoils for themselves,but
were also strongly influenced by the expansionist policies of South Africa, largely
propoundedby Louis Botha and Jan Smuts.
By September1916, the British forces, commandedby Smuts, had occupied the bulk
of German East Africa with all the railways, towns and ports in their possession.
However, he had failed to bring the GermanSchuatruppe to battle and it remained a
powerful and well-motivated force. Furthermore, his reliance to manoeuvre and
reluctance to fight battles led his troops ever-deeper into enemy territory and
dependenton inadequatelines of communication.
Smuts continued his advanceuntil January 1917 when he left for the Imperial War
Conference. His forces were in terrible condition and unfit for further offensive
operations. He was succeededby the British GeneralHoskins for a bare three months,
but, who neverthelessinstigated badly neededreforms and reorganisation. In May
1917,the South African, General, van Deventer assumedcommand, an appointment
that he would hold until the end of the war. Vah Deventer continued to build on
Hoskins's work while instigating an aggressive policy of fighting hard battles
wheneverpossible, while concurrently trying to destroy German food supplies. These
methodswere continued throughout the remainder of 1917 and until November 1918
when the war endedwith the Schuatruppe being pursuedfrom PortugueseEast Africa
into Northern Rhodesia.

For both sides, the campaignwas dominatednot by heavy fighting, but by the
questionsof health and supply. The levels of sickness,particularlymalaria,were
many times worse than other theatresand constantlyhinderedmilitary operations.
The provisionand distributionof food and other supplieswas an enormousproblem
that was only partially solved by the widespreaduse of motor vehicles and road
construction.For the British, relationswith their BelgianandPortuguese
allies were
neversmoothasimperialrivalriesoftencreatedfriction andmisunderstanding.
In the
'
end,the East African campaignwas one of mobility
and evasionand quite unlike
foughtin Europeandthe Middle East.
campaigns

ACKNOWLEDGEMENTS
I wish to acknowledge the excellent assistance and guidance provided by my
supervisor,ProfessorHew Strachan.
Crown copyright material is reproduced with the permission of Her Majesty's
StationeryController.

ii

LIST OF ILLUSTRATIONS
MAP
I
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28_
29

PAGE
TITLE
8a
The StrategicSituation at the Outbreakof War 1914
26a
The StrategicSituation at end September1914
28a
GermanEast Africa
35a
The StrategicSituation I" October 1916
49a
The Situation in the West, September- October 1916
55a
The Kilwa Area, September- October 1916
OperationsAgainst the German Western Forces, 11 - 29 58a
October 1916
71a
OperationsNorth of the Rufiji
Operations Around Kibata, November 1916 - January 73a
1917
80a
The Rufiji Area: GeneralSmuts's Plan for the Offensive
The Advance to the Rufiji, 31 December 1916 - 15 82a
January1917
OperationsNorth of the Rufiji, December 1916- January 89a
1917
Operations of the 2ndDivision, 20 December 1916 - 10 94a
Janu 1917
129a
Operations of Northey's Force, 19 December 1916 -2
February 1917
Operations in the South and West, February - March 132a
1917
159a
2ndDivision Lines of Communication, September1916
163a
GermanLines of Communication, September1916
222a
The Operational Situation, June 1917
229a
The Wintgens/NaumannRaid, February- October 1917
Operationsat Lindi, June 1917
231a
Operationsof Hanforce and O'Grady, July 1917
237a
Van Deventer's Plan, August 1917
244a
Northey's Situation, I" Week July 1917
248a
Hanforce and O'Grady, September- October 1917
256a
Lettow's Escape
277a
GermanEntry into PortugueseEast Africa, 1918
282a
British Counter-Moves,January- February 1918
292a
Operations,PortugueseEast Africa, April - May 1918
302a
Operations, Portuguese East Africa, June - September 308a
1918

iii

TABLE OF CONTENTS

CHAPTER
TITLE
I
Introduction
2
The Situation at the End of September1916
3
The Operationsof September1916to January1917
4
Reorganisationand the Operationsof Februaryto May 1917
5
Supply and Transportation
6
Diseaseand Medicine
7
The Operationsof Juneto December1917
8
The Operationsof 1918
9
Inter-Allied Cooperation
10
Conclusions
T Bibliography
11 -------

iv

PAGES
1-20
21-40
41-100
101-144
145-165
166-215
216-280
281-319
320-340
341-345
346-37

CHAPTER I- INTRODUCTION

The First World War has been the subject of intense description and analysis for
over eighty years.

Given the scale of operations there, the Western Front has

understandablyreceived the bulk of English language attention, although recently, the

secondarytheatresof Mesopotamia,Gallipoli and Palestinehave come to be considered
in much greaterdetail. The opening of archives and increasedaccessto hitherto private
papershave enabled scholars to develop their understandingof the conflict while also
disposingof a number of popular myths enroute. The backgroundto the formulation and
the executionof grand strategy,operationsand tactics has been examinedunder a critical
andlessnationalistic light with very interestingresults.
It is true that comparedto the Western Front, the campaign in East Africa was
very small scale and strategically unimportant. British troop strengths there peaked at
58,000 in August 1916 while total wartime casualties of 349,311, of whom nearly
330,000were sick, representedsome 5.62 per cent of the Western Front's and 3.15 per
'
British
Empire's
losses.
Yet it is easy to become fixated on numbers
total
cent of the
aloneand it is worth recalling that the war there lasted from August 1914until November
1918,covering a huge swatheof Africa. It ranged from the modem statesof Kenya and
Ugandain the north, through the Congo, Ruanda,Burundi, and Tanzania in the centre,to
Zambia, Malawi and Mozambique in the south. Few inhabitants, Europeanor African,
escapedits effects or ravages,while the colonial empires were irrevocably changedby
the conflict. It is worth noting too that the casualtiessuffered were comparableto those
incurred in the Anglo-Boer War that had endedjust twelve years previously.2 Perhaps
insignificant in global terms, the war was of overwhelming local consequence.

1WO 33/ 858,European War Telegrams,SeriesD, Volume1,29' January 1915 29h February 1917, London: War Office, 1918.
HenceforthTelegramsD 1. No. 1235,TelegramN 1394GeneralHeadquartersto War Office, 14 September1916,pp. 327-329;
Mitchell, Major TJ and Smith, G M, Official History of the War, Medical Services,Casualtiesand Medical Statistics, London:
HMSO, 1931, reprintedby Imperial War Museum and Battery Press,[n.d.]. Henceforth, Official History Medical Statistics. See
Table I "Approximate Total Casualtiesin British Expeditionary Forcesduring the Great War", p. 12. East African troop casualties
reached349,311over the courseof the war as comparedto 6,218,540sufferedon the WesternFront and 11,096,338throughout the
British forcesworld-wide.
2Official History Medical Statistics, SeeTables 2,3
and 4 on pp. 253-254 for East Africa and Tables 2,3 and 4 on p. 269 for South
Africa. Casualtiesamongstthe troops are fewer in the EastAfrican
campaignthan the South African, but the inclusion of the
followers' figures for the former reversesthe
situation.

HISTORIOGRAPHICAL BACKGROUND

Over the past forty years, it has been popular accounts rather than academic
studiesthat have provided the bulk of the limited coverageof the campaign. It has been
only very recently that the scholarly neglect has been rectified by the publication of the
first volume of Hew Strachan's The First World War which devotes a substantial and
informative chapterto Africa. 3 Apart from this excellent and much-neededwork, which
is a wide-rangingsynthesisof the existing literature in English, Germanand French,most
attentionhas beendirected onto detailed studiesof subjectssuch as labour, medicine and
transport.

The question of strategy and operational aims has never been covered in great
depth,perhapsowing to the theatre's lack of strategicimportance. Certainly, the question
of why the campaign was actually fought requires greater examination as the politics
involved were complex and involved many players outside of London. However, David
French has produced two useful studies of British decision making, namely British
Strategyand War Aims 1914 1916 and TheStrategyofthe Lloyd George Coalition.4 At
present,there is nothing of similar value concernedspecifically with East Africa.
This is not from want of resourcesas a vast amount of primary and secondary
material exists in several languages,but it has not been fully exploited in the published
official histories. Study has not been helped by the relative paucity of reliable secondary
information in English. In English, the author of the authoritative Official History died
beforethe secondand concluding volume could be finished and the period of late 1916to
the war's end in November 1918 remains vague and contradictory.5 On the other hand,
the Belgians produced a detailed and thorough history of their own forces' operations.
3 Strachan,Hew, TheFirst World War: ToArms, Volume 1,Oxford: Oxford University Press,2001,
pp. 495-643.
4 French,David, British Strategy
and War Aims 1914-1916, Unwin and Allen, London, 1986; and The Strategy of the Lloyd George
Coalition, 1916-1918, Oxford: Clarendon Press, 1995;
other earlier, but still useful sourcesinclude Gooch, John, The Plans of Wor:
The General Staff and British Military Strategy 1900-1916, London: Routledge & Kegan Paul, 1974; Rothwell, VF British War
c
Aims and PeaceDiplomacy 1914-1918,Oxford, 1971; and Guinn, Peter,British Strategy
and Politics 1914-1918, Oxford: Clarendon
Press,1965. The German view is consideredin Louis, W R, Great Britain
and Germany's Lost Colonies, Oxford: Clarendon Press,
1967.
5 Hordem, LieutenantColonel Charles,History
of the Great War: Militapy OperationsEast Africa: August 1914- September1916,
Volume 1, London: HMSO, 1941,
reprintedNashville, Tennessee:The Battery Press,1990. Henceforth, Hordern, Military

Thesevolumes are well-documentedand possessexcellent maps, with good background

direct
Belgian
But,
Portuguese
British
as
operations.
and
coverage of concurrent
the
it
is
to
late
in
1917,
in
corrective
the
a
partial
only
ended
participation
campaign
deficiency.6 It appears that the Portuguese Government never produced an official
forces
largely
their
of
the
unsuccessful,operations
account of
considerable, although
during the campaign.
On the German side the coverageis variable, with the naval history providing a
land
defence
information
the
deal
the
the
operations
colony
while
of
on
great
of useful
7

from
detail.
Apart
the narrative memoir
disappointingly
thin
on
short
and
volumes are
largely
in
books
the
years
aimed at chronicling
post-war
emerged
accounts,a number of
from
indignation
debate
justifying
Much
German
the
at the victors'
resulted
of
rule.
and
hold
It
German
to
colonies.
seemsthat the
unfitness
alleged
accusationsof
atrocities and
in
further
interest
War
World
d6bacle
Second
to
Nazism
the
the
put paid
of
rise of
and
the subject.In recent years,secondaryworks emergedmainly from the former GDF, and,
although slanted against the Wilhelmine and capitalist regimes, they contained some
8
insights.
useful
Much of the historiographical deficiency can be traced to the fact that the bulk of
all modem English languageaccountsare based on the version of events presentedby
General von Lettow-Vorbeck on the German side and Colonel Meinertzhagen on the
British.9 While both were first-hand participants and keen observers,neither could be
describedas impartial and each had his own domestic points to score. The former's
OperationsEast Africa. A partial draft of Volume II can be found in the Public RecordOffice (PRO), London in the CAB 44 series.
A numberof mapsand sketchesfor the plannedvolume are containedin the CAB 45 and WO 808 series.
6 Royaumede Belgique, Les CampagnesColoniales Belges 1914 1918, TomesI-3, Bruxelles: Imprimerie Typographiquede
11C.M., 1927-1932.
7The naval works interest include Assmann,Kurt Konteradmiral, Der Krieg
of
zur See1914- 1918: Die Kampfe der Kaiserlichen
Marine in den DeutschenKolonien, Berlin: Verlag ES Mittler & Sohn, 1935;Raederand Manley, Der Kreuzerkrieg in den
ausldndischesGewdssern,3 Volumes, Berlin: Verlag ES Mittler & Sohn, 1922-1937. The military works are only bare summariesof
the known situation and are much lessvaluable. SeeDer Wel1krieg1914 bis 1918:Die Militarischen Operationenzu Lande: 9. Band,
"Der Krieg in den Kolonien bis Ende 1915. Die Ereignissein Deutsche-Ostafrikabis zum Ende desJahres 1915", Berlin: Verlag ES
Mittler und Sohn,pp. 480485 with Skizze28; also 13. Band "Der Krieg in Ostaftika seit 1916", pp. 452462.
11Stoecker,Helmuth (ed), German Imperialism in Aftica. From
the beginnings until the Second World War, Transt Bemd Zoellner,
London:C Hurst & Co, 1986(originally publishedin DDR 1977);Helbig, Klaus, Legendeund Wahrheit.-Der erste Welikriegin 0stafrika
und Rolle desGeneralsLettow-Vorbeck,PhD Thesis,Leipzig: Karl-Marx University, 1968.
9 Lettow-Vorbeck, General Paul
von, My Reminiscencesof East Africa, London: Hurst and Blackett, 1920,reprinted Nashville,
Tennessee:The Battery Press,[n.d.j. Henceforth,Lettow, Reminiscences,
Meinertzhagen,Colonel Richard,Army Diary 1899-1926,
London:Oliver and Boyd, 1960.

from
largely
it
detail
from
lack
memory and
of precise
as was written
accountsuffers
a
has a distinct whiff of hindsight, particularly when dealing with tactical or operational
decisions.10 The esteem accorded by his former enemies to the commander of the
Schutztruppehas given his work an undeservedaura of authoritativeness;that it was
its
increased
in
English
German
attractiveness
the
greatly
almost
sole
account produced
to monoglot authors. Von Lettow was the decisive personality on the German side and
The
for
the
the
work seemsquite reliable when
campaign.
of
strategy
was responsible
dealingwith von Lettow's own experiences,but its accuracysuffers wheneverdiscussing
his
immediate
from
his
and
often glib comments
presence
eventsand personalitiesaway
do not always standwhen set alongsidethe surviving documents. Provided that it is used
in conjunction with other sources, it does, however, remain an important memoir of
considerablevalue.
On the British side, Meinertzhagen'saccountof his time as an intelligence officer
in General Headquartersprovides an insider's view that is both immediate and highly
high
It
the
the
the
command.
also
especially
of
and
most
critical of
campaign,
conduct of
provides a number of charactersketchesand battle descriptions that are of considerable
value. However, doubts have been raised about the veracity of all of Meinertzhagen's
"
recollectionsand whether the diary was "corrected" after the event. He was a talented
soldier, but an unusual and difficult personality who was keen to present his version of
events. Again, his coverageof events is uneven and ends in December 1916, when he
was evacuatedfrom the theatre on medical grounds. Like von Lettow, Meinertzhagenis
a useful guide, but should not be usedin isolation.
A number of popular accountshave been written in the last 30 years, but most of
them suffer from the undue reliance on Lettow and Meinertzhagen,and detailed coverage
of the period beyond January 1917 is virtually non-existent. Charles Miller's book, The
Battle for the Bundu, is often cited as being the standard popular account of the
10Lettow, Reminiscences,
p. vii. In the prefaceto his work, written soon after the war's end, von Lettow frankly admitted that the loss
of his records,lack of time and inability to consult former colleaguesmade errors of detail inevitable.
" Amongst the reasonsfor this
scholarly suspicion is the fact that Meinertzhagen'snumerousvolumes of diaries are composedof
typed, ring-bound pagesinterleavedwith captionedphotographs. It would have been impossibleto producesuch an immaculateand
error-freework under wartime field conditions and the possibility of subsequentamendmentscannotbe discounted. His book, Army
Diary, reproducesselecteddiary entries almost
word for word.

campaign,but it suffers from a number of drawbacks, most notably the lack of any
in
list
fighting.
While
drawing
the
consideredanalysis of
of sources,
on an extensive
both English and German,the book tends to accept statementsuncritically and generally
ignores the importance of supply and health as vital factors. Several other popular
accountsappearedin the same period, although none could be said to have advanced
knowledgeof the campaignsignificantly.12 As a genre,thesebooks are largely confined
to narrativeaccountsbasedlargely on the printed literature in English and make no use of
archival sources. In a number of cases,factual errors and a number of myths get recycled
without seriousquestion.
This is unfortunate as a great deal of very useful material exists both in print and
in various national repositories. From the Germanperspective,probably the most useful
is the semi-official account, Die Operationen in Ostafrika, produced by Major Ludwig
Boell, a participant in the campaign and later an historian in the Reichsarchiv.13 This
clear and detailed work describesthe Germanwar effort from start to finish, giving much
valuable information on plans and intentions together with the detail of operations. The
composition of detachmentsor Abteilungen is laid out as are casualty figures and the
determinationof successor failure. It is written with a degreeof detachmentand analysis
that makesit invaluable as a source. Sadly, a large number of sketchesintended for the
volume were destroyedby Allied bombing, although severaluseftil, large-scalemaps are
provided.
Equally valuable as a primary source is Governor Heinrich Schnee's book,
Deutsch-Ostafrikaim Weltkrieg, written immediately after the war's end.14 Schneekept
a careful set of diaries and was a perceptiveobserver. Although not a professionalsoldier
himself, in many ways his accountof the campaignis clearer and more accuratethan von
Lettow's. Certainly, it covers important, but frequently neglected areas, such as the
medical system, supply and transport, civil governanceand war finance in some detail.
12Mosley, Leonard, Duel for Kilimanjaro, London: Weidenfeld & Nicolson, 1963, Sibley,
Maj J R, Tanganyikan Guerrilla,
Pan/BallantyneIllustrated History of the First World War, Book No 4, London: Pan/Ballantyne,1971; Hoyt, E P. Guerrilla: Colonel
von Lettow-Vorbeckand Germany'sEast African Empire, New York: Macmillan, 1981; James,Lawrence, Savage Wars:The British
Campaignsin Africa 1870-1920,London: 1985
Farwell, Byron The Great War in Africa, New York: WW Faber, 1986.
and
13Boell, Ludwig, Die Operationen Ostafrika, Hamburg:Walter
Dachcrt, 1951. Henceforth,Boell, Die Operationen.
in
" Schnee,Heinrich, Deutsch-aytaftika
im Welikrieg,Leipzig: Quelle et Meyer, 1919. Henceforth,Schnee,Deutsch-Ostafrika.

Centralto his story is the failure of civil-military relations that resulted in the bitter clash
betweenhimself and von Lettow over who would decide the direction and scopeof the
Deutschletzer
Gouverneur
is
later
Als
Also
colonial war effort.
work, entitled
useful a
Ostaftika Erinnerungen, in which he reconsidersthe conflict after the passageof over
15
twenty years. A synopsisof the disputetogetherwith analyis of the respectivepoints of
16
Solf
former
Wilhelm
colonial minister,
view is containedin a book review by the
.
Taken together, both Boell's and Schnee'sworks shed a rather different light on
German operations than is provided by von Lettow alone. This can be further
supplementedby the perceptive accounts of several other participants, most notably
doctors,who servedaway from the headquartersin the various Abteilungen." A number
during
inter-war
the
their
period,
memoirs
own
of commandersand soldiers produced
18
for
less
Perhaps
traditional
a
source
many of which are of some considerablevalue.
military history, but nonethelessa useful one, is the accountof the Governor's wife, Frau
Ada Schnee,as it gives clear picture, but also focuses on the economic and medical
19
forces.
fighting
The Germancolonial statehas also been little
systemsthat underlaythe
examined in the context of the war, yet the fighting effectiveness of the Schutztruppe
dependedon the civil administration for the provision of food, equipment and porters. A
very useful analysis is provided by John Illiffe, while a newer study by Juhani Koponen
helpsreadersto understandthe natureof the colonial stateand Wolfgang Eckart examines
the politics of colonial medicine.20
The Belgian contribution to the East African campaign is not fully appreciated.
Apart from the excellent official history on the Africa colonial campaigns, further
backgroundinformation on the pre-war organisationof the Force Publique is provided by

15Schnee,
Heinrich,AlsLe&erGouverneur
Heidelberg:
QuelleundMeyer,1964.
Erinnerungen,
in Deutsch-Ostafrika:
16Solf,Wilhelm,"Schnee Lettow-Vorbeck",
Nation,2 (1920),pp.87-95.
Die Deutsche
und

17Deppe,Ludwig, Mit Lettow-Vorbeckdurch Afrika, Berlin: Verlag August Scherl, 1919;Hauer, August Dr, Kumbake:ErIelmisse
einesArtzein Deutsch-Ostafrika1914-1919,Berlin: Hobbing, 1922.
" Lettow-Vorbeck,Gen Paul Emil
von, Mein Leben,publishedby Ursulavon Lettow-VoTbeck,Biberachan der Riss:Koehlers,1957;
Kohl, HauptmannFranz,Der Kampf um Deutsch-Ostafrika,Berlin: Verlag KameradschA [1919]; Methner,Wilhelm, Unter drei
Gouverneuren:16 Jahre Dienst in den deutschenTropen,Breslau:Kom, 1938.
'9 Schnee,Ada, Meine Erlebnisse
wdhrendder Kriegszeitin DeutschesOstafrika, Leipzig: Verlag Quelle und Meyer, 1918.
20Iliffe, John,Tanganyika
under GermanyRule 1905-1912,Cambridge:CambridgeUniversity Press,1969;Koponen,Juhani,
DevelopmentforExploitation: GermanColonial Policies in Mainland Tanzania,1884-1914,Helsinki: Finnish Historical Society, 1994;
Eckhardt,WolfgangU, Medizzinund KolonialimperialismusDeutschland1884-1945,Paderbom:Schoningh,1997.

Flament while WR
21
region.

Louis has written a very useful study of Belgian policies in the

The Portugueseparticipation in the war was undistinguishedby martial success.

Despitecontributing forces for over two years and having one of their most important
coloniesas a major battleground,there appearto be few published primary works on the
22
Of
down.
difficult
the secondary
do
track
to
those
that
subject and
are remarkably
by
Malyn
Missier
by
first
Ren6
the
books
two
the
second
and
sources,
stand out,
Newitt.23 P61issiercovers the war in greater detail, but both help to explain the
1918.
throughout
Germans
to
most
of
that
the
conditions
continue resistance
enabled
fate
German
determining
in
factor
the
the
key
"Sub-imperialism"
of
was a
II
from
London.
divergent
those
in
South
Africa
of
colonies as
particular pursued goals
Much can be gained from ProsserGifford's and William Roger Lewis's study, Britain
24
highlights
This
Rule.
Colonial
Germany
Rivalry
Africa:
Imperial
in
work
and
and
Of
helped
the
local
to
frictions
that
conflict.
shape
many of the
and aspirations
African
South
Failure
book,
The
is
Hyam's
Ronald
of
considerable value, too,
Expansionism,which examinesthe attemptsto incorporatemuch of southernAfrica into
25
a single state,most notably by Jan Smuts.
Overall, despite the many different studies, there can be said to be no generally
authoritative work on the entire campaign and certainly none that looks at the conflict
from more than one national perspective. This gap is not due to lack of archival material;
on the contrary large amounts remain, untapped and largely unseen in various
repositories. Britain has a huge amount of material in the Public Record Office in the
form of Cabinet Office, War Office, Admiralty, Colonial and Foreign Office files, to list
21Flament,F
et at, La Force Publique desanaissance a 1914: Participation des militaires a I'histoire despremiaresanniesdu
Congo,Brussels:Departments State
of
and Public Institutions, 1952;Louis, W R, Ruanda-Urundi 1884-1919.Oxford: Clarendon
Press,1963.
22 Pires, Antonio J, A Grand Guerra
em Macambique,

Porto: [n pub. ], 1924; and Martins, General Ferreira DSO, Portugal's

Co-

op.eration with Great Britain in the Great War 1914-1918,Lisboa: Servicosde Informacao e Imprensada EmbaixadaBritanica, 1943.
2 Pdlissier,Rend,
La Noissancedu MozambiqueRgsistanceet Revoltes,2 Volumes, Orgeval: Pdlissier, 1984; and Newitt, Malyn, A
Hlsto?Y of Mozambique,London: Hurst & Company, 1995; Isaacman,Allen, The Tradition Resistancein Mozambique,Berkeley:
of
UCLA Press,1976.
24Gifford, Prosser
and Louis, William Roger, Britain and Germany in Africa: Imperial Rivalry and Colonial Rule, New Haven: Yale
University Press,1967.

the mostprominent. The British Library, the Imperial War Museum and National Army
Museumin London also contain a great deal of relevant material as do archivesin Oxford
andCambridgeUniversities. Both the South African National Defence Force (SANDF)
DocumentationDirectorate and the South African National Archives contain a large
amountof valuable information on the war. The Belgian Musee Royal de lArmej has a
considerableamount of printed and original material relating to the campaign, while it
appearsthat a number of important military reports are in the Archivo Historic6 Militar in
Lisbon - althoughthesehave not yet beenconsulted.
1
The German records in the BundesarchivlMilitdrsarvchiv

in Freiburg contain the

terse wartime diaries of General von Lettow together with the huge account of the
26
invaluable in
by Major Boell. These

campaignwritten subsequently

are

supplementing

the limited printed sourcesand making up for the loss of most of the official colonial
records during the war. Other Bundesarchiv sources, such as the Schnee papers in
Koblenzand the Reichskolonialsamtcollection in Potsdam,promise to be useful although
time haspreventedtheir exploitation.
The greatestproblem with archival sourcesis the sheermassand weight of paper;
the British sourcesalone go well beyond the capability of one individual to searchand
examine comprehensively. These efforts must be viewed as a pioneering attempt to
locateand discussthe most important sources. It is highly likely that much of value will
continueto be discoveredover the coming years.
AIM

The aim of this dissertationis to examine the critical period in the East African
campaignfrom September1916 to November 1918. This period covers the events from
the time that the advancingAllies had reachedthe centre of GermanEast Africa, through
the Germans' evacuationof their colony in late 1917 and the subsequent
campaign in
25Hyam,Ronald,7heFailure

ofSouth African EqAansion1908-1948,London: Macmillan, 1972.

7M

SKETCH

it
o+x
Mo-,

LAKE
RUDOLF

ale

l1.

M
EBM
aM
EBM

Q MorongoleOle

Sere
Sere-

oKuW
M

Maroto
Maxich

Ta
-lp

Gobwen(
E3

Baringo

LAKE
ALBEJ?T
Bombo

oKENYA

Kisumu
li

ENTEBB9
10

RUWENZORI

NAIROBI
l1v

LAXE
VICTORIA
Bukoba

LAKE
ED14ARD .
I
7
chuiu

Ao
K6wyi

---

'414
'Pl.,
kiLIMANJARO
M

11
LAKE

Arushu

11
wanza

KIVU
ya

An

MO

PARE

Lukemba

ASA

USAMB

RA
Tt

N4

Tanga
4Y

Kondoa
Irangi

Usurnbura
Tabora

Kigorria

ZANZIBA

04

DA-R-ES-SALAA

K limatinde
uga
02

0 hinga
Baudoinvdle

2
M ahenge

Q.
Neu
Langenburg

Bis=rckburg

Pweto

itope

Kit iiu ii-

LAKE

-- -

3 je, Lindi

*'*

61-

MWERU

CKasama

REFERENCE

THE
STRATEGIC SITUATION
at the Outbreak of War

VK. A.Rifl*s, company

........
3/K. A.Rifles, do

13, Fort
Jarnesixt

German Fold Kornparsic ......

"lo"British

Scale- 210 Milea to I inch

0

100

COMPILED 8Y 1,413TORICAL SIECTION

(MILITARY DRANCmi

Nian

1914
.
Miles
100 so
.

do.
half- company........
0
4/K. A.Riflas, company.
.......
Rhodesia Police, do.. _, _C3
Belgian Military Post ......... -A L,

ZOMBP3

200

300 Milts

M 11)

forces

41
.4

CR.

80-

Copyright

Rescived

as

located on 2nd. August 1914.

German units shown at their
peace

Crown

shown

stAillons
0, d.. -re

S-1-Y

19J5

look
dissertation
The
1918.
in
at
November
will
PortugueseEast Africa until the armistice
being
level,
high
from
placed on the strategic and political
emphasis
with
a
operations
imperativesthat drove the campaign for both sides. It will use a dual thematic and
fighting.
knowledge
the
in
in
of
the numerousgaps our existing
narrativeapproachto fill
Africa,
in
East
background
to
be
The thematic chapterswill
operations
usedto provide a
in
tactics,
the
in
from
of
theatres,
areas
differed
those
particularly
other
considerably
which
then
The
inter-Allied
will
chapters
narrative
transport,
relations.
supply,
medicine and
1918.
November
1916
September
to
in
how
the
period of
explain
eventsunfolded
INITIAL

BRITISH

STRATEGY

AND SOUTH AFRICAN

INVOLVEMENT

War initially came to Africa not for reasonsof expansionism,but for those of
British imperial security. From the outset,the British put in place a strategyof protecting
their maritime interests and attacking the German system of overseascommunications
27
In
terms
this
in
detracting
from
theatre
practical
the
of
war.
the
main
efforts
without
transmitters
German
destruction
as well as
of
wireless
the
the
system
colonies'
meant
of
denyingbasesto their fleet. This policy was also attractive to the governmentas it could
be largely achieved by amphibious operations and did not require the conquest of
had
its
British
the
inland
In
allies,
territories.
order to maintain unity with
substantial
imperial
for
the
expansion,stating
territories
of
purposes
specifically ruled out acquiring
29
that all permanent decisions would be subject to any post-war peace conference.
However,this high-minded declarationdid not rule out the conquestof enemycolonies,as
29
if
informally,
It
that
be
only
also recognised,
suchprizescould useful negotiatingpieces.
the self-governingdominions, such as South African and Australia, might have their own
aspirationstowardstheir Germancolonial neighbours.

' The diaries are pocket-sizedand written in a

short-handcovering details of movements,stocks,and casualtiestogetherwith an end
to over
of year summary, They cover from 1914to 1918. Boell's work, upon which he basedDie Operationenis enormous
running
i;
4,000 sidesof paper. Unfortunately, there are a few gapsowing to wartime destruction,but overall it is invaluable the historian.
'7 Hankey,Lord, TheSupremeCommand,Volume 1, London: GeorgeAllen
and Unwin, 1961,p. 168; French,British Strategy,pp.
27-28; Oxford and Asquith, Earl of, Memories and Rey7ections
1852- 1927, Volume II, London: Casseland Company, 1928,p. 25.
23CAB 21/3, Proceedings
of a Sub-Committeeof the Committee of Imperial DefenceAssembledon the 141hAugust, 1914", 14
August 1914."... no formal proclamation annexing any such territory should be issuedwithout specific instructions from His
Majesty's Governmentand further, that an agreementshould be come to with our Allies to act on similar lines." ; French,British
Strategy,p. 15.
2"Churchill, Winston S. 7he World Crisis: 1911-1914,Volume 1, London: Thomas Butterworth Limited, 283.
p.

The other Europeanpowers in Africa had their own views on the subject,but the
British made it quite clear, in diplomatic language,that their allies' offers of military
French
Portuguese
in
The
Belgian,
East
Africa
and
even
assistance
were unwelcome.
in
the
the
to
early
attempts contribute contingentsagainst
common enemy were rebuffed
30
war. This may be attributedto British over-confidencein their own abilities to eliminate
the Germansas rivals as well as a wish to deny other countries a claim in the potential
spoils. Despite the pressing need for military victory over the Central Powers, imperial
factors
throughout the war, and
significant
rivalries and colonial aspirationswould remain
EastAfrica would be no exception.
However, this littoral strategy did not last very long in its pure form as other factors
induced the British to invade German colonies shortly after the outbreak of the war. Subimperialism and local fears for security played their part in forming a more aggressive
forces
Togoland
fell
by
August,
the
were
colonial
expeditionary
while
attitude.
end of
being prepared to attack the Cameroons in the north and German South-West Africa in the
31 While
it
initially
longer
drag
than
these
to
was
expected,
south.
on much
campaignswere

in EastAfrica that this alterationto the initial British strategywas to have the most visible
effects.
Germanyfaced the war in East Africa from two points of view: one from Berlin
andthe other from Dar-es-Salaam.From the imperial viewpoint, colonieswere not a major
priority although,prior to the war, efforts had beenmade to reducecolonial tensionswith
Britain, actions that had been only partially reciprocated.Indeed, the two powers had
initialled the draft of a secrettreaty aimed at carving up the Portuguesecoloniesin the case
32
financial
default.
However, once hostilities broke out, thoughts of
of an expected
conciliation were quickly forgotten and a more aggressiveposition was taken. A group in
I
30MS Harcourt,dep 507, "Colonial Office TelegramsCirculated
to the Cabinet August 1914- August 1915", 22 September1914,
TelegramNo. 1, Secretaryof Statefor the Colonies to the High Commissionerfor South Africa; 3 October 1914,TelegramNo. 3,
Secretaryof Statefor the Colonies to the Governor EastAfrican Protectorate(for the Belgians); dep 507,27 August 1914,Telegram,
No. 2, Secretaryof Statefor the Colonies to High Commissionerfor South Africa;
and dep 590 Foreign Office Print, 26 August 1914,
TelegramNo. 176, Sir Edward Grey to Mr Carnegie(for the Portuguese);CAB 21/3,12 August 1914,TelegramForeign Office to Sir
F Bertie; and Andrew, Christopher and Kanya-Forster,France Abroad, London: Thamesand Hudson, 1981,
pp. 62-63 (for the
French).
3t French,British Strategy,
pp. 27-28.

10

the government,exemplified by the Colonial Secretary,Dr Wilhelm Solf, aspired to a

GermanMittelaftika, built on existing colonies and augmentedby helpings from the
33
links
1914
by
However,
Portuguesethe
Belgian-held
territories.
all effective
end of
and
Solf
his
been
had
understoodthe
the
aspirations,
of
with
cut off and regardless
colonies
follow
territory
that
a general
the
would
any exchangesof
realitiesof
strategicsituationand
34
peacesettlement.
On the other hand,the Germancolonial authoritieswere faced with the much more
immediate concern of survival.

They realised both the extent of their isolation and the

initially
Schnee,
Dr
The
Governor,
tried to claim neutrality
their
weaknessof
own means.
by
but
the
1885,
Berlin
Act
these
the
the
shared
concerns were not
under
provisions of

British, nor indeedthe Belgiansoncetheir

35
homelandhad beeninvaded. Furthermore,the

commanderof the Schutztruppeactively opposedthis attitude of non-belligerenceand a

disputebetweenthe two ensued36 In fact, pre-war German plannershad consideredthe
.
situationandhad correctly concludedthat the British were unlikely to acceptneutralisation
be
directed
it
had
direct
interests.
Accordingly,
that
they
their
preparations
as went against
37
if
Therefore,
for
defending
interior
the
than
early
the
coast.
made
of the colony rather
surrenderin the manner of Togoland was not to be contemplated,the only effective
Von
Lettow's
long
for
delay
to
personality
try
to
as
possible.
strategywas
as
and
survive
andthe courseof eventseventuallywon out, and his strategyof distracting as many enemy
troopsandresources,even at the expenseof evacuatingthe colony, continueduntil the end
of thewar.
While a Belgian colonial adventuremay appearunusual with most of the home
country under occupationby the Germansand the remainder largely a battleground,the
"Fischer, Germany's War Aim Aims in the First World War, London: Chatto & Windus, 1967,
pp. 38-39; French,British Stratesy, p.
9-10 and 14; Langhorne,Richard. "Anglo-German Negotiations ConcerningThe Future of the PortugueseColonies 1911-1914",
Historical Journal, XVI, No. 24, (1973), pp. 378-379.
" Andrew and Kanya-Forster,France Abroad, 57; Fischer, Griff
nach der Weltmacht, p. 94.
p.
34Fischer,Fritz, Germany's War Aims,
102-103.
pp.
35Schnee,Deutsch-Ostafrika, 37-38; Hordern, Military OperationsEast Africa,
pp.
pp. 527-528, CAB 8/5, Colonial Defence
Committee,No. 431 M. "East and West African ProtectoratesPosition in the Event of War with a EuropeanPower", 24 January 1911.
The British had examinedand ruled out a declarationof neutrality for its East African colonies as early as 1911. It saw such neutrality
as favouring the Germans.
36 Lettow,

My Reminiscences,

pp. 27-28.

37MS Boell, N 14/14,Denkschrift aber Mobilmachungsvorarbeitenfar den Fall

einesKrieges mit Grossmachtftr Deutsch-Ostaftika,
27 April 1912.

11

prospectof a campaign held a number of attractions for the government-in-exile. The

includedboth nationalpride and diplomacy; in short they neededsometangible
advantages
38
in
from
in
to
gains order negotiate
a position of strength any peaceconference. As the
official Belgian instructions directed, they wanted "a pawn in the form of a portion of
Germanterritory".39 The proximity of German East Africa and the support promised by
the British madethe campaignattractive,as did the wealth and fertility of its provincesof
Ruandaand Urundi. A limited campaignto seize and control some of the best parts of
Germanterritory at relatively low risk was a most attractive option. It also had the
advantagesof placing the much strongerBritish in their debt, of enhancingtheir Congo
colony, and of inflicting pain on the despoilersof their country. Under these conditions
andfrom a Belgian point of view, an offensivemadestrong strategicsense.
The last of the four powers involved in the East African campaign,the Portuguese
also emergeas by far the weakestand least effective, virtually drifting or being pushed
along by the will of their stronger neighbours. Hated by their African subjects for the
venality and brutality of their rule, the Portuguesewere also despisedby the European
40
including
for
incompetence.
Germans,
the
their military and administrative
protagonists,
They were also awarethat the British and Germanshad beeneyeing up their coloniesand
that both consideredtheir administrationto be hopelesslyinefficient and cruel 41 With the
.
outbreakof war, Portugal stood aloof, declaringneither neutrality nor belligerence,largely
in accordancewith British desires.42 This policy also accordedwith Portugal's desireto
protect its African colonies against encroachmentwhile recognising Portugal's own
financial and military weakness. However, a combinationof internal politics and external
pressureled it into the war in March 1916with ultimately disastrousresults.
All of the Europeanpowers were in comparablepositions with very small white
populationsruling over millions of blacks who had no say in questionsof war or peace.
" Louis, Ruanda-Urundi,
pp. 216-217.
39CampagnesColonialesBelges,Volume
11,p. 125,"Instruction minist6rielle no 563 du 22 avril " quotedand the terms "3. Occuper
a litre de gage uneparlie du terriloire allemand: 1effortprincipale devait
le Ruanda."
etreportd
vers
40Pdlissier,Naissancede Mozambique,Volume 11,
650-651 and 656.
pp.
41Teixeira, Nuno Severiano,LEntreif du Portugal
dons la Grande Guerre, Paris: Econornica, 1998,pp. 114-119;Vincent-Smith, J D,
"The PortugueseRepublic and Britain, 1910-1914",Journal ContemporaryHistory, X, No. 4, (1975)o 714-717.
of
pp.
'2 Vincent-Smith,John,"Britain, Portugal
and the First World War, 1914-19", EuropeanStudiesReview, IV, No. 3, (1974), pp. 210211; Teixeira, LEntrie du Portugal, pp. 188-194.

12

Yet the African population was to be drawn into the war in a manner and scale quite
be
it
in
its
history.
Whether
to
through
soldiersor porters,
volunteering
was
unprecedented
force,
by
frequently
information,
food
labour,
being
and
or
compelledto provide manual
few communitiesescapedthe effectsof the war and many were devastatedby it. Owing to
illiteracy and the undevelopednature of their societies,their voices are largely silent in
described
few
the
historical
their
could
events
of
participation
most
accounts,yet without
43
importance
lack
be
have
Silence
and the
of
with
confused
must not
ever
occurred.
African contributionto the campaignwas absolutelyessentialif not yet fully explained.
THE DEVELOPING

STRATEGIC

SITUATION

FROM 1915 - EARLY 1916

The East African campaignevolved from the relatively simple plan of destroying
illinto
Dar-es-Salaarn
denying
facilities
the
an
as
a
port
use of
coastalwireless
and
judged attempt to seize the whole of German East Africa with two lightly equipped
brigadesof the Indian Army. The decision to attack was made on a number of erroneous
inter-departmental
that
through
poor
planning,
assumptions
arose
a combination of
long-lasting
have
it
leadership;
lack
firm
to
effects on the
was
rivalries and a
of
campaign.
44

The result was the disastrousattempt to land at Tanga in early November 1914
.

Decisivelydefeatedby the numerically inferior Germanforces,the British retired to British

EastAfrica where, on the orders of Lord Kitchener, the War Secretary,they went over to
the defensive. He correctly realisedthat EastAfrica was a strategicbackwaterand refused
further substantialreinforcementsas he generallyopposedany movesto diminish the main
45
Western
Front.
effort on the

4' Hodges,Geoffrey, 77jeCarrier Corps: Military Labor in

the East Aftican Campaign 1914-1918,Westport, Connecticut:
GreenwoodPress,1986. Seepp. 223-226 for his African
oral history sources;CAB 45/14, Transport Difficulties in EastAfrica
During the Great War, by Sir HD Lawrence, [n.d.j. In North-easternRhodesiaalone,the Administrator estimatedthat any one time a
third of the taxableadult male population (c. 40,000) was directly engagedin carrier servicewhile their wives and children were
growing food crops.
"A fuller accountof the
political backgroundand of the battle of Tanga itself is at Anderson,Ross,"The Battle of Tanga2-5
November 1914", War in History, VIII, No. 3, (2001), pp. 294-322.
45WO 33nl4, European War Secret TelegramsSeriesB, Volume1,31*July 1914 31-*January 1915, London: War Office, 1915.
No. 1522,4 December1914,Telegram2343, War Office to Wapshare.Henceforth, TelegramsB 1.

13

The abject failure at Tanga and the inability to provide significant reinforcements
for East Africa led to a change in attitude. By January 1915, official meetings were
best
determine
the
to
Belgian
British
between
in
the
governments
and
underway London
South
being
to
feelers
out
put
were
methodof military co-operation,while unofficial
Africa.46 Despite London's reluctanceto bring Portugal into the war, the Governor of
47
Nyasalandwas authorisedto requestmilitary assistanceas a measureof last resort.
had
less
they
had
only
Germans
On the other side, the
strategic choice as
much
intermittent communicationswith their colony that were based on frequently distorted
48
The
blockade.
British
through
the
wireless signals and occasionalmessagesspirited
destruction
Cameroons,
the
together
Africa
the
South-West
with
German
and
conquestof
East
Africa's
German
position much more
German
overseas raiders, made
of the
backwater
49
it
1915,
1914
Throughout
and
military
remaineda strategic
and
vulnerable.
into
been
had
fleet
directly,
little
do
their
Germans
penned
for the British while the
as
could
50
increasing
their
having
blockade
war efforts.
on
North
Sea
effect
the
an
the
was
and
Althoughthe new U-Boat weaponbeganto emergeas a major threatto the Allied merchant
fleet, the technology of the day limited the range of these vesselsto mainly European
insufficient
the
bases
available,
numbers
Without
to
with
and
access overseas
waters.
in
East
Africa
impact
the
although
situation
on
submarinewas unable to make an
51
increasingshippinglosseswould restrict the amountof supportto that theatre.
This defensivepolicy would hold true for the better part of 1915. The forces in
British East Africa suffered not only from insufficient numbers, but also widespread
lack
This,
from
the
of an energetic commander,
and
sickness
malaria and poor morale.
both
in
largely
sides
spent static warfare, with
sappedtheir effectivenessand the year was
relying on small patrols and raids to disrupt their opponents,while simultaneouslytrying to
build up their own limited forces.The one major initiative approvedby Kitchener was the
46MS Harcourtdep 583,4 January 1915,TelegramNo. 2, Kidston to Sir Edward Grey; dep 507,14 April 1915,TelegramNo. 4,
GovernorGeneralof South Africa to Secretaryof Statefor the Colonies.
47MS Harcourt,dep 507,14 January1915,TelegramNo. 2, Secretaryof Statefor the Colonies to GovernorNyasaland.
48Schnee,Deutsch-Ostafrika,
Verlag,
Herbst
Wilhellm
Ostheim/Rhon:
Nouen,
162-163;
Klein-Arendt,
Reinhard,
Kamina
pp.
ruft
1996,p. 292 and300.
49Strachan,TheFirst World War, p. 570.
'0 Halpern,Paul,A Naval History World War 1, London: UCL Press,1994,pp. 328-329.
of
31Halpern,Naval History,
pp. 306-308.

14

pushingof a spur of the UgandaRailway close to the British-Gennan border, a factor that
52
be
in
would crucial any offensiveoperations.
By late 1915,the British Cabinetwas divided into two main factions with strongly
opposedviews on strategy,the managementof the war effort, and conscription. Notably,
therewas strongoppositionto further offensiveson the WesternFront and someministers
soughtto find alternativetheatresin which to achievestrategicsuccess.Furthermore,Lord
Kitchenerhad lost the confidenceof his civilian colleagueswho respondedby schemingto
diminish his powers.53 One area of contention was East Affica, where Kitchener had
maintainedhis opposition to any further offensive action while keeping its reinforcements
to thebareminimum required.
Finally in November 1915, when he had been deliberately sent away on an
inspectionof the Dardanellesby the Cabinet, his opponentsstruck. Using a series of
pessimisticand alarmist reports from the general in command in East Africa, the CIGS
producedan appreciationthat called for at least 10,000reinforcementsand a more active
54
stance. In this he was supportedby his Director of Military Operations,who later
admitted to using Kitchener's absenceas an opportunity for going against his express
55
wishes.

The GeneralStaff did this with the knowledgethat there was broad supportfor the
56
scheme. Indeed,the King had taken a personalinterest in the matter for sometime and
hadhad his Private Secretarywrite to Bonar Law, the Colonial Secretary,aboutthe needto
involve the SouthAfricans in GermanEastAfrica. 57 He neededlittle convincing ashe had
long welcomedsuch assistanceand had beencarefully negotiatingwith the South African

52WO 33/858, TelegramsD 1, No. 23,6 February 1915,TelegramS 215, Wapshare

to War Office; Hordem, Military OperationsEast
Africa, P. 130. War Office approvalwas given on 16 February.
" Turner, John,British Politics
and the Great War:Coahtion and Conflict, 1915-1918,New Haven: Yale University Press,1992,pp.
68-69; French,British Strategy,pp. 158-161.
'4 WO 106/310,"The Military Situation in British East
Afria', CIGS 8 October 1915. In this report, GeneralSir AJ Murray cited
GeneralTighe's Telegram322 of 2 August 1915,his despatch 31 July 1915,
his Telegram325 of 14 August 1915.
of
and
55Callwell, Major GeneralSir C E, Experiences
Dug-Out 1914-1918, London: Constable,1920,pp. 178-179.
ofa
56Gilbert, Martin, WinstonS Churchill, Companion
III Documents,London: Heinemann,1971,p. 1296. Cites 3 November 1916,
Letter Curzonto Churchill.
57MS Buxton, dep 9930, File August 1915,13
August 1915,Letter Lord Stamfordhamto Buxton.

15

58
in
Furthermore,
for
the
Governmentthrough the Governor General
time.
victory
some
October,albeit with a reducedmajority, of Botha's South African Party, had clearedthe
mainobstacleto an offensivethere.
In the circumstances,the CIGS then produceda memorandumadvocatingan attack
in
South
African
troops,
East
Africa,
German
order to enhance
mainly
using
against
59
Britain's position at the end of the war. On his return to Britain, Kitchener wastedlittle
time beforeweighing into the proposal,stating:
"This schemefor offensive operationsin the centre of Africa is, in my opinion, a
Me
in
the
dangerous
the
generalmilitary policy
war...
of
presentstate
project
very
in
South
African
lead
troops
to
positions
therefore,
us place
now advocated,may
be
from
to
disaster,
be
liable
to
able
extricate
not
they
we
will
which
where
will
them,as our troops will be fully engagedelsewhere.I think that the recentexample
in
Mesopotamia
based
have
had
premises
on wrong
of similar proceedings
we
60
"
this
in
be
nature...
of
teach
to
operations
similar
should
cautious undertaking
us
Nevertheless,such was Kitchener's weakenedposition that the War Committee
having
1915,
December
28
the
gainedthe support of
endorsed new offensive strategyon
61
few
days
GeneralSir William Robertson,the newly-appointedCIGS a
earlier. It was a
in
from
British
East
Africa
British
to
the
attack
vast and ambitiousproject with
planning
the north-eastand from Northern Rhodesia and Nyasaland in the south-west,with the
Belgiansadvancingfrom the Congo in the west. The prospectiveCommander-in-Chief,
General Sir Horace Smith-Dorrien had identified many of the salient difficulties of
campaigningthere, but, nevertheless,it is clear that the politicians expecteda quick and
62
easyvictory. The scenewas now set for a major strugglein the heart of Africa.
soMS Buxton, dep9930, File May 1915,9 May 1915,TelegramBuxton to Harcourt; 11 May 1915,TelegramHarcourt to Buxton; 19
May 1916,Letter Bonar Law to Buxton; File August 1915,9 August 1915,TelegramBonar Law to Buxton.
" CAB 22/3, "An Appreciation by the GeneralStaff
on the Situation in East Africa", 10 December1915. It foresaw about 15,000
reinforcements,of whom 14,000were South Africa and 1,000were Indian, joining the nearly 15,000troops already in-theatre.
60CAB 22/3, "Minute by the Secretary State for War",
of
appendedto the GeneralStaff Appreciation, 14 December1915;CAB 22/3,
War CommitteeMeeting of 15 December1915.
61CAB 22/3, War CommitteeMeeting 28 December1915. Robertson's
of
agreementwas containedin the "CIGS Note" of 23
December1915.
62WO 106/310,"Appreciation the Situation in EastAfrica", GeneralHL Smith-Dorrien, I December1915,
pp. 3-4. Smithon
Dorrien put greatemphasison the needfor careful and thorough preparationon accountof the climate and endemicdiseases.He

16

SUB-IMPERIAL

INTERESTS

The involvement of the Union of South Africa in German South-West and

large
fact,
In
the
East
Africa
a
German
presenceof
was outwardly puzzling.
subsequently
SouthAfrican contingent,led by one of the country's foremostpoliticians and a veteranof
the Anglo-Boer War, may be considered remarkable given the traditional Afrikaner
hostility to British imperial actions. Yet while the Union of South Africa had only been
formedin 1910and many remainedhostile to their new overlords,it must be remembered
being
Great
Trek
the
the
founded
been
had
Boer
with
expansionism,
on
that the
republics
best-knownexample. Leading politicians, most notably Louis Botha and Jan Smuts,
Rhodesia
British-run
Africa,
South-West
German
territories
the
of
viewed
neighbouring
Indeed
South
Africa.
for
Greater
East
Africa
Portuguese
a
as attractive acquisitions
and
been
had
Boers
in
the
ftustration
had
been
the
which
there
manner
at
considerable
63
late
"hemmed-in"during the colonial expansionof the
nineteenthcentury. The outbreak
the
these
to
had
under
claims
advance
of war
presentedan unparalleled opportunity
auspicesof loyalty to Britain's cause.
The planned attack on Gennan South-West Africa had been delayed by the
had
forces
dissident
However,
the
in
October
1914.
Rebellion
as soon as
outbreakof the
been suppressed,the Prime Minister, Botha, took personal command of the operations,
64
future
field.
The
in
threat
later
from
Smuts'
Pretoria
the
of
risings
and
with
assistance
justify
from
the seizure:
to
territory
that
andphysicalattack
were used
"He [Botha] addedhoweveremphaticallythat the cost was worth incurring, if, ashe
believed, it would make any further rising impossible for the future, and if it
it
becoming,
from
by
German
the
territory
as
otherwisewould,
secured annexation
both a military and intriguing menaceagainstSouthAfrica."65
thoughtthat the Governmentbelieved that GEA could be completely subjugatedby April 1916,but warned that no decisive result
could occur beforeJuly or August 1916.
63Hyam,Ronald,SouthAfrican Expansionism, 23-24.
pp.
" Strachan,TheFirst World War,
pp. 554-555.
65MS Harcourt,dep 471,2 November 1914,Letter Buxton to Harcourt. This letter followed
a meeting with GeneralBotha.

17

Smuts certainly agreed with Botha and had no intention of relinqui ing any
African
66
for
South
less
target
East
Africa
German
obvious
was a
capturedterritory.
Nyasaland
Mozambique
Rhodesias,
by
it
the
and
the
massesof
ambitions,separatedas was
from the northerntip of the Transvaal. However, Smutssaw it as a useful bargainingtool
for
in
the
its
hoped
southern
to
Portuguese
exchange
area
southern
the
swap
and
with
portionof PortugueseEastAfrica:
Port:
to
to
of
Govt.
Union
some part
"The
are of courseanxious put out a claim
E.A. and think this an opportunity to do so, and the above re-arrangementof
67
idea.
,
territory is Smutsspecialpet
Suchan exchangewould provide first classports for an enlargedSouth Africa and
68
British
had
the
It
from
the
with
coinciding
of
the
merit
also
scene.
removea rival power
belief that Portugalwas unfit to be a colonial power and might welcome an expansionof its
British
hard-pressed
Furthermore,
banner.
the
African
interests
South
as
colonial
under the
for
the
Indian
campaign, a
Armies
to
reinforcements
significant
spare
and
were unable
strong South African contingent would be welcomed on military as well as political
in
firmly
had
Smuts
the
mind:
grounds.
spoils
already
11 But they [the British Government] now practically intimate that in future
...
GermanEast Africa will be our destination. If that country were conqueredby us,
we could probably effect an exchangewith Mozambique and so consolidateour
49

territoriessouthof the Zambesiand Kunene...

Smuts's ambitions, however, were getting aheadof themselvesas the fighting in

German South-WestAfrica was far from over. It was the Prime Minister, Botha, still
66MSBuxton,dep9930,File September
1914,24September
1914,LetterBuxtonto Harcourt.Smutsis quotedassaying,"if the
UnionGovernment
to
in theseoperations,
theterritory,it wouldbeimpossiblein thecircumstances
weresuccessful
andconquered
"
relinquishit to Germany,
67MSHarcourt,dep472,18May 1915,LetterBuxtonto Harcourt;MS Buxton,dep9930,File May 1915,15May 1915,Letter
Buxtonto Harcourt.
Hyam,SouthAfricanExpansionism,
pp.26-28
Hancock,
KeithSir,andPoel,Jeanvander. Selectionsfron;
theSmutsPapersVolumeIIIJune 1910- November1918,Cambridge:
Cambridge
UniversityPress,1966,30August1915,LetterSmutsto JX Meffiman,p. 310.

18

front,
directing
the
in
the
who was more cautious,
campaign at
engaged personally
70
before
finish
startinganother.
to
campaign
off one
prudentlypreferring
Despite this, and as early as May 1915, speculationreached a point where the
Governmentfelt obliged to issuea pressstatementstating that no decision had beenmade
in
Furthermore,
in
Africa
the
Europe
campaign
as
or elsewhere
aboutsendingtroops to
.71
South-WestAfrica wound down in July 1915, a new and difficult problem arose,that of
heavy
Botha
The
Election.
General
pressure
was
under
government
winning the required
72
Ministers were seriously
from the Nationalists and the Afrikaner population was split.
before
friendly
the
voters
potentially
many
the
so
effects of sendingaway
concernedabout
election:
"The next Election is the most critical this country probably will ever have to
face. It has not beenpossibleto prevent the First Brigade from being formed, and
that they are leaving before the Election takes place; but the Government are
disenfranchisement
further
the
to
do
that
would add
nothing
extremely anxious to
of voters.
in
do
to
they
in
Besides, the uncertain state of politics
not want move the
Contingent
Both
Election.
the
the
Contingent
East
African
until after
matter of an
73
,
itself, andthe financing of it might be usedaspoints againstthem electorally.
With the election set for October 1915, no outward preparationscould be made
train
it
to
last
take
and
equip
the
recruit,
the
months
several
and
would
until
quarterof
year
Affican
South
interim,
In
about
the
any contingent.
nothing was publicly announced
in
been
have
in
it
East
Africa,
a close
a
vote-loser
participation
as would almost certainly
Nationalists
by
Hertzog's
The
British
Cabinet
and
the
threat
contest.
posed
run
realised
heeded
London
Botha's
be
South
African
Party
to
to
accordingly,
wanted
returned power;
74
Lord Buxton's adviceand refrainedfrom public pressure.
70MS Buxton, dep 9930, File May 1915,25 May 1915,TelegramD 166 Smutsto Buxton citing 22 May 1915,TelegramB 45 Botha
to Smuts.
71MS Buxton, dep 9930, File May 1915,27 May 15, PressCutting CapeArgus.
7' Meintjes, Johannes,General Louis Botha, London: Cassell, 1970, 273.
p.
73MS Buxton, dep 9930, File August 1915, It August 1915,Letter Buxton to Bonar Law.
74MS Buxton, dep 9930, File August 1915,19 August 1915,Letter Buxton to Asquith; 28 August 1915,TelegramBuxton to Bonar

19

For sometime Buxton had believedthat the passive,defensivestrategywas failing

to protect Northern Rhodesia and Nyasaland adequately, and had been pressing the
Colonial Secretaryto take the offensive.75 Despite War Office opposition, Bonar Law
ageedwith Buxton's assessment:
"The position in EastAfrica and surroundingdistricts is of a natureto cause
the greatestanxiety... it would be of the greatestpossibleadvantageto senda large
enoughforce to conquerGennanEastAffica onceand for all.s976
With the victory in October, albeit with a reduced majority, of the Botha
Government,the main obstaclein South Africa disappearedand together with the British
Cabinet'sagreement,the way for war in EastAfrica was clear.

Law.
71WO 33/858, TelegramsD 1, No. 163,10 April 1915,Telegram3970,
War Office to Buxton, p. 39.
76Yorke, EdmundJames,A Crisis Colonial Control: War
of
andAuthority in Northern Rhodesia:] 919-19, CambridgeUniversity PhD
thesis, 1983,p. 132. Cites Bonar Law to Buxton, 3 September1915.

20

CHAPTER 2- THE SITUATION

AT THE END OF AUGUST 1916

SMUTS'SAPPOINTMENT

With the plan to invade GermanEast Africa approvedby the British Cabinet and
the South African Government providing the bulk of the contingent, General SmithDorrien was expectedto lead the force. However, he had fallen dangerously ill of
deemed
despite
during
long
journey
medically
the
was
a recovery,
pneumonia
sea
and,
unfit to take up his command. In the circumstances,the British offered the appointmenta
secondtime to Smutsin February 1916. He had beenapproachedprior to Smith-Dorrien's
77
With
but
had
declined
his
and
political
responsibilitieS.
nomination,
citing
parliamentary
thegovernmentnow re-elected,his commitmentsbecameless demandingand he agreedto
takeup the post of commander-in-chiefof the East African Expeditionary Force (EAEF).
Theappointmentof a colonial officer in supremecommandof imperial troops receivedthe
assentof both Lord Kitchenerand the War Committeeon 3 February1916.78It was to be a
fatefulandfar-reachingdecision.
The appointmentof a politician, although a major general in the Union Defence
Force,was extraordinaryto say the least. It partly reflected political disillusionment with

the Army's seniorleadership;to Lloyd George,BonarLaw and Churchill he seemedto

personifythegreatdreamof the"giftedamateur"politicianwho wascapableof showingup
79
blinkered
thenarrow-minded
and
soldiers.
professional
Smutshad also helped manoeuvrehimself into the position. As has been seen,he
had been instrumental in instigating an aggressive policy in East Africa as well as
pushingfor a large South African contingent. Although the war in German South West
Africa had beenfar from popular in someAfrikaner circles, he judged that the benefits of
77MS Buxton, File February 1916,1 February 1916,Telegram
Buxton to Bonar Law-,8 February,Telegram Buxton to Bonar Law;
Hancockand van der Pocl, Selectionsfrom the SmutsPapers,111,29November 1915,Letter Smuts to AB Gillett, p. 326.
73CAB 22/3, War CommitteeMeeting 3 February 1916.
of
71MS Long, 947/547,29 March 1917,Letter Derby Long; MS Long, 947/602,11 May 1917,Letter Buxton Long; Roskill,
to
to
Stephen,Hankey:Man ofSecrets, London: Collins, 1970,p. 388;
and Hughes,Matthew, Allenby andBritish Strategy in the Middle
East 1917-1919,London: Frank Cass, 1999,p. 24; Gilbert, Martin, WinstonS Churchill, 111,London: Heinemann,
pp. 565-566,cites
12 November 1915,Letter Bonar Law to Asquith.

21

supportingBritain in East Africa would outweigh the unpopularity of a new campaign.

For its part, the British Governmentwas anxious to supportthe Botha government;it had
suppresseda major rebellion before conquering one enemy territory and was now
actively supporting the Imperial war effort in Europe as well as Africa.

In the

circumstances,Smuts' appointment, with his Cambridge education and connections,

Anglo-BoerWar serviceand more recentexperience,seemedto be a good one.
On the other hand, there were serious limitations to Smith-Dorrien's successor.
80

He was not a professionalsoldier and he lacked experienceof higher commandin war.

He had been a very successfulleader of a commando in the Anglo-Boer War, but his
troops had seldom numberedmore than 300 and the fighting was in friendly and fertile
territory. Subsequently,he had served for some time as second in command in the
campaignto conquer German South-WestAfrica, leading a subsidiary column from the
south. However, while arid, that country was reasonably favourable for the use of
mountedtroops and had few of the deadly tropical diseasesthat plagued the East Coast.
Importantly, the German defendersthere were not particularly well-led and conducteda
mainly passive defence that succumbed easily to Botha's tactics (and logistical
81
planning). Now, Smuts was being called upon to lead over 50,000 soldiers of many
nationalitiesand languagesthrough extremely difficult and often unmappedterrain while
battling a harshclimate and a host of deadly diseases.Apart from operationaland tactical
skill, his appointmentalso required a firm graspof administration and supply which had
beenshownto be so important in previous African campaigns.
Smuts'sposition was also not madeeasierby his statusas a politician and leading
governmentminister. It was unavoidablethat politics would play an important role in his
decision-making,be it the appointmentof officers, selectionof strategyor the issuanceof
communiqu6s. With Botha under heavy pressurefrom the Nationalist party, Smuts was

goMS Buxton, Semi-Official Letter File, 12 February 1916,Letter Smith-Dorrien Buxton. Despite
the public announcements
to
statingthat he had recommendedSmutsto be his replacement,he privately made it clear that this was not the case."Concerning
GeneralSmuts' powersas an up to date GeneralI know
nothing and I certainly should not have selectedhim had there beentime to
get a Generalwho had devotedHis life to War from Home."
3' Strachan,77zeFirst World War,
pp. 559 and 567-568.

22

"
list.
He was also well awareof the
dangers
large
acutelyconsciousof the
casualty
of a
need for success,as any setback would be used to attack and possibly weaken the
govermnent.
SMUTS'S AIMS AND INTENTIONS

On taking over the chief commandfrom Smith-Dorrien, Smutsreceiveda copy of

the former's instructions,which read in part:
"is to undertake an offensive defensive with the object of expelling the enemy
from British territory and safeguarding it from further incursion. The decision as
to the ultimate scope of the offensive operations to be undertaken against German
East Africa after the rainy season should be postponed until General Sir Horace
Smith-Dorrien has reported in light of the experience gained before the rainy
83
,
season.

However, Smuts had no intention of waiting for the dry weather and had been
He
for
his
before
pressing
an early advancewell
appointment as commander-in-chief
had prevailed upon the Governor Generalto contest Smith-Dorrien's plans to start after
the end of the rains in late May, claiming that delay would have deleterious effects on
84
health
morale,recruiting and the
of the troops. In this he was successful,obtaining the
War Committee's permission to start before the rains, provided that he was confident of
success. Accepting the appointment on 5 February 1916, he sailed for East Africa
shortly thereafter, arriving at Mombasa on 19 February. He had a meeting with the
incumbentGeneralOfficer Commanding,followed by a rapid reconnaissanceof the areas
likely to be involved. Despite the shortnessof time for detailed consideration of the
situation and his lack of familiarity with the region, he signalled to London on 23

12Hancockandvan den Poel, Selectionsfrom

the SmutsPapers, 13 April 1916,Letter Botha to Smuts,p. 356; Meinertzhagen,Army
Di.ary, p. 166.
23Hordem, Military OperationsEast Africa, 222.
p.
" MS Buxton, File February 1916,1 February 1916,TelegramBuxton Bonar Law.
to

23

Februarythat he wished to undertake an immediate offensive, receiving approval two

85
dayslater.

This senseof over-riding haste was to be typical of Smuts's time in command.

While it was interpretedas part of his implacable willpower and desire for success,it can
alsobe seenas part of his lack of understandingof the situation. He had pressedfor an
acceleratedoffensive timetable before he had been given all of Smith-Dorrien's
86
had
inspected
the terrain. He made up his mind to attack in less
campaignpapers or
than four days in the country, of which the better part of two were neededfor travel, and
beforethe all-important transport arrangementswere put in place. Furthermore,his own
troops were still in the processof arriving (one brigade would not arrive until after the
opening of the offensive) and had had very little time for acclimatisation or
reconnaissance. Most significantly, he had already decided that his mounted troops
would play a leading role through a country that was known to be infestedwith the tsetse
fly with its fatal effects on all domestic animals.87 This was not just relevant to the
mounted brigades, but also to the overwhelmingly animal-drawn supply system.
Ironically, his predecessorhad specifically requestedthat only the front line units have
88
Smuts chose not to wait or
mules with the remainder equippedwith motor vehicles.
rectify this critical weakness.
SMUTS'S STYLE OF COMMAND

Furthermore,the unique combination of Smuts's previous military experience,his

political role and territorial aspirations together with his personal characteristics,gave
him a decided approach to campaigning. In order to achieve his ambitious ends, he
plannedto use manoeuvreas the meansof defeatingthe Schutztruppe. He was opposed
to fighting heavy battles and preferredto use limited frontal holding attackssupportedby
" Smuts,Lt Gen J C, Military Despatchesftom
the Commander-in-Chief East Africa Force, Despatchdated 30'* April 1916,
V, blished in the London Gazette,'Supplement'. No. 29630,20 June 1916,p. 187.
MS Buxton, File February 1916,7 February 1916,Letter Brigadier GeneralSimpson-Baikieto Buxton. The
relevant files %ere
only despatchedon that datewhile the requestfor an acceleratedattack was sentto London on I February.
'7 Off, Colonel G M, "Some Afterthoughts
on the War in EastAfrica", Journal of the Royal UnitedServicesInstitute, LXIX, (1924),
p. 697.
" WO 106/310,"Appreciation
on Situation in EastAftica! ', GeneralHL Smith-Dorrien, I December 1915,p. 4.

24

hitting
by
highly
the
turning
at
wide
movements
mobile mounted columns aimed
89
flanks
his
enemy's
and cutting off
escape. He countedon their unexpectedappearance
in the rear areas and withdrawal routes to cause disruption and disintegration. These
tacticshad worked well on the South African veldt and in the bush of South-WestAfrica,
but did not take the rather different conditions of East Africa into account.90
Above all, he wanted to conquer territory and, despite his pronouncements,he
neverappearsto have been really interestedin defeating the Germansin pitched battle.
This was not due to physical fear, as he was personally brave and was frequently in or
91
front
line.
The quality of his intellect stood out and he understoodmanoeuvre
aboutthe
very well; his difficulties came when it came to pressingthe fight home and translating
his grandintentions into attainableand sustainableorders.
That Smuts would be unfamiliar with the British Army's methods was expected
and an experiencedofficer, AR Hoskins, was supplied as his Chief of General Staff
(CGS). However, while taking over the bulk of Smith-Dorrien's staff, Smutshad insisted
on a number of changesin key appointments. The most important substitution was of
Hoskinsfor the SouthAfrican JJ Collyer as CGS. As this was by far the most influential
staff post, Smutswas within his rights in making such a decision, but it was at the cost of
losing a highly experiencedprofessional officer who had been Inspector General of the
King's African Rifles for one who had only served in South Africa. Matters were not
helpedby the relative inexperienceof many of the more junior staff officers, particularly
in the administrative branches.92 While Collyer was able enough in purely operational
matters,he never seemedto grasp the importance of underpinning Smuts's plans with a
sound logistical basis. The importance of this weakness would become glaringly
apparentin late 1916.

19Smuts,Lt Gen JC Military Despatchesfrom

the Commander-in-Chief East African Forces, Despatchdated 28'hFebruary 1917,
publishedin the London Gazette" Supplement",No. 30020 of 18"'April 1917,pp. 121,125, and 128-129. Thesetactics were
frequently usedby van Deventerand other South African
commandersaswell. Henceforth, Smuts,DespatchIII.
" Strachan,TheFirst World War, 1, 605.
p.
91Meinterzhagen,Army Diary, 199.
p.
92Crowe, General Smuts'Campaign,
pp. 4 and 7.

25

The bringing together of British and Imperial forces under South African
higher
friction
led
the
to
at
particularly
commandwas unprecedentedand
rivalry and
levels. While Smuts's qualities of leadership and decisiveness were appreciated,
from
the professional
doubts
his
tactical
abilities emerged,particularly
considerable
over
British officers in the force. Indeed, there was some speculation about the motives of
Smuts and the other senior South African commanders,most of whom were close
home.
A
in
involved
in
life
number of
at
politics
associates private
and were often
British officers believed that the inability to catch and defeat the Germans in pitched
battlewas due to a distinct reluctanceto incur heavy casualtiesand therefore opprobrium
93
in
Smut's
lasted
tenure
This
throughout
command and
the
sentiment
with
electorate.
contributedto ill-feeling betweenthe two groups.
Unhappiness with the commander-in-chief was not limited to British Army
by
deeply
distrusted
had
himself
Smuts
a considerable segmentof the
officers.
made
South African population, both Afrikaners and English.94 He was admired for his
intellect and energy,but was also aloof and remote, lacking Botha's personalmagnetism.
He was consciousof his superior mental agility and adroitness,and this confidenceoften
be
headquarters,
his
to
He
preferring
main
came across as arrogance.
was seldom at
forward with a few, trusted officers. Smutswas not a man for discussingmatterswith his
95
subordinatesor seekingadvice.
Thus by the end of August 1916, Smuts had outwardly achieved a great deal
reachingthe heart of the German colony. Much further to the west, the Belgians under
GeneralTombeur, and aided by a British brigade, had driven the enemy out of Ruanda
and Urundi and were threatening Tabora. To the south-east,a brigade-sizedforce under
Brigadier GeneralNorthey had advancedfrom the Rhodesia-Nyasalandborder and had
pushednorth-west against light opposition reachingthe hilltop town of Iringa. In the far
south, the Portuguesehad recently enteredthe war on the side of the Entente, and had
" Meinertzhagen,Army Diary,
pp. 198-200,CAB 45/44, Colonel Fendall's Diary, Entry for 29 August 1916;Crowe, General Smuts's
Campaign,pp. 6-7 is the most tactful.
" MS Harcourt,dep 474,7 May 1915,Letter Buxton to Harcourt,
and 17 May 1915,Letter Buxton to Harcourt; Reitz, Deneys,
TrekkingOn, London: Faber & Faber, 1933,pp. 68-69 and 105-106.
95CAB 45/44, Colonel Fendall's Diary, Entry 12 October 1916;MS Buxton, File June 1915,6 June 1915,Letter Buxton to Bonar
Law.

26

...........
I-4LAff

BRITISH
EAST

MA.MOX

1-4'-=

%...

ZAV

Vol

...

GERMAN
F, 4ST
An? IC4
TABOJU

IWONA

Z-)

BIEL

77-ANJA

Kff-LWTME

5
WAIH

VAN
DEVEN-TER

MA

JWOROGORO

m
Im
rJZJWGA
UJAWA

ol

M IM

NORT=

DAR- fZ. 4
DA.
-r4.4-V

Shims
LEXTOW

Z0

a MVIENGE

ONEULANGEAWR6

I-,

IM

kRAUT

r
NYASA
MMES

100

0123

500 NULES

STRATEGIC SITUATION AT
END SEPTEMBER 1916

ALLIED FORCES
GERMAN FORCES

M
EM-

k
Z

landed an expeditionary force preparatory to commencing offensive operations against

the Germans.

To Smuts, with the richest and most fertile areas of western and northern German
East Africa in his possession and having superior numbers, the prospects of an early
9" While he had not been able to surround and cut-off tile Germans
looked
bright.
victory
territory
as well
he
had
of
amounts
the
vast
there
capturing
of
satisfaction
was
as
wished,
97
as the majority of the population and all of the main towns.
The German situation appeared to be difficult as the main body in the cast under
Colonel von Lettow was beginning its withdrawal south ofthe Central Railway. Several
hundred miles of sparsely populated bush separated them from the Wesurul)I)en under
Major General Wahle, still

Tabora.
around
concentrated

In the south-west, an

Ahteilung, 98 now under the cornmand of Major Kraut, had been reinforced and prepared
in
Finally,
highlands.
Iringa
tile
determined
the
south
to present more
along
resistance
the
lorces
Ocean
with
Indian
were
available
weak
tile
relatively
coast,
only
and along
lakes
Affican
The
the
being
lines-of-communications
were
troops.
great
sea and
majority
firmly under the control of the Royal Navy, while the Germans had lost access to their
main means of overland mobility, the railways.

Logically. and to preserve as much of'

While
for
them
to
many
surrender.
to
their possessionsas possible,it seemed make sense
Lettow's
it
German
favoured
took
account
of
von
the
this
no
approach,
on
side
determinationto fight to the bitter end.99
Yet, despite appearances and the optimistic communiques sent to the War Office,
things were not quite as they seerned. It was true that by the standards of the Western
front, Smuts's advances had been both speedy and lengthy.

Furthermore, his battle

high
disconcertingly
by
light,
the
sick
this
casualties were relatively
although
was offset

"' Smuts, Despulch 11, 132.

p.
,,7 WO 33/858, Telegrams D 1, No. 1500,11 January 1917, Telegram X 1182, Smuts to Secretary of State for War, p. 412.
9' The Germans seldom
used conventional battalion and regimental organisations in Fast Africa, preferring to create ad hoc
Abledungen (detachments). The Abteilungen (Abi), which generally ranged from two to six companies with one or two guris wcre the
backbone of von Lettow's system. The leadership and composition oftliese detachments changed frC(jUelllly- the commandcr's nanic
was used as the main florm ot'designation.
1.1Lettow, Renumsceilces, p 149,

27

in
defeated
been
had
Schutztruppe
neither
rate. Despite his claims to the contrary, the
100
It remained an effective and cohesive force, still
it
demoralised.
battle or nor was
first
for
the
this
was
There
8,000.101
two
situation;
major reasons
were
numberingover
Smuts's
defence
the
of
Lettow's
the
unsatisfactory nature
second was
and
skillful
von
tacticalmethods.
He was faced by a highly professional and determined opponent in Colonel von
Lettow. A career soldier, he had served as a Guards officer in Germany as well as in
being
to
Africa,
to
South-West
German
in
China
appointed
prior
and
colonial campaigns
102
highly
had
He
Schutztruppe.
although
and
personality
a very strong
commandthe
demandingand sometimesdifficult to deal with, he commandeda great deal of respect
Lettow
beginning
his
From
the
was of the unshakable
the
war,
von
of
amongst soldiers.
belief than the colony's fate would be decided on the battlefields of Europe and that his
103
front.
from
the main
duty was to tie up and divert as many allied troops as possible
This he would do regardless of the fate of the colony and unlike his South African
from
head
did
to
battle
decisive
believer
in
he
shrink
not
and
counterpart, was a great
head clashes. More importantly, he was fully aware of the importance of keeping his
104
army adequatelymaintainedand his diaries bear testamentto that concern.

Sincebeingturnedout of the Kilimanjaropositionin March 1916,he hadusedthe

had
he
in
being.
his
force
In
keep
for
this
delay
to
time
tacticsof
andthe tradingof space
beenmateriallyaidedby the densebushandlack of roadsin the interior of the German
Smuts's
forward
in
difficulties
British
together
The
the
with
advance
colony.
sustaining
This
his
initially
in
fight
battle
Lettow
to
aims.
achieving
reluctance
an all-out
aided von
is not to say that the Germanshad it all their own way as their counter-attackat Kondoa

"0 CAB 44/9, Draft ChapterXVII, The Campaignof 1917,p. 1. Smutswas cited twice in Vie Times,the first time in South Africa on
29 January1917and the secondin London on 13 March 1917. In the secondstatementhe asserted"... The campaignin GermanEast
Africa may said to be over but it is merely the remnantof an army that is left, and not a formidable fighting force."
...
101
WO 95/5292,War Diary GHQ, 25 January 1917,Appendix 46a, TelegramHoskins to GeneralEwart 22 January- this repeats
Telegram28209, CIGS to Hoskins. It cites a figure of 600 whites and 6000 Askaris. The reply is in War Diary GHQ, 27 January
1917,Appendix 54, TelegramOA 936 Genstaff to CIGS, 27 Januaryand gives the higher figure of 7,500. Seealso CAB 44/9, p. 3.
102
Lettow, Mein Leben,pp. 20-98.
103
Lettow, Reminiscences,pp. 18-19.
'0' MS von Lettow-Vorbeck, N 103/33-39.From mid-1916 onwards,von Lettow recordedthe amountof available food supplieson
almost a daily basis.

28

41)
34

kiu L

2387

E 1.

6 1E.

/uLog,

F-pi c,4

Irangi in April had been an expensivefailure and they suffered heavily from sicknessand
thelossof food growing areas.
The German commander was well known for his willingness to strike isolated
enemydetachmentsand then to withdraw before becoming decisively engaged. He also
had the substantialadvantageof falling back into known and friendly territory on wellpreparedlines of communication while his opponents had to bring all supplies and
reinforcementslong distancesforward. But above all, he led a strong and well-trained
force with good morale and the will to fight. This was an advantagethat he would retain
throughall of 1916and much of 1917.
Smuts's tactics were influenced by his own experiencein South and South-West
Africa with only some modification for the unique conditions in German East Africa.
That country imposed severelimitations on tactical formations and movement,ranging as
it did from tropical jungle to forested mountains to parched scrubland."' This, together
with the undeveloped state of the road network meant that communication between
columns was often impossible and units could pass within a mile of each other and
106
remain undetected. The sheer thickness of virgin bush often limited movement to
narrow and unimproved tracks that wound their way through forest, bush and jungle.
This forced the columns of fighting troops and their heavily laden carriers into single file,
often stretchedover many miles. Movement was slow and predictable as it took several
hoursjust to get the leading elementssufficiently aheadof the main body for it to set off.
Furthermore, the fighting troops were accompanied by a long line of carriers, who
usually matched or exceeded their own numbers; frequently too, there were large
quantities of pack and slaughter animals. Behind them, the lines of communication
troops would be straining to maintain touch in order to push the required quantities of
food, ammunition and equipmentto the forward areas."'

10-1
Hordem, Military OperationsEast Africa, pp. 12-14.
106
Sheppard,Brig Gen S H, "Some Notes on Tactics in the EastAfrican Campaign",Journal
the United ServiceInstitution of
of
India, XLVIll, No. 215, (April 1919),pp. 142-143;Off, Colonel G M, "Random Recollections East Africa 1914-1918",Army
of
Quarterly, X1, (1928), pp. 292.
"' Ridgway, Brig-Gen R T, "With No. 2 Column: German
EastAfrica 1917", 1,Army Quarterly, V, (1922), pp. 13-17; Orr,
"Afterthoughts", pp. 697-701.

29

Strong rivers, rocky outcrops and swamps caused time-consuming diversions,

while navigation itself was extremely problematic owing to the virtually non-existent
stateof survey and consequentlyinaccuratemaps. It was frequently necessaryto march
cross-countryon compassbearings with distancesmeasuredonly by human pacing and
timing.

Climate played its part, as for much of the year there was insufficient

precipitationto support grazing, while for the remainder there was a superabundanceof
water that impeded movement, made life generally miserable and promoted sickness.
Furthermore,densevegetationoften reducedvisibility to a few metresand often even less
when moving through bamboo thickets or elephantgrass. In the dry season,dust raised
by the marching columns enshroudedeverybody,while heavy rain and mist madethe wet
seasonevenmore trying. "'
In a land largely devoted to subsistenceagriculture, it was impossible to sustain
the indigenous inhabitants and the passing armies without long supply chains and
magazines. Diseaseof both man and beast was to prove decisive; malaria, dysentery,
pneumonia and typhoid weakened the human ranks while trypanosomiasis, horse
sicknessand rinderpestwiped out the animals.
The emphasison speedleft little time for accuratereconnaissanceand locating the
exactposition and depth of the enemy's positions was difficult. For the flanking column,
even the initial finding of the defenceswas a major concern and they often relied upon
the soundsof firing in order to close in for the fight. Liaison by telephone or runners
between friendly columns was only possible once the objective was near, but these
methods were hindered by the activity of the German defenders who often captured
runnersand cut communicationcables. While the British deployeda wireless transceiver
with each column, they were insufficient in number as well as being too heavy and
unreliableto be usablein controlling a tactical battle. The strain on the advancingallied
troops was magnified by the pressuresto advance as quickly as possible in virtually
unknown country where the first sign of the enemy was usually a short range rifle
ambushor a devastatingburst of machine gun fire from well concealedtrenches. From
108
Hordem,Military OperationsEast Africa, pp. 12
-14; Crowe, General Smuts'Campaign, pp. x-xiii.

30

based
followed
battle
on
then
sequence,
the
a predictable
this unpleasantopening,
Germandispositions."'
Despite these disadvantages, Smuts was to apply his methods repeatedly,
"'
However,
his
troops.
at
the
the
terrain,
the
of
state
weather or
seeminglyregardlessof
flanking
local
distinguish
the
it
is
important
the
attack, which
to
this point,
concept of
than
both
by
several
more
extended
seldom
to
and
which
sides,
effect
used
great
was
by
favoured
from
battle,
the
from
extended moves
the centre of the
thousandyards
Smuts. Thesewide turning moves frequently involved marchesof up to 10 to 30 miles
into
The
these
before
the
from
of
area.
results
rear
the
cutting
positions
enemy
parallel
"'
desired.
less
frequently
than
effective
much
moveswere
Throughout most of the campaign, the Germans' defences followed a familiar
facing
in
the
track
expected
the
area
an
they
sole
on
centred
usually
pattern;
were
direction of enemy approach. Carefully sited machine guns with interlocking arcs of fire
formed the backboneof the system. These were supplementedby cleared fields of fire
The
NCOs.
German
by
being
rest of
the
selected
manned
guns
and rangemarkers,with
for
fire
trenches
built
line
riflemen
the
was then
up around the guns with a series of
locality
then
flanks
in
depth.
flanks
The
the
anchoredon
were
entire
the
of
placedon
and
A
impenetrable
hills
it
impassable
be
undergrowth.
terrain,
or
swamp,
some type of
further seriesof depth positions were then constructedto the rear of the forward trenches,
kept
deep.
Finally,
was
motivated
reserve
a strong and well
sometimesas many as eight
infiltrate
flanks,
fend
breakthroughs
to
the
to
any gapsthat appeared
or
off any
ready
on
betweenthe attacking forces."'
The difficulty of the terrain and restricted visibility made the attacker's progress
slow and time-consuming while also masking the movement of the reserve. Once
109
Sheppard,"Notes on Tactics", pp. 152-153," Ridgway, "With No. 2 Column", 1,pp. 25-27.
110
Hordem, Military OperationsEastAftica, pp. 392-393; Off, Colonel G M, "The Winter Campaignof 1916 in East Afirica7,Journal
of the UnitedServiceInstitution ofIndia, LX, (1930), p. 78.
... Off, "Smuts vs Lettow: A Critical Phasein East Africa: August to September1916", Army Quarterly, VIII, (1925), p. 296;
Downes,Capt W D, With the Nigerians in GermanEast Africa, London: Methuen & Co, 1919,p. 146; Lettow, Reminiscences,pp.
202-203.
"' Sheppard,"Notes on Tactics",
pp. 150-155; Ridgway, "With No. 2 Column", L pp. 18-19 and 23; Off, ""From Rumbo to the
Rovuma The Odysseyof 'One' Column in EastAfrica in 1917",Army Quarterly, V11,(1924), pp. 111-115.

31

initiated, counter-attackswere fast and furious with the aim of surprising and disrupting
the attackersat a time when they were beginning to tire and lose formation. The
Germanswould then probe to find and, if possible,turn their opponent's flank, and break
into the vulnerable rear area with its attendantcarriers, ammunition and stores. Success
would be reinforced by further reservesif possible. Such tactics required the British to
deploytheir own local reservespromptly and decisively.
On the other hand, the British usually were advancingin a brigade-sizedcolumn,
with three to four battalions in single file, of which one would form the advancedguard.
The leading company of the advancedguard would be usually be the first to run into the
Germanforward picquets. After reacting to short-rangefire, the battalion would deploy
the remaining companiesforward and attempt to drive back the relatively weak security
elementsas quickly as possible while the rest of the column kept closed up behind it.
Eventually, the main defensive position would be reached and the advanced guard
preventedfrom further advanceby heavy machine gun and rifle fire. As the position
would normally be well sited and impossibleto rush, the column commanderwould bring
up his remaining battalions to form a firing line on either side of the advancedguard
battalion while retaining a strong reserve. These actions were intended to enable the
maximum number of troops to engagethe defenderswhile simultaneously searchingfor
an open flank. This processwould take at least an hour as the rear units had to move
forward and then hack through the bush to get to the flanks. The column commander
would be careful to keep his units tightly groupedand in physical contact with eachother
while he maintained control through hand-laid telephone lines. Generally, he and his
staff would be no more than 400 yards behind the leading troops while commanding
officers of battalionswould be immediately to the rear of the forward companies."'
If an open or weakly defendedflank could be found, then immediately a battalion
would be sentto turn it while the firing line continuedto engagethe enemy. At this time
it was vital that the column commander
retained a strong reserve under his personal
control as the defenderswould be attempting to find the end of the British line with the
'" Ridgway,"With No. 2 Column", 1,
pp. 24-25; Off, Colonel G M, "From Rumbo to the Rovuma7,pp. 111-112;Stronge,Capt HC

32

both
to
a
flanking
for
and
collide
intentions.
sides
the
It
of
moves
unusual
not
was
same
be
seriesof hasty attacks and counter-attackswould ensue and reserveswould rapidly
it
defences
guns,
their
reliance on machine
and
committed. Owing to the strengthof the
flanking
a
frontally
engagedtroops to make significant progress until
was rare for the
into
brought
be
Stokes
later
mortars, could
move succeededor the mountain guns, and
forward
had
to
but
crawl
fire
the
artillery observers
support was critical,
action. Indirect
in
back
laid
be
had
any
the
to
crews
line,
whose
to
firing
guns,
into the
telephonecable
When
the
for
bush
felling
in
clearance.
be
muzzle
trees
and clearing
casewould engaged
Germanshad artillery deployed,they were usually able to bring it into action much more
laid
had
telephone
fire
and
cable
had
already
and
they
positions,
pre-prepared
quickly as
reconnoitredthe target areas.
The fighting usually lasteduntil darknesswhen either further counter-attackswere
latter
In
bush.
into
the
decided
the
case, the
Gerinans
launchedor the
to withdraw
If
down
to
disorganised
all
force
effectively.
pursue
too
or
wom
was
usually
advancing
be
if
they
dead
would
defenders
and wounded; not,
would carry away their
went well, the
left on the field of battle for the British to treat and clear. With at least four carriers to a
least
for
deplete
few
the
a
at
it
to
column
mobility of a
casualties
stretcher, took relatively
day.114

In discussing tactics, careful distinction must be made between the advanceto

be
former
the
deliberate
In
would
the
advancing columns
case,
attack.
contact and the
be
locations
intentions
his
but
would
and
the
enemy
aware of the general presenceof
had
Accordingly,
the mission of advancingas quickly as possible,
the
columns
unknown.
to locate any positions and then to engageand drive off any rearguards. The details of
the attackoff the line of march would then generally follow the sequenceoutlined above.
However, if the position was found to be too strong for a hasty attack or had been
deliberate
Scouts,
located
Intelligence
attack
through
the
the
a
more
previously
efforts of
would become necessary. In this instance, a thorough reconnaissanceof the position
(1927), pp. 608-610.
T, "Bush Warfare Against Trained Troops", Journal of the Royal United ServicesInstitute. =I,
114
CAB 45/19, "A Record of the Doings of the OverseaContingent Nigeria Regiment,during the Campaignin GermanEast Africa:
December1916- December1917", [n.d.) [E St Stobartappendedin writing], pp. 35-38; Downes,"With the Nigerians", pp. 178-179.

33

be
would be attemptedwhile severalcolumns equippedwith artillery and mortars would
brought up. A holding attack would be launched against the centre of the German
defenceswith one or two column-sized flanking movementsattempted from the outset.
Suchmoves would often involve a march of no more than two to three miles off the
centreof the action before striking inwards. Thereafter, the normal round of probes,
decided
to
follow
side
either
until
counter-attacksand commitment of reserveswould
Battles would often last until past dark; their duration thereafter
dependedon the importance of the position to the German commanderwith a night-time

breakoff the battle.

in
to
try
both
If
and
place
entrench
the
would
sides
not,
result.
withdrawal often
"'
for
the
next phase.
reorganise
Such tactics were effective, but at a cost in casualties. Whilst the Schutztruppe
from
their pursuers,
darkness
bush
their
to
escape
effect
and
normally managedto use the
they suffered from steady attrition of their irreplaceabletroops, particularly the German
leaders.It was true that the British casualtieswere usually greaterthan the Germans',but
they were normally on the attack and had much greater reserves. Furthermore, von
Lettow's aggressivetemperamentand desire for decisive victories often led to heavy
losseson his side that, on a percentagebasis,equalledthose of his British opponent."'
If the British were hampered by lack of operational originality in 1916, the
Germansalso had enormoustactical advantages.They had a huge area in which to fight
had
least
If
difficult
they
traverse.
to
the
terrain
was
poor,
at
mobility
and
was uniformly
the time and resourcesto plan and organise their retreats. They were on the defensive
forces
British
hand,
largely
battles.
On
the
the
their
advancing
and could
other
choose
into
lines
from
depots
their
and
communication
of
were moving away
on precarious
be
finding
The
imperatives
trying
to
territory.
able to
unknown
water and
urgent
of
supply the forward troops alternately forced and hindered the pace of advance.
Opportunities were lost through exhaustion of the fighting force or lack of supplies;
under-nourishmentand sickness played major roles in dissipating the strength of the
"s This occurredat Bweho Chini, Narunyu, Narungombe,
and Mahiwa to list a few.
116
Despitevon Lettow's claims of great victories, at Mahiwa both sideslost nearly 25 per cent of the engagedforces. Similarly,
lossesat Kibata, Narungombeand Lioma were high on both sides.

34

117Certainly, Lettow madethe most of theseadvantages,which, combinedwith

attackers.
Smuts'sreluctanceto fight heavy battles, enabledhim to keep his force largely intact
throughoutthe year."'
THE OPPOSING FORCES IN SEPTEMBER 1916

By late September1916,the EAEF had advancedover 300 miles from its starting
positions in the Kilimanjaro area. During the advance south, two halts had been
the first owing to the annualmonsoonin April and May and the secondat the
necessary,
endof July to replenishsupplies. The second%%ras
particularly worrisome as the lack of
adequatelogistical preparationsmeant that much of the force was on the verge of
119
force
but had
driven
his
hard,
his
had
physicalcollapse. By
own admission,Smuts
Wed to appreciatethat physical endurancecould only go so far and that soldiers and
followershadto be fed, clothed,and suppliedadequatelyif they were to be fit for ftirther
120
from
his
Despite
military operations.
staff, he had
considerableadvice and warnings
pressedthe advance without ensuring that the unglamorousbut critical matters of
transport,medicalsupportand rear areaorganisationwere in placebeforehand. He now
contemplated
plunging into the low-lying, pestilentialand totally undevelopedhinterland
southof the railway with a force that was already suffering heavily from sicknessand
121However,the
malnourishment.
weaknessof the troops and the nearing onset of the
shortrainy seasonmadean operationalpauseessential.
In the east, the Royal Navy had disembarked landing forces at Dar-es-Salaarnand
at the remaining coastal ports bct%vccnthat town and the southern border %%ith
Portuguese
East Africa. 122 The Central Railway
was almost entirely in Allied hands although
German demolitions had
put it out of order for some time to come. If General Smuts's
"7 Crowe,General
&"ut, ' Ca"Wigo%pp. 9-10; Orr "From Rumboto the Rovum&7,pp. I 10-112;Ridgway,'With No. 2 Columd"qpp.
15-16;Ridgway.Brig-Ge, R T.
'With No. 2 Column:GermanEnt Africa 1917:Part2", Amry QwHerty, VL pp. 260-26 1.
"I Mcincrtzhagcn.
Army Diary. pp 165-166.
119
Crow, General&,, - CamWign.
"Is
p. xI. Smutshimself said in the foreword"It maybe said dud I expectedtoo muchof my men.
andthatI imposedtoo harda task them
die a%fulconditionsof this tropicalcampaigning.Idonotthinkso. lamsurcitwas
on
under
rVHP: s; to conductthis cAmPaIgn
in anyother way."
successfully
d
4,1,,,.,Y OPrrwEastAfi*v. p. 393; Off. "Rarkkxn Recollections",pp. 290-291.SouthAfrican NationalDefence
Force(SANDF)DocumentationDirectorme,
GSWA.
GSWA.
Box
2k
2
Div
Smuts.
licnccforth
II
Sq*cmber
1916.
Telegram
to
121
WO 95/5291,War Diary GIIQ. 22
September1916.Appendix32a.TelegramCIA4 IZ Smutsto CIGS,21 September.
123
liordern, Ulh&vy Opra,
$OmraIjO$c4, pp. 381-393and386-389.

35

j, q
oo
11-4

0
0

B:.

0.
cr

40

cn
.9

am

P
b-4

li
El

-4

:4E.

w..Uolr

pr
0

Pt

wi

7:

-D
0.

(4,

_N-
!4

N.

3:1-

ft

It-cor
VIL

_ C. -

2..

C:
cl

II.

R,

lip

;a2.
or

a
rr

00

N-

..

-..

....

'!

to,

.Ia CY

d*

:3
IIa,

C) ; C30

--1x

2
X.

Cf,

;.

>
X

4r
J,

x0*

--,

a0L'

.09
In

had
They
Belgians.
had
been
the
had
those
then
of
so
reached,
physical objectives
123
They
fielded two brigades, each with six battalions, totalling about 5,000 soldiers.
to
had
down
their
they
namely
aim,
political
achieved
were now winding
operationsas
hold
keen
the
to
in
bargaining
captured
were
and
gain a
any peace negotiations,
chip
124
Tabora
held
Nord
The
Brigade
the
Ruanda.
Urundi
and the
of
area
regions of
and
north, while the Brigade Sud was deployed to the south and south-eastof that town,
facingAt Wahle.125They had beenwom down during the campaign,but were still more
than a match for Wahle's much smaller force. However, the termination of Belgian
down
he
the
turned
discouraged
by
Smuts
thereafter
shortly
and
operationswas not
126
finally
disagreement,
Belgians
After
further
the
to
much
opportunity seek
assistance.
force
leave
for
to
5,000
of some
to
their
an
occupation
ally and
agreed recruit
porters
127
back
2,000 Congolesetroops with the remainderbeing withdrawn
to their own colony.
The situation along the Gennan-Portugueseborder had been quiet following the
disastrousfailure of their advancein April and May of 1916. Despite the setback,they
retained ambitions of retaking the disputed Kionga triangle and capturing the southern
3,100
force
Portuguese
Mikandani.
To
some
this
third
of
expeditionary
port of
end, a
19
in
1916
June
the
to
had
Major
General
F
Gil,
supplement
arrived
soldiers, under
128
force
2,700
September,
By
in
Askaris
of
about
the
a
companiesof
country.
already
had crossedthe River Rovuma, occupying the northern bank and seizing the swampy
Kionga triangle that lay in the delta.129

123CampagnesColonialesBeiges,11,p. 62-63.
"" CampagnesColonialesBeiges,11,pp. 593-595.
12'CampagnesColoniales Belges,II, p. 532 for Brigade Nord and p. 574 for Brigade Sud
126
WO 33/858, TelegramsD 1, No. 1257,22 September1916,Telegram23102, CIGS to Smuts,p. 336. "... although the Belgian
Governmenthasexpressedreluctanceto extendthe operationsof its troops beyond Tabora,yet it has never, categorically, refusedto
do so, and it probably could be inducedto order co-operationin any way that you wished, with the object of disposingof the German
forcesretiring from Tabora. Pleasekeep me informed of your wishes in this matter"; No. 1316,18 October 1916,TelegramF 5867,
Smutsto CIGS, p. 356. "Even taking into accountthe heavy wastageof presentnumbersfrom malaria, I hope to be able to finish the
campaignwithout Belgian assistance.If the Belgians pressfor unacceptableconditions it will be better to finish without diem. I shall
not, however,be able in that caseto releaseSouth African Infantry as I had hoped,and the formation of additional South African units
for servicein Europewill be delayed."; No. 1320,21 October 1916,TelegramF 5880, Smutsto CIGS.
127
CampagnesColonialesBeiges,11,p. 596.
12sBoell, Die Operationen, 243-244. An
pp.
expeditionaryforce of about 1500had landedin 1914with anotherof similar size
arriving in February1916. Both had effectively beendestroyedby sicknessand the bulk of the soldiers evacuatedto Europe. Ile
third expeditionaryforce and the existing Askari companieswere split betweenprotecting the 600 kin long frontier and preparingfor
invasion;Cann,John,"Mozambique, GermanEast Africa and the Great War", Small Warsand Insurgencies,XII, No. 1, (2001), pp.
132-133.
"9 WO 33/858, TelegramsD 1, No. 1262,26 September1916,TelegramX 757, Smutsto CIGS, London, 337; Hordem, Military
p.
OperationsEastAfrica, pp. 388-391.

36

The Germanmain body had evadedencirclementin the Uluguru Mountains south

River
Mgeta
low-lying
in
the
Morogoro
the vicinity of
steamy and
of
and was now
their
from
Belgians
broken
had
the
making
The
Westtruppen
now
and
was
clean
valley.
detachments
Small
in
Mahenge
in
the
the
troops
area.
way eastward an attempt to rejoin
deployed
the
Ateilung
southern
British
on
was
the
a
weak
enclaves
and
watched
coastal
borderoppositethe Portuguese.
DETAILED

DISPOSITIONS

Looking now in more detail, the main body of the British force under General
Smutshad pushed south of Morogoro and swept around the sides of the imposing and
forest clad Uluguru Mountains. Major General AR Hoskins's I" East African (EA)
Division, composedof British, Indian and African units, had moved around the eastern
South
rd
Division,
3
EA
C
Brits's
General
Dutumi
Major
overwhelmingly
slopesto
while
African in composition, had moved along the western sides to Kissaki. Further to the
largely
South
African,
nd
Division,
EA
2
Major
General
JL
Deventer's
again
west,
van
had more or less ground to a halt in and south of Kilossa Station on the Central Railway.
Much further north and west, Brigadier General Sir Charles Crewe's brigade-sizedLake
Force was in the process of disbandment,having completed its work with its Belgian
E
Northey's
General
Brigadier
Finally,
to
the
allies.
south and west,
many miles
Rhodesia-NyasalandFrontier Force, or Norforce as it was usually known, was
having
Iringa,
hilltop
its
town
the
advancedthrough
of
consolidating positions around
forest
tangled
miles
of
and mountain to get there.
many
The 1" EA Division consisted of two understrengthbrigades, I" and 2nd EA
mustering four and six understrengthbattalions respectively. It had one mountain and
three field batteries of artillery together with four companies of mounted infantry and
130
cavalry. The exertions of the previous months togetherwith the onset of heavy rain in
130
Perry, FW (comp), History of the Great War.- Order of Battle Divisions Part 5a: TheDivisions ofAustralia, Canadaand New
Zealandand thosein East Africa, Newport, Gwent: Ray WestlakeMilitary Books, 1992,p. 85. Henceforth Perry, "Order ofBattle
Part 5a". The V EastAfrican Brigade was composedof 2d Loyal North Lancashire(2 LNQ, 2' RhodesiaRegiment (2 RR), 130'h
Baluchis, and a compositebattalion of the Kashmir Rifles (KR). The 2"d EA Brigade had 25' Royal Fusiliers (RF), Yd KAR, 129b
Baluchis, 40'hPathans,57'hRifles, 29thPunjabisand the Gold CoastRegiment (GCR). Unbrigadedwas the mixed race I" CapeCorps
(I CC) battalion from South Afirica.

37

late Septemberforced the division to halt on the north bank of the River Mgeta, centred
131
rd
Kissaki.
on the village of Dutumi while patrols linked up with 3 Division now at
The 2ndEA Division was madeup of two brigades,one mounted and one infantry.
The I't South African (SA) Mounted Brigade had five weak mounted regimentstogether
rd
SA Infantry Brigade was
3
battery.
The
with a motorcycle unit and an armoured car
basedon four infantry battalions while the division had one mountain and four field
batteriesof artillery in support.132Divisional headquartersremained at Kilossa while the
I" SA Mounted Brigade had been detachedto 3rd Division and a column from 3rd SA
Infantry Brigade had been pushedsouth to Kidatu, just north of the Ruaha River. By its
4,752
it
men capableof operationsand a
about
own reports, was greatly weakened,with
further 1,946unfit. 133
The 3rd EA Division was similarly composed to the 2nd with a mounted and
,
infantry brigade, both numberedthe 2nd The mounted brigade had four regiments plus
.
an armouredcar battery, while the infantry brigade had four battalions under command.
134
howitzer
batteries.
The mounted
Artillery supportwas provided by three field and one
troopshad beenreinforced as mentionedabove,but taken togetherthe two brigadescould
hardly muster 1,200 soldiers. It had concentratedat Kissaki where they met up with an
135
nd
2
Infantry
Brigade.
equally weakened
Norforce was an anomaly as it was technically under the control of the Colonial
Office, having been raised to protect the Northern Rhodesian and Nyasaland frontier
from Germanincursions. As the war progressed,the pretenceof independencevanished,
asNorthey, a professionalofficer who had seenserviceon the WesternFront, was careful
to liaise closely with General Smuts and to coordinate his actions with that of the main
force. He also put a great deal of care and attention into the preparationsfor the advance
"' Hordern,Military OperationsEast Africa,
pp. 371-374.
"2 Perry, Order ofBattle Part 5a, 90. The I' SA Mounted Brigade had V, 2", Yd, 0
p.
and9'h South African Horse (SAH), the SA
Motorcycle Corps and 4hArmoured Car Battery. The 3' SA Infantry Brigade had Vh, 10", 11' and IP South African Infantry (SAI).
'3' Hordem, Military OperationsEast Aftica, 374-377. Cites War Diary 2d EA Division 27 September1916.
p.
of
of
134
perry, Order ofBattle Part 5a, p. 94. The 2"dSA Mounted Brigade had 5th,6"', 7' and 8' SAH togetherwith 5'hLight Armoured
Car Battery. The 21 SA Infantry Brigade had 5h, Vh,Th and 8" SAI.
'" Hordem, Military OperationsEast Africa, 371.
p.

38

he
he
to
as
lines.
Consequently,
progress
his
good
make
able
was
as well as
supply
Although
the
German
into
was
opposition
the
colony.
advancednorth-westward
his
led
relatively weak, the combination of methodical planning and aggressiveness
136
His
August.
29
Iringa
hilltop
important
was a mixed
town
on
troopsto occupy the
of
it
in
Nyasaland;
Rhodesias
from
Africa,
South
all
the
formation with units drawn
and
137
Tactically,
fighting
3,700
troops.
4,700
were
all ranks, of whom about
numberedabout
it was split into three independentcolumns,with Colonel Rodger's South Africans eastof
Iringa at Hange, Colonel Murray's Rhodesiansfurther east near Muhanga, and Colonel
Hawthorn'sKAR was south-westof Mahengeat Mkapira, a distanceof over 200 miles of
difficult mountain, forest and lowland valley. All were further dependenton another 180
down
Nyassa
the
Lake
back
thence
to
lines
and
of communication
miles of vulnerable
138
Zanibesiriver to the coastof Mozambique.
On the other side, Colonel von Lettow's troops were split into three main
body
29
Railway
Central
the
of
in
main
the
was
groupings: the east and south of
Rufiji,
Mgeta
the
his
with
16
and
companies,with
under
personal command along
further
hundred
Several
Mahenge.
Kraut
west and now
13
miles
around
another under
Wahle
General
Major
WestbefehIshaber,
from
Tabora,
the
trying to withdraw east
was
SadbefehIshaber,
far
in
the
12
Finally,
appointed
the
newly
south,
with
companies.
CaptainLooff, was due to take over the three companiesfacing the Portuguese,making a
total of 44 overall.
The detailed dispositions were as follows: the main body had two sub-groupings
Uluguru
Dutumi,
the
bulk
Lettow
of
the
south-east
troops
near
with
under von
of
Mountains. Abt von Lieberman and Abt Schulz, eachof three companies,were facing 1"
Division supported by Abt Klinghardt of two companies and Abt Tafel with two
back
held
Stemmermann
in
Abt
battery
behind
with
them.
was
companiesand a
echelon
the equivalent of four companies. The Kommandeurkept two companiesas a personal

136Hordem,Military OperationsEast Aftica, p. 484.

137
CAB 44/4,p. 1, andSketch 67.
1311
Hordem, Military OperationsEast Africa, pp. 504-505 and Sketch 67; WO 95/5229,Norforce War Diary, September1916,
"Detailed Stateof Troops as at 30'hSeptember1916".

39

reserve. Further west and on the other side of the mountain range, Abt Otto with six
139
rd
holding
3
Division
Kissaki.
companiesand a gun was opposite
Major General Wahle was in the processof moving eastwardto rejoin the main
body near Mahenge. He had a force of about 12 companiesand two guns divided into
four groups. Two of the columns, At von Langenn and At Zingel, consistedof three
companiesand a gun; whereasAt

Wintgenshad five companiesand At Habener had

the equivalentof a companyand much of the support services.140
In the Mahenge-Iringaarea,Major Kraut assumedcommandof an enlargedforce
in mid-September. He had an enormousareato cover and deployedAbt Schoenfeldwith
three and a half companiesand two guns along the Ruipa River at Kidatu to prevent any
further southwardmove by 2ndDivision. At Lincke was given the task of holding the
line of the Ruipa River, midway between Mahenge and Iringa, with five and a half
companiesand two guns. Its mission was to contain Northey's columns operating out of
Iringa. Finally, At Krager consisting of four companies,one of which was mounted,
was placedto the south-westof Mahengenear Mkapira to guard the Schutztruppe'sleft
flank againstthe combined columns of Hawthorn and Murray. 141
Songeahad been left virtually unoccupied,owing to lack of troops, and once it
had been taken by Northey, a small German detachmentwas sent to prevent a further
advance. Finally, in the south and north of the Rovuma River was a three-company
strongAteilung facing the Portuguese.142

139Boell, Die Operationen, 23 1.

p.
t40Boell, Die Operationen, 284-285.
pp.
141
Boell, Die Operationen,pp. 251-253.
142Beell, Die Operationen, 244.
p.

40

CHAPTER 3

THE OPERATIONS OF SEPTEMBER 1916TO JANUARY 1917

PRELIMINARIES

TO THE REMAINING

OPERATIONS

OF 1916

I-, As the 1916 dry seasonended,General Smuts was increasingly unhappy with his
role in East Africa and worried about the effects of the campaignon his South Africans'
morale. As a soldier-politician and conscious of the vulnerability of his government at
home, the huge sick-rate and difficulty in keeping his units at even half-strength made
him anxious to complete the campaign as quickly as possible. Discontent with the
administrativeand medical arrangementswas growing as it was becoming impossible to
ignore the effects of severe under-feeding and the failure to evacuateor treat the sick
effectively.
Forced to halt by a combination of rain and exhaustion,the commander-in-chief
beganto considerthe next phaseof operations. Although the Germanshad been pushed
into as unhealthy and low-lying ground as the British, reaching them and then beating
them in battle would be a very difficult process. Lacking the resourcesto advance
further, he tried bluff, writing to Governor Schneeat the end of Septemberand calling on
him to surrender. However, this stratagemwas never likely to tempt von Lettow and the
letter was ignored.143No longer able to deludehimself that his opponentswould throw in
the towel if given one more push, Smuts now had to face the reality of a prolonged
campaign.
The ground would also play a major role in determining the course of future
operationsas the prospectiveareaof operationswas difficult in the extreme. In the east,
the low coastalplain extendedinland to the broad and low-lying plains of the RufiJi and
Ulanga River basins. North-west of Kilwa were the Mtumbei Hills which dominatedthe
coastaland delta areas. From there, the land dipped back into the fluvial plains of the
Mgeta, Ruf1ji and Ruaha rivers until it met the
northern tip of the Utschungwe
143Boell, Die Operationen, 233-234; Lettow,
Reminiscences,p. 158.
pp.

41

Mountains,a wide spur of highland that extendedfrom Iringa in the north to Lake Nyassa
in the south. Almost in the centre of this broad central plain was the Mahenge plateau
which toweredabovethe steamingforest below.
The lowlands were hot, humid and pestilential. There were no roads and
The
leading
from
few
footpaths
by
to
the
numerousrivers
village.
movementwas
village
Owing
jungle,
high
to the combination of
by
dense
swamp.
grass,and
were surrounded
broad, flat valleys surroundedby steep escarpmentsand massifs, they were known to
flood extensivelyduring the rains. The Ruf1ji itself was over 400 metres wide during the
dry seasonand during the wet, flooding could reach six miles on either side of its banks.
The physical difficulties were compoundedby the presenceof vast quantities of malarial
mosquitoesand the dangeroustropical diseasesthat were so prevalent throughout their

area.
The highlands,particularly around Iringa, were wild and thickly forested although
the climate was much less trying than in the valleys below. If the roads were poor and
better
difficult,
they
approachthan through the
communicationsalso
still offered a much
Rufiji basin. German-held Mahenge was known to be a population centre based on
fertile uplandsand a relatively healthy climate, but any approacheswould have to cross
"
from
River
Ulanga
from
Ruaha
the
River
the
the
the
north.
either
west or
PLANS FOR THE OFFENSIVE

Faced with these and other difficulties, Smuts began to look at options for
wrapping up the campaign. He wired his plans to the CIGS on 22 September,outlining
both the difficulties and possibilities of a future advance. He noted that the Germansno
longer appearedto be falling back onto the Mahenge plateau. They were deployed in
three major groupings;the main body under von Lettow to the south of the Mgeta River
running east and south toward the coast, a force under Kraut in the Mahenge area, and
finally the westerngroup under Wahle still in the vicinity of Tabora.
WO 153/111,Map of East Africa; Crowe, General SmutsCampaign, Sketch4; CAB 45n3, Sketch 74, "The StrategicSituation, I

42

While Smuts was naturally focused on dealing with von Lettow's main body in
the east,he was becoming increasingly concernedabout the progressof Wahle's force in
the far west. Abandoning Tabora on 19 Septemberto the Belgians, it was believed to be
making its way eastwardto rejoin von Lettow, probably via the Great Ruaha or Ulanga
River valleys. The gap between2ndDivision and Norforce was still more than sufficient
to allow the Westtruppenan uninterrupted passageto the east, which in turn would
threatenthe vulnerable lines of communicationsincluding the small and isolated garrison
in Iringa.
Smuts was quite clear that his Belgian allies would not pursue the
withdrawing force owing to severe constraints in their supply system, lack of political
145
indeed,
he
had
will, and
madeno effort to convince them otherwise. His own troops in
the area,the Lake Force, were unable to move further south owing to lack of sufficient
transportand were in the processof disbandment.
In light of the situation, and despite his tremendousdifficulties with supply and
health, Smutsproposedto keep the bulk of 2ndDivision betweenthe Great Ruaha River
and Kilimatinde to protect his lines of communicationand to prevent Wahle from linking
P
Kraut.
The
Is'
Divisions would advance south from the Mgeta to the
up with
and
Rufiji while a brigade recently landed at Kilwa would advancenorth-eastwardsto Utete
on the lower Rufiji. Northey's force was expectedto push east from Iringa and reach
Mahenge within a few days. In the far south, Smuts left offensive operations to the
newly arrived Portuguese Expeditionary
146
from
them.
assistance

Force, although he expected no practical

Just five days later, his opinion had changedas the sick rate along the Mgeta rose
alarmingly. Learning that the area between there and the Rufiji became a vast swamp
during October and November, he pulled back Brits's 3rdDivision to Morogoro for
rest
and recuperation. Modifying his plans, he decided againsta major advanceon the Rufiji
from the north and proposedto cut down the troops
on the Mgeta to a brigade. Kilwa
December1916".
145WO 33/858, TelegramsD 1, No. 1248,20 September
1916,TelegramOA 411, Smutsto CIGS, p. 333.
146WO 95/5291,War Diary GHQ, 22 September1916,Appendix
32a, TelegramOA 412, Smutsto CIGS, 22 September.

43

from
Lettow
the
brigade
then
operate against von
would receive a second
and would
nd
2
light
In
Wahle's
south-east.
continuedwithdrawal eastwards, Division was ordered
of
to sendits infantry brigade to Nyukwa's, a ford on the Ruaha River, to be followed later
by the mountedbrigade, currently supporting 3rdDivision. 147
A personalvisit to van Deventer's division in early October confirmed his worst
fears,asthe route forward from Kilossa Station to Nyukwa's:
"... proceeds30 miles through the area which must be swamp or lake in rainy
seasonand requires heavy engineeringwork in other parts. Troops sent that way
fast
On
be
in
from
same
approaching.
now
season,
must
cut off
supplies rainy
ground operationsfrom KISSAKI to RIFIR [sic] are ruled out. In view of these
facts and seriousstatehealth of troops and need for thorough rest I doubt whether
148
,
final advanceagainstenemycan be resumedbefore beginning of January.
These gloomy prognostications were confirmed shortly afterwards by a report
from 2ndDivision that poor roads and lack of transport would prevent it from moving
before the first week of November.149This unwelcome news meant that Northey would

be on his own for the critical period when pressurefrom both west and east was
beginningto develop.
Despite having had his plans acceptedin principle by London, Smuts could not
ignore the dire physical state of his force, which urgently required rest and
reorganisation. Optimistically, he informed Robertson that he believed that his supply
situation could be resolvedwithin a fortnight and that the lines of communicationrunning
north to British East Africa could be closed down as soon as the Central Railway was
reopened. This was temperedby the advice that, as the rains beganin late Septemberthe
low-lying river valleys would become increasingly waterlogged and impassablefor any
" WO 95/5291,War Diary GHQ, 27 September1916,Appendix 36a, TelegramOA 491, Smutsto CIGS, 27 September. This
change
of mind is puzzling as the facts aboutthe climate and geographyof the region had beenknown to his staff since early September.
141WO 95/5292,War Diary GHQ, 8 October 16, Appendix 10a,TelegramOA 615, Smuts CIGS, 8 October.
to
149WO 95/5292,War Diary GHQ, 10 October 1916,Entry 1 2"d&r Dtvislons. The building
of roadsto Iringa would take at least
one month while transportwas barely sufficient to cover day to day needsand was completely inadequateto build up stocks.

44

troops
it
to
In
many
term,
as
the
withdraw
essential
transport.
was
short
sort of wheeled
forward
in
the
Central
Railway
back
the
to
to the
problems of supply
ease
as possible
areas.
While the above assessmentwas superficially correct, it glossedover a number of
his
develop
formations
forward
to
desire
supply
Smuts's
to rest the
and
problem areas.
lines was sound,but grossly underestimatedthe time and resourcesrequired. The troops
had alreadypassednearly 50-100 miles beyond their original destination on the Central
Railway and there had been no opportunity to stockpile supplies. The railway itself was
Korogwe
from
300
the
had
on
at
railhead
to
travel
miles
over
out of action and supplies
be
links
before
be
the
Northern
It
Railway.
would
time
and
rail
port
the
some
would
fully operational,and until then it would be impossible to build up the large stockpiles
horse
fly
for
further
tsetse
The
the
sicknessruled
and
ravagesof
necessary a
offensive.
600
despite
the
new wagons
over
of
transport
arrival
out any sustaineduse of animal
"'
Mechanical
transport,
long
be
kept
June;
enough.
alive
animals could not
since
did
but
help,
they
light
Jigger
300
Ford
not reach
cars,offered some
particularly the
extra
151
insufficient
Furthermore,
there
forward
September.
were
the
areasuntil the end of
drivers or vehicles to build up large forward reservesof food and ammunition. The poor
in
dramatically
the nearing rains;
likely
few
to
worsen
condition of the
existing roadswas
in the low-lying areas they would vanish under water altogether. This left the only
If
the
human
of
war.
pressures
under
alternativeof
porterage,a systemalready suffering
highly inefficient as a meansof moving material, it was flexible and could cross ground
becoming
However,
themselves
the
that
were
carriers
or water
no vehicle could attempt.
increasingly difficult to recruit (or coerce) as the military requirements escalatedand
casualtiesmounted.
It was also an unpalatablefact for the British commandthat, even if the railway
could deliver supplies from the port of Dar-es-Salaamto Morogoro or Kilossa, those
"o Blenkisopp,Maj-Gen Sir LJ
and Rainey, U-Col J W, History of the Great War: Basedon Official Documents: Veterinary
Services,London: MISO, 1925,pp. 416-417. Henceforth, VeterinaryServices.
151
Crowe, GeneralSmuts'Campaign, p. 217-218. Animals seldom lastedlonger than six weeks and it generally took a month to
breakthem in beforethey could be usedto pull wagons. The Ford Cars arrived in August and it took over a month to bring them into
full operationaluse.

45

from
line
The
forward
from
distance
the
areas.
railheadswere still a considerable
been
it
had
difficult
Mgeta
as
Mikesseon the Central Railway to the
was notoriously
The
improvised
forest
rivers.
crossingsof
with many
carvedout of mountain and virgin
from
but
less
patches
from
Kidete
Kilossa
extensive
suffered
to
mountainous,
was
route
it
Furthermore,
impeded
movement substantially.
of sand and black cotton soil that
for
long
the
few
belt"
passage.
bad
"fly
after
survived
through
animals
and
a
very
passed
In his haste to cut off von Lettow, Smuts had ignored the fact that the landings
had
diverted
Ocean
Indian
scarce
ports
the
to
remaining
and shipping needed capture
the
immediate
task
of
from
port
of
clearing
the
and
critical
away
and
resources
shipping
Dar-es-Salaamand developing the lines of communication along the Central Railway.
He had eliminatedthe risk of anotherblockade-runnerlanding supplies,but at the cost of
delay to his own logistical build-up. This was far from academicas the fighting troops
Without
a
in
a state of semi-starvation and severely afflicted with sickness.
were
functional port and railway, he had absolutely no chance of bringing forward the
improve,
let
the
situation.
food
to
alone
stabilise,
needed
quantitiesof
and equipment
Grand plans for sweepingencirclementswere fine for the map, but the enforced
immobility of the 2ndDivision in the west and the 3rdDivision at Morogoro, togetherwith
Northey's
troops
Kilwa,
that
were
build
l't
Division
the slow
only
the
meant
at
up of
152
fact
Given
future.
that
the
in
so
the
near
capableof any sort of offensive operations
decided
Smuts
disease,
by
debilitated
on medical grounds
many soldiers were seriously
153
A series of medical boards
to reduce numbers and to repatriate the worst affected.
South
12,000
bleak
to
the
the
some
answer was
were set up
assess
situation and
Africans were medically downgraded and ordered home. This representedsome twoFor
the
April
the
force
in
March
had
the
same
year.
thirds of
of
that
and
arrived only
for
1"
Division,
British
Indian
arrangementswere made
predominantly
and
units of the
the piecemealreplacementof the worst affected units. It would be severalmonths before
the worst-affectedunits would be ordered to Egypt or India, as ongoing operationsand
152WO 95/5292,War Diary GHQ, 10 October 1916,Entry It 2d and 3rdDivisions. Van Deventer informed GHQ that his formation
could not move until at least the beginning of November.
153WO 95/5300,War Diary DMS, 24 September1916,Appendix V, Letter DMS to DA&QMG, 24 September.

46

but
indigenous,
"'
It
delayed
the
still small,
their
was
shipping shortages
withdrawal.
KAR that survived the rigours of the campaignbest."'
The first formation to be disbandedwas the exiguous 3rd Division under Major
GeneralBrits on 15 October. It had withered away during the final advanceto the Mgeta
behind
left
brigade
to reinforce van
was
and only a much reduced composite mounted
Deventer.The now redundantLake Force was broken up on the following day. In the 2nd
Division, the weakest units were amalgamated and joined by the survivors of Brits's
decimated force, amounting to a reduced Ist SA Mounted brigade and 3rd SA Infantry
Brigade.

The Is' Division moved to Kilwa with the 2nd and newly formed 3rd EA

Brigades while the Is' Brigade remained on the Mgeta Front. It and the Force Reserve,
the reduced 2nd SA Infantry Brigade, remained under the direct control of General
SMUtS.156

In comparison, Norforce's losses through illness were substantial, although

This
large
bad
Smuts's
not
required.
was
nowhere near as
scale evacuation
as
and
but
it
had
the
healthier
the
also
great
through
operated,
which
reflected
slightly
climate
by
Northey.
General
to
personalattentionpaid medical and supply matters
THE MGETA LINE

While the high command was pondering its future moves, much of the force
remained in close contact with the Germans separatedby the swampy and unhealthy
Mgeta River. Neither side relished the unpleasantliving conditions nor the densebush
that madeit ideal for raids and ambushes.In fact, it was singularly ill-chosen as a forward
line with swamps extending for some distance on either side of the river and being
distinctly malarious. Even by East African standards,the lines of communication were
exceptionally difficult and the humble porters bore the main burden. Both sides had to
1`4WO 95/5292,War Diary GHQ, 25 January1917,Appendix 46a, TelegramOA 846, Hoskins to Ewart 25 January;Appendix 46b,
TelegramOA 847, Genstaff,Dutumi to Genstaff, Dar-es-Salaam,22 January.
155Moyse-Bartlett,Lt-Col H, TheKing's African Rifles, Aldershot: Gale & Polden, 1956,
pp. 332-333 and 70 1. At I July 1916,the
KAR stood at nearly 8,300 all ranks; by I January 1917the numbershad risen to 15,600.
"6 WO 95/5292,War Diary GHQ, 15 October 1916,Entry P 2"d&P Divisions; 16 October 1916,Entry Lake Det. & Belgian
Force; 24 October 1916,Appendix 29, TelegramF 5892, Smutsto CIGS, 24 October.

47

dust
dense
clouds of
struggleto move supplies forward on narrow tracks that produced
keep
difficult
it
to
found
Germans,
in
The
dissolved
too,
very
rain.
when dry and
feed
in
the
troops.
to
trying
effort
considerable
suppliesmoving and expended
As the rain fell, and operationswound down, much effort was expendedin trying
These
by
disrupt
achieved no
to
the other side's supply chain
small-scale patrols.
decisiveresults, but did much to lower morale and increasethe senseof isolation. The
holding
Mgeta
Vt
the
largely
Brigade
the
EA
while
troops
continued with
withdrawal of
As
further
Kidete
to
the
forward
held
Bde
2nd
SA
Inf
west.
the
positions at
remnantsof
in
Africans
the
labour
large
front
cutting of
engaged
were
the
of
gangs
remained static,
bridges
the
improving
the
numerous rivers
across
the
engineersworked on
roads while
November
in
October
for
Unfortunately
and
the
them,
continuing rain
and streams.
157
into
dissolved
"road"
large
the
mud.
this
of
sections
undid much of
valiant effort and
In October, the British sensibly abandonedattempts to hold the southern bank,
keeping only piquets on the northern side with the main body pulled back north onto
higher ground. The Germansrespondedsimilarly, keeping only At Otto on the forward
118
Otto's force numbered
line with the remainder of troops drawn back to the Rufiji.
159
He
two
950
14
concentrated
guns.
and
someeight companieswith
machineguns,
rifles,
Alt
junction
holding
track
bulk
the
his
20
at
the
troops some miles south of the river,
of
160
from
Dutumi.
Kissaki
Behobeho,wherehe could reactto any movescoming
or
NORTHEY'S ADVANCE IN THE SOUTH
Norforce was handicappedby a serious insufficiency of soldiers for the task at
hand. It was a brigade-sizedcommandthat fielded less than 4,000 rifles, split into three
161
high
Despite
the
by
200
terrain.
columns,separated
miles of wild and mountainous
157CAB 44/6, p. 8; WO 95/5329, War Diary Norforce, Entry 30 November 1916. Northey's difficulties began slightly later with the
first rains starting on 30 November.
"' CAB 44/6, p. 11.
15"MS Boell, N14/30,26. Kapitel,
pp. 1903-1904.
160MS Boell, N14/30,26. Kapitel, p. 1906.
161Brown, Ian Malcolm, British Logistics the WesternFront 1914-1919. By comparison,in early 1915,the BEF held 50 km of
on
front with 350,000men. A typical brigade frontageon the WesternFront would have beenabout one mile with its lines of
communicationa maximum of five miles.

48

lack
his
that
Northey
his
was rightly concerned
quality of
columns and their commanders,
Westtruppen
detail
in
defeated
being
the
dangers
move
should
the
of
of concentrationand
disease,
from
his
As
although
the
and
troops were suffering
effects of climate
east.
well,
nowherenear as badly as the main body to the north.
In early September,Smuts instructed Northey to advance on and occupy the
Mahengeplateau.162The plan was for Norforce to move westwardsthrough the Ulanga
body
German
before
Mahenge
the
main
withdrawing
plateau
valley, then to ascendthe
his
Mahenge
itself.
The
that
grand plan to cut-off
of
was
part
capture of
reached
place
the Schutztruppe;this, coupled with the advancefrom Kilwa would make further retreat
impossibleand force a surrender.
As was becoming usual, this plan was based on major assumptionsthat were
100
Northey
false.
from
Apart
to
miles across
another
advance
quickly proven
expecting
it
that
the
flood
during
the
presupposed
season,
rainy
an, enormous
short
plain
Westtruppenwould pose no threat to his exposedlines of communication."' The second
:s that Smuts's
further
for
in
offensive
condition
troops
absolutely no
wa.
own
were
operationsand restoring the Central Railway to full capacity was going to take months
Germans
be
it
the
to
it
Finally,
than
that
prevent
possible
would
rather
weeks.
presumed
from breaking through the 75-100 mile gap that would remain between Norforce and the
2 ndDivision.

Furthermore,the strength of Wahle's force, still some 350 miles to the west, was
seriouslyunder-estimated;it was deemedto be only 1,500 strong and ready to surrender.
In fact, it exceeded2,500 and was determined to fight its way through to Mahenge.164
Even taking the lower estimate of 1,500 at face value, it was still sufficiently large to
inflict serious damage on Northey's dispersed command, not to mention his virtually
defencelesslines of communication. The option of a continued Belgian pursuit of the
162
WO 95/5291,War Diary GHQ, 2 September1916,Appendix 4, TelegramF 5213, Smutsto CIGS, 2 September.
161WO 95/5291,War Diary GHQ, 3 September1916,Appendix 6, Intelligence Report for 3 September1916. smuts,s staff were
certainly awarethat the Ulanga and Ruhudje Rivers would be impassablein the rains and later evidenceshows his understandingof
the problem. How he expectedNorthey to survive at Mahengeis unknown.
164
MSS Afr. s.1715(300), Williams, Col H P, History of I KAR, p. 158; CAB 44/4, pp. 2-3; Meinertzhagen,ArmyDiary, maps 17-19;
CAB 44/4,p.1, Sketch67.

49

p-a
.1

let

.4

.o.

U6-:

,-ss

IC-

Ar

IaI

V.

611

't

.
z

. '"t ....

..

*.
.

qlL

6 Z.

. -1

i!

:a
tl

Z-

t: t 0

t'

't

Westtruppen does not seem to have been seriously considered.

Despite Northey's

to
issue,
Smuts
the
this
continued
to
commander-in-chief
on
repeated representations
treat the threat as a minor one and provided no real assistance until it was nearly too
late.165

Despite these misgivings, Northey resumedhis advance in mid-September. He

realisedthat there were insufficient resourcesat Iringa to support a major advance and
decidedto sustain operations from Lupembe, further to to the south-west. To achieve
this, he left Rodger's small column at Iringa, and ordered Murray to march south-eastto
link up with Hawthorn at Mkapira on the Ruhudje River. Once united, the combined
lowlands
the
through
draw
from
Lupembe
and advancenorth-east
columnwould
supplies
towardsMahenge."'
Murray began his preparatory move on 16 September,moving forw ardf rom
Iringa through Muhanga on the easternedgeof the escarpmentsome 5,000 feet abovethe
Ulanga valley. Brushing aside minor resistancefrom At Braunschweig, he began the
join
instructions
Under
to
descent
into
the sweltering and steamy valley.
arduous
Hawthorn as quickly as possible, he was also looking for potential crossing points along

theriver ashe movedsouth.

While these movements were underway, Colonel von Lettow was making
nd
incapable
2
Division
Realising
Mahenge
that
to
the
of
was
arrangements reinforce
area.
further advancefor the immediatefuture, he despatchedMajor Kraut to take chargeof the
defenceswest of the plateau. Despite British fears that it would be a major centre of
defence,it was in fact very weakly held and servedas an administrativehub. The local
commanderdisposedof no more than 80 rifles, while farther to the southAbt Songeaheld
the rich areafrom Wiedhafento Songeato the Portugueseborder with about 30 rifles and a
numberof auxiliaries. West of Iringa on the valley floor was Abt Braunschweigwith five
"" WO 95/5329,War Diary Norforce, Entries 16
and 30 Septemberand I October 1916;WO 95/5291, War Diary GHQ, 24
September1916,Appendix 33a, TelegramOA 445, Smutsto CIGS, 24 September. Smutssignalled "According to Germanwireless
messageinterceptedthe enemyhave ordersto retire to the ITUMBA Mountains South-Eastof KILIMATINDE. I have advised
NORTHEY to keep a small force at IRINGA and have orderedVAN DEVENTER to senda force to MAHORE and NJUKWA.
1" WO 95/5329,War Diary Norforce, Entries 28 August
and 16 September1916. Northey consideredthat advancingto Mahenge
would put too great a strain on his alreadyoverstretchedsupply lines.

50

companiesof 900 rifles and 10 machine guns, supported by 12 FK further south at

Tanganikawith another250 rifles and2 machineguns.
Kraut brought important reinforcementssouth with him, some ten companies,12
machinegunsand six guns,totalling about 1200rifles; by the beginning of September,the
167
force
17
24
2,450
6
total
amountedto about companiesof
rifles,
machinegunsand guns.
The Mahengeforce was now strong enoughto prevent a further advanceand offered the
chance to deal a sharp blow to any isolated column. Most importantly from the
Kommandeur'spoint of view, Kraut was a dependableand aggressiveofficer who would
not give up ground without a fight

Kraut's aims were twofold; the first was to block any movementeastwardfrom
Iringa and the secondwas to preventa link up betweenNorthey and van Deventerover the
RuahaRiver at Kidatu. He correctly recognisedthat the former posedthe greaterthreat and
decided to leave only Abt Schoenfeldas rearguardat Kidatu to hold the enfeebled2nd
Division. He then began to move his powerful force southwestof Iringa and took Abt
Braunschweigunder command. As ever, the supply situation was a major concernas the
important magazineat Ifakara on the Ruaha and his own lines of communicationneeded
168
foodstocks
low
protection,while
were
owing to a shortageof carriers.
While Kraut was moving south,A bt Braunschweighad beenthinning out its forward
positionson the escarpmentand moving into the Ulangavalley owing to a fear of being cut
off. Halting the move, Kraut decidedto give Braunschweiggreatersupport. He placedAt
Lincke in the valley to the rear of the latter's forces, while detaching three companies
further south to block possible exit routes from the Utschungwe Mountains and
169
maintaining a small reserve.
This shift was detected by the British, and Northey was growing increasingly
concernedabout having to deal with Kraut in the east at the same time as Wahle was
167Boell, Die Operationen
p. 250.
'" Boell, Die Operationen:
pp. 250-25 1.
169Boell, Die Operationen,
p. 253.

51

in
Iringa
in
Deventer
to
He
to
the
order to
so
south
push
moving
wanted van
west.
his
former
but
Norforce
the
the
threat,
to
move
until
could not
concentrate
meet
new
shrunkenunits were reorganised. Furthermore, Smuts remained anxious to continue the
advanceon Mahenge."
Hawthorn had arrived opposite the Ruhudje crossing of Mkapira on 10
September,and discovereda strong enemyposition held by Abt Kriiger on the west bank.
After careful reconnaissance,his column forded the river and manoeuvredthe Germans
out of their defenceson 27 September. The next day KrOger led a strong counter attack
'
71
dislodging
British.
500
5
the
with some
rifles and machine guns with the aim of
However, Hawthorn was an able commander and was well prepared for the move,
holding off the assaultseasily. With Murray now only 10 miles to the north-west, the
Germansbroke off the battle and withdrew further east.
Despitethis successand the imminent arrival of Murray, plans for the advanceon
Mahengewere now altered. Hawthorn had been informed van Deventer had not even
startedto move and would not be a position to so for at least a fortnight. This meant that
he would have to face the full weight of Abt Kraut while operating from an isolated
forward position, linked to his baseat Lupembeby a very difficult bush trail. He decided
to hold firm and withdrew to the west bank of the Ruhudje where he was joined by
Murray on 29 September. Togetherthey set up a strong defensiveposition while waiting
for word of van Deventer's move. At the sametime, they attemptedto bring forward and
172
from
Lupembe.
stockpile asmany suppliesas possible
Much ftu-ther to the east, von Lettow was viewing these developmentswith
disfavour. He was concernedwith what he consideredhis subordinates'propensity to
dispersetheir forcesunwontedlyand he wantedto strike a blow againstwhat he considered
a vulnerableforward column. Accordingly, he issuedordersfor Kraut to attackHawthorn's
170WO 95/5329,War Diary Norforce, Entries 16,17
and 21 September1916;GSWA, Box 26,11 September1916,TelegramK 475,
2 Div to Smuts. By mid-September,van Deventerwas reporting that his troops were largely incapableof further effort owing to
diseaseand lack of food.
l" Boell, Die Operationen, 254.
p.
172MSS Aft. s.1715(300), History I KAR,
of
pp. 159-160.

52

column at Mkapira on I October, although the latter had only six companies in the
immediatearea,and requiredreinforcementfrom At Falkenstein.173Despitevon Lettow's
impatience,it was impossible to regroup so quickly in the difficult conditions of the
Ruhudjevalley and Kraut would require severalweeks to make adequatepreparationsfor
theattack.
While these moves were taking place, Northey was given yet another task for his
stretched forces. The town and rich farming district of Songea was guarded by less than
a platoon and its denial to the enemy would reduce their room for manoeuvre and help
restrict them to the central portion of the colony.

Equally importantly, its loss would

reduce their food-producing capability considerably, although it is not clear whether

Smuts considered this in his decision-making.
reinforced by a newly-raised African

Accordingly,

as soon as he had been

battalion, half of it was despatched as an

independent column to Songea. Sailing across Lake Nyassa by steamer, it landed at

Wiedhafen on 12 September and immediately marched inland. The tiny German garrison
there evacuated the place on arrival of the column and it was occupied without a fight on
20 September.174

It was not until 25 Septemberthat Kraut heardof the landings and he immediately

I
orderedFalkensteinto secureSongea. Learning the next day that the British had already
taken that place, he reinforced Abt Falkenstein and agreedto its withdrawal to Likuju to
maintain observation of the enemy. He also decided to detach two of Lincke's
companiesto the south to deal with the growing menace,only to be over-ruled by von
Lettow on 30 September.He was concernedabout the dispersionof Kraut's force as well
175
Liwale.
as aboutprotectingthe open road to the undefended
The situation did not remain quiet as the Songeaforce was ordered to advance
further eastto MkuJu, reaching it on 5 October. Again, the Kommandeur was unhappy
with the loss of initiative and he ordered Major von Grawert, the military commanderof
173Boell, Die Operationen,
p. 254.

174
WO 95/5329,War Diary Norforce, Entry 20 September1916.
17-
Boell, Die Operationen,pp. 253-254.

53

Mahenge, to attack on 12 October with three companies now moving south to reinforce
the threatened area. Ominously, the mood of the African population was changing for the
deployments
German
information
British
and also
the
they
on
with
worse as
provided
beganto attack isolated posts and their lines of communication. This region had suffered
heavily during the Maji-Maji

for
love
had
the colonial regime.
no
rebellion and

Von

Grawert was well aware of the disaffection and became decidedly cautious about rear
176
area security.

THE SWITCH OF EMPHASIS TO KILWA

As General Smuts had told the CIGS, the obstaclesof the Mgeta line made it
bulk
I"
Division
to the coast
He
the
to
to
of
worthwhile switch emphasis. wanted redeploy
177
to
the
its
base.
Initially,
Kilwa
this
opportunity
the
offered
option
using
seaportof
as
lines
before
line
River
the
Matandu
of
the
the
on
enemy
moving
strike west along
of
East
Africa.
German
to
the
the
of
south
communicationas well as cutting off
main route
However,while Kilwa Kisiwani had an excellentanchorage,there were absolutelyno port
facilities and it would be necessaryto build up the basebefore any advancein strength
September,
8
be
having
been
the remainder of the
Kilwa
on
occupied
could
attempted.
be
formation
first
The
landing
to
in
facilities
developing
its
troops.
and
month was spent
P
deployedwas Brigadier General Hannyngton's newly established EA Brigade and it
178
by
first
in
October.
was complete the
week
Hannyngtonwas instructedto push rapidly inland in the hopes of cutting off the
Early
forces
further
decided
Hannyngton
to
reconnaissances
quickly.
act
enemy
north.
revealed that the village of Njinjo, some 42 miles inland along the Matandu valley,
provided a jumping off point for either a move north towards the Rufiji, or alternatively,
south towards the food-producing areasof Liwale. However, a secondand more direct
route was found. Instead of following the Matandu inland, it was possible to cross that
'76Boell, Die Operationen,p. 255.
177
Therewere two placesknown as Kilwas; Kilwa Kivinje was at near the mouth of the Matandu River and was usually known as
Kilwa; Kilwa Kiswani was locatedon a island to the south of Kilwa Kivinje and had an excellent deepwater anchorage.It would be
subsequentlydevelopedas the port.
172
WO 95/5292,War Diary GHQ, I November 1916,Appendix 1, TelegramF 5948, Smutsto CIGS, I November.

54

Hills.
Mtumbei
following
to
the
the
track
river
north-west

The key to the area was the

junction at the hilltop station of Kibata that dominated the local region. From there, a route
lower
Rufiji
lay
Utete
hills
the
to
through
the
and was a major
on
which
ran
straight
German supply centre. An advance through the Mtumbei Hills promised to be much
healthier
than the coastal
Matandu
they
than
the
and
were also cooler
shorter
route, while
179
plain.

180Captain Looff, learned of the landings, but decided to

The OstbefehIshaber,
forces.
bare
This
Rufiji
leaving
Utete,
the
the
order
of
virtually
areanorth of
concentrateat
harvesting
Looff
by
Lettow
to
and
concentrate
on
who wanted
was countermanded von
collecting the crops in the lower Rufiji. He declined to do so and was promptly sacked.
On. 12 September,Major von Boemken was designatedas OstbefehIshaberwhile Looff
wasdespatchedto the southernborderto take up the new appointmentof SiOsbefehIshaber
Looff
Sternmermann
Captain
Portuguese
took
the
threat.
s troops
over
against
growing
together with responsibility for the defence of the food producing areas of the lower
Rufiji. 181

In order to have a more direct presence, von Boemken and his four companies were
by
journey
lengthy
September.
After
for
12/13
Kilwa
to
the
a
ordered march
on
evening of
river steamer and much marching, they reached Kimbarabara, on the southern slopes of the
Mtumbei Hills on 25 September. Only a few miles from Kilwa, von Boemken was still
some distance from Njinjo, which apart from its key location also housed an important
182
magazine.

The British were also awareof Njinjo's importanceand patrols had alreadyreached
there. Desperatelytrying to evacuatethe valuable stocks of food to Mpotora, on 22
Septembervon Lettow sentan urgenttelegramto von Boemken,ordering him to securethe

magazine.However,thelatterwasmoreconcerned
with his own securityandwithdrewthe
179
Moyse-Bartlett,TheKing's African Rifles, p. 336.
'soBoell, Die Operationen,p. 235. The former Kommandant Dar-es-Salaam,Looff,
of
was given the new title of Ostbefehishaber,on
25 June 1916. He was maderesponsiblefor the defenceof the coastalregion.
Boell, Die Operationen,page238.
Boell, Die Operationen,p. 238.

55

rt, A

1
;::?

JW

1sJ

tq""-.,

*!
'1i
go

i-I

NIC j6

.--.

ki

ii

_:.

S.

lag

l
1
1

rl

I" L-Z41
.
,

4 -Z

ee

qr.

!i

0,

14.

Ii

417!

Il

1.

u
.III

1
.1

e.

If

lip

IV

4qk
It

10,

La

-4
40
10
'a-1

.1

i
'.
%

-1

ol

11
116

A J6

bulk of his Atedung to the west, leaving Njinjo open to a raid that that destroyed most of
the supplies and forced the evacuation of the village. Instructed to attack, a reluctant von
Boemken launched a half-hearted assault that was easily repulsed and then returned to
Kimbarabara.183

This left the Matandu Valley open and Hannyngton's troops occupied Njinjo
without opposition on 7 October. The misfortunes were capped two days later, when
184
Njinjo.
It was an
reinforcementssent from Liwale were ambushedand scatterednear
unpromisingstart for the Germancounter-movesagainstKilwa.
Von Boemkennow shifted his defensivepositionsto the west of Njinjo in order to
block the road to Mpotora. Unfortunately for him, this meant that he was no longer in
telegraphiccommunicationwith von Lettow, and messengerswere now the only meansof
passinginformation for the next eight days. The lack of adequatecommunicationsand
misunderstanding
now led to a blunderthat left the roadto Kibata opento the British.
Von Lettow had wantedto ensurethat Kibata was adequatelyprotectedand had told
von Boemkento sendtwo companiesthere. However,they left with out-of-dateorders,and
neithercompanyfell back on the critical road to Kibata and the Mtumbei Hills as intended;
onewent easttowardsthe coastandthe other withdrew to the west. This left the main route
from Kilwa to the Rufiji River unguardedat the sametime as the British decidedto move
northwards.Hannyngtonsenttwo battalionsto occupy Kibata and the surroundinghills on
10 October,taking the unoccupiedfort four dayslater. One battalion left to garrisonKibata
while the other was withdrawn back to Kilwa. At the sametime, the brigadepushedone of
its units to a position four miles west of Njinjo, at Mchemera,and a screenof outpoststo
the north andwest was established.

On receiptof the news,von Lettow realisedthat his forceshad effed and senta
companydown the coaston 14 Octoberto reinforcethe isolatedtroopsin the Mtumbei
Hills.'Anotherwassentto join Abt vonBoemken,
but like its counterpart,
it arrivedwell after
181
Boell, Die Operationen,p. 238-240.

56

food
in
it
the
British
to
supplies
to
the
of
the
evacuation
guard
order
north
and withdrew
from the magazine at Tawa. This undoubted failure led to the second command change in
two months, as on 17 October the Kommandeur ordered Captain Schulz to take command of
the Kibata area and sent von Boemken south to assist Abt Looff against the less capable
Portuguese.185

A great deal of time was then spent on patrolling and the gathering of accurate
infonnation on the ground,asthe few mapsavailablewere poor and frequently misleading.
Despite his concernsabout the difficult situation being faced by Northey's forces much
ftu-therto the west, GeneralSmutsdecidedto make a personalreconnaissanceof Kilwa at
the endof October. Clearly impressedby its potential,he saw the Kilwa Forceasplaying a
key part in his plan to surroundvon Lettow, with one elementmoving westwardalong the
Matandu and the other striking north-westwardsfrom the Mtumbei Hills. His chances
Germans
indications
that
the
favourable,
there
were moving
were no
seemed
especiallyas
from the Rufiji area.186He decidedto sendanotherbrigadeto the areaand to place Major
187
into
deployments
These
in
GeneralHoskins and his I st Division
were put
command.
Kilwa
I
Division
the
Smuts'
areaon
of
command
assuming
effect shortly after
visit with st
188
November.
15Novemberandthe units of the 2ndEA Brigadecompleteby 29
THE FIGHT AGAINST THE WESTTRUPPEN

VVhileboth Northey's and Kraut's columns were manoeuvringfor advantage,the

Wesuruppenwere moving eastward. Although telegraphiccommunicationsbetweenthe
two main groupshad beenlost with the cutting of the Central Railway in August, General
Wahle had gained important intelligence. He still had his wireless receiver and had
intercepteda number of useful British transmissions. From these he learnt that Dar-esSalaamhad been lost and that the main body had been forced south of Morogoro. More
1" CAB 44/5, p. 15. The Germanslost 18 deadand 18 wounded out of 60 present. British losseswere two killed and six wounded.
"' Boell, Die Operationen,pp. 241-242.
'36WO 95/5292,War Diary GHQ, I November 1916,Appendix 1, TelegramF 5948, Smutsto CIGS, I November. Smutsnoted that
someten Germancompanieshad withdrawn from the Mgeta acrossthe Rufiji with somereinforcementof Utete.
'37WO 95/5292,War Diary GHQ, 7 November 1916,Appendix 9, Telegram OA 969, Genstaff, Morogoro to Genstaff, Dar-esSalaam,7 November.
18'SmutsDespatchIII, p. 138.

57

appositely,these communicationsrevealed that Iringa. was only lightly held and was
dependent
on supplylines running back to Neu Langenburgat the headof Lake Nyassa. He
decidedto march and attack Iringa. so as to disrupt the British advanceand relieve the
pressureon von Lettow's easternflank. In order to concealhis true intentions,Wahle then
spreadrumours and false messagesthat his objective was Kilimatinde on the Central
Railway:a plausiblealternative.189
After evacuating Tabora on 19 September, the Westtruppen had withdrawn
southwardsto Sikongewhere they regroupedand the companieswere reorganised. While
still largely intact, they now had the dauntingtask of crossingsome 300 miles of largely
undevelopedand uninhabited land to hinga. The countryside in the triangle Tabora Bismarckburg- Kilimatinde was known to be virtually foodless and water was scarce.
Despite careful preparations,the march began badly as over 5,000 locally conscripted
carriersdesertedat the onset. With 5,500 loads preparedfor movement,Wahle had only
800 carriers,200 British askari prisoners,350 chain-gangedAfrican civilian prisonersand
80 donkeysto move them. On 24 September,he decidedto destroythe excessand between
60-70,000kg of food, plus spareclothing, tobacco, alcohol and private belongingswere
burnt. To complicatemattersfurther, the ammunitioncolumn soon lost all 160 loadsas the
packdonkeysrapidly died from the effectsof the tsetsefly.
The loss of so much food and the needto forage widely forced Wahle to split his
troops acrossthree routes, each some distance apart. Abt von Langenn was given the
northernroute of Kiromo to Idodi, and was to be followed by At Zingel and Wahle's staff
oneday later. Abt Wintgenswas allocatedthe southernroute of Sikongeto Madibira, while
Abt Hfibener was to follow Wintgenstwo dayslater, but 110bener'spath was to diverge at
Mobogo and thence followed the more central route to Idodi. It was planned that von
Langennand Wahle would concentrateat Idodi by 18 October with Hiffiener arriving the
following day. Owing to the longer distancesinvolved, Wintgenswas not expectedto link
up until 22 October. Once together,the entire force would then move up to Iringa and
deliver a convergingattackfour dayslater.
189
Boell, Die Operationen,p. 287.

58

r--1'

-- zz-- -z-- -:;

!2

1.1-13

---

/
':

2--, W
,,,
I

te -,
.

,,

I:

"i

lu

Lij

C)

-0

, ---

4) -- .0A
zio

IL
ui
z
J

,-

11

C14.

z
-q!

C3

.a
3

WLN
-

0
0

0
.

00

cz

------

z,clli,
k -t
cc

E
-

cc

-alk
3.

tt,

.-

0
0.

?:5 11
C,4
.1

41

01-w

as

11

o",

_i%

o0

co
C-4

"---Z
.6

1%

.)0
-l

-,,

al

.-

-. -

---b

en

i
ii
.,
14 11

Cl)
Cs

415,1

C-\..

. -7

Oi2
,'fl-'

c-

to

-,;Zc

sel,

On 29 September, with the force largely concentrated in Kiromo, the lead elements
began their three-week march to the east. Unimpeded by any serious follow up despite the
presenceof the main Belgian force a few miles to the north, the Westtruppen disappeared
Fromview and nobody had any idea of their exact location or intentions. 190

None of this was known by Northey, although even a cursory examination of the
191
likely
Certainly
it
Iringa
a
objective.
was
map made
clear
and its small garrison
Rodger's column was no match for Wahle's force and could be easily cut off and
Northey continued to ask for assistance from 2nd Division.

Unaware of the true state ol'

that formation, Northey had to be content with the despatch of a mounted squadron to
Nyukwa's, which duly arrived on 27 September. In the circumstances, it was the best
that van Deventer could do, but it still left some 70 rugged miles of forest and mountain
between it and Iringa. In practical terms, it was too far away to be of any real use and tile
threat to Norforce remained undiminished.

Despite the growing dangers, Northey was unwilling to withdraw south. lie saw the
holding ofIringa as being important in terms of local prestige and the local population had
been very helpftil to him. It also offered a vital link to the main body, and for supply
purposes it was closer to the Central Railway than Lake Nyassa. Smuts, too, did not want to
lose Iringa, although he appearsnot to have fully appreciated the difficulties of Northey's
situation. Apart from the slender reinforcements sent to Nyukwa's, the only advice he gave
Northey in the event of the Gennans appearing in strength was "he should concentrate with
transport line via LUPEMBE". 192

Moreover, he appears to have ignored the consequencesof the dual threat posed by
Kraut and Wahle as Northey's mission of taking Mahenge at best speed remained
unchanged. The commander-in-chief was well aware of van Deventer's inability to send a
' 'Boell, Die Operationen,
p. 286-287,

"' WO 95/5292, War Diary GHQ, 3 October 1916,Appendix 4, 'relegram OA 569, Smuts CIGS, 3 October. It
to
was estimatedthat
the withdrawing Germanswere headedfor Mahengewith Wahle's 600 rifles moving along the railwa and then south and Wintgens
with 900 rifics going via Iringa.
1-12
WO 95/5292,War Diary GI IQ, 13 October 1916,Entry Nya.valand Force.

59

significant force south to secure Iringa in order to protect Norforce's rear. WishFUl thinking
ratherthan practical measuresseem to have been all that he could offer.

In summary, by the beginning of October, Northey had Rodger's column defending

Iringa, with Hawthorn and Murray dug-in at Mkapira vainly awaiting the advance of 2d
Division and with a half battalion in the Songea area. Apart from a single mounted
squadron,van Deventer was still unable to move south to link up with Norforce and showed
fiew signs of being able to do so soon. On the German side, von Lettow was pressing for
Kraut to attack both Songea in the south and Mkapira in the centre, while Wahle was just
beginning the long trek towards Iringa.

The first contact with the approaching Wesuruppen came on 12 October as one of
Rodgers's patrols was driven back across the Great Ruaha River at Kiganga by the vanguard
Furthen-nore, intelligence estimated that 6-7 companies were close behind with

company.

the intention of taking Iringa. 193 It was impossible to ignore the threat any longer and Sinuts
was compelled to send a large proportion of 2 nd Division's

usable strength to Northey.

One

battalion and a section of mountain guns marched from Kilossa on the Central Railway to
Iringa via Nyukwa's 194 while another was despatched on a lengthy and tortuous journey to
,
Nyasaland. Leaving Morogoro on 15 October, it moved by floot, rail, ship via South Africa,
and river steamer to Wiedhafen

before finally

going ashore and marching

forward

to

Songea. 195 It did not arrive until 27 November after an epic journey. 196

In the meantime, the marching battalion, now only 380 strong with six machine and
two mountain guns, was struggling southward. but could not reach Nyukwa's until 17
October. The remainder of Northey's force, the columns under Colonels I lawthorn and
Murray, lay well out of reach at Mkapira, over 100 miles away as the crow flies. They were
at least a week away from Iringa, and even if either had been ordered to cancel the march on

CAB 44/4, p. 11,

WO 95/5292,
as having
""

CAB

War Diary

about 400 rilles

44/4,

pp. 9-9,

61 IQ, 14 October
together

Sketch

with

2 guns.

1916, Appendix
The journey

17, Telegrarn
was estimated

F5 839, SMUtS to CIGS, 14 October.

7"' SAI 'A as gi% Co
at 6 days but actuall took closer to 9 days.

67.

' m'WO 95/5329, War Diar Norforce. Entries 14 October

mid 27-29 November 1916.

60

Mahenge, they could not have made the journey, as there were insufficient supplies or
197
carriersto sustain them.

By 19 October, Wahle's vanguard was skirmishing against Rodger's detachment.

As the degree of threat visibly increased, Northey ordered Rodger to concentrate around Alt
Iringa while Baxendale's Northern Rhodesian Police were split between that place and the
halt'
last
his
had
dump
He
Ngominyi.
a
available reserve,
already committed
vital supply
at
19'
There
lines
battalion of Africans, to Buhora near the southern of his
of communication.
The
for
do
but
brace
the
to
road network and supply
onslaught.
expected
was nothing
depots en-route were absolutely essential to Norforce's survival. South ofNgonunyi lay the
important depot of Malangall and thence Ubena.

The latter place occupied a critical

junction that linked Iringa in the north with Neu Langenburg in the south-west, while also
forward
lost,
be
Lupembe
in
Should
Ubena
the
then
the
columns would
all of
joining
east.
be isolated from their basesand the approach to Northern Rhodesia would be left wide open.

The attack on Norforce would take place, but not in the manner expected. The
patrol action of II October had slowed down the advance and it was not until 17 October
that the united columns of von Langerni and Zingel cleared the banks ofthe Rualla River
AN
British,
Suffering
the
the
as
and pushed on.
same problerns of command and control
Wintgens and I-Ifibenet-had been out of touch since the beginning of the march. This was to
have major consequences for the timetable as Wintgens, who had been entrusted with
escorting a howitzer, had tbLind the going impossible and had dumped the gun on I Wbener.
This in turn slowed down the latter's progress and would delay his arrival very considerably.
Of this Wahle would remain ignorant until battle was joined. 199

The whole of the Iringa garrison was dependent on the supplies held at Ngorninyi
and Malangali, both of which were thinly held and isolated. On 21 October, reinforcements
were sent Ngominyi, which had only a garrison of 50 together with a pair of old gulls.
Iringa itself was defended by only 100 soldiers who anxiously awaited the arrival of
CAB 44/4, p 12.
CAB 44/4, pp>9-11.
19 Boell, Die Operationen,
p. 289.

61

by
hastily
being
from
Malangali
a
the
entrenched
was
north.
reinforcements still en-route
200
from
Buhora.
detachment
brought
small reserve
up

Unfortunately for Northey. the reinforcements from Iringa were delayed for 24
hours in their departure.

This enabled the Gen-nans to close the road leading south,

ambushing first a supply column oil 22 October and then the reinforcements themselves the
next day, losing 37 out of 50. Successful in his efforts, Wintgens now moved north towards
Ngominyi to rejoin Wahle's main body. 201

Rodger arrived from Iringa on 24 October, but realised that he was completely
day,
help.
On
in
dug
tile
Unable
he
halted
the
to
to
same
await
outmatched.
attack,
and
further reduced reinforcing battalion from 2"d Division, now only 225 rifles strong, reached
Iringa, plus another 70 men from another unit.

On realising that the telegraph line to

Northey had been cut, the South Africans hurried south to assist Rodger. It was evident that
202
the enemy was astride the vulnerable lines of communications.

The tensesituation at Iringa was relieved by the unexpectedarrival of a mounted

infantry
had
launched
from
Dodoma
25
October.
It
the
a
arrived
newly
regiment
and
on
pre-emptive attack on At

into
Langenn,
the
its attack
of
moving
process
in
was
which
von

positions, and disrupted its plans. Another vigorous attack on the following day Further
hampered the German preparations.203 Although two of his Abledungen were in position to
attack Iringa, Wahle decided that the British were too strong and cancelled the attack
planned for 26 October. Furthermore, as lie was still without news of Abi Hiibener, who
was actually still far to the south, he decided to bypass Iringa and head south. A rearguard
204
left
inyi
Ngorn
to mask the move while the main body took the road south to
was
.

"" CAB 44/4, p, 15-16, Sketch 68 WO 95/5329, War Diary Norl'orce, I-ntries l9and 21 October 1916.
201CAB 44/4, 17. The RNR lost 3 KIA
'
22
PW,
p.
and
2111
WO 95/5292, War Diary G1 IQ, 26 October 1916, I: iitrv 2"" Division, However, Sinuts gave a rather difTerent account in his report
to CIGS the same day, Appendix 33,1'elegrarn F 59 10, Smuts to CIUS. lie said "While tile enemy is apparentIN in no position to do
any substantial damage to General NORTHEY it is also clear that the eliciriv cannot be prevented from dribbling through in small
parties along innumerable tbotpaths in the bush. "
..
2111
WO 95/5292, War Diary GHQ, 27 October 1916,
Appendix 35, Telegrarn F 5917, Smuts to CIOS, 27 October
2114
Buell, Die Operationen. pp 290-29 1,

62

A key supply depot, Ngominyi had been poorly laid out for defence and had been
under siege since 22 October.

Wahle ordered the newly arrived AN Wintgens to take

Ngominyi, with the attack going in on 28 October.

Despite a determined fight, the

defenders were greatly outnumbered and cut off, with the post falling to a dawn assault tile
next morning. It was a major blow to Northey as a large quantity of much-needed supplies
was captured and the situation in fringa still appeared serious.

The vulnerable lines of communication

cut in half

were now severed and Norforce

had been

The loss of further depots would seriously delay fature offensive operations as

well as threatening the survival of the troops at Iringa.

Furthermore, the week's events and

prisoner interrogations had revealed that the Wesitruppen were stronger and more aggressive
than previously thought.

Given the withdrawal

from Iringa and the loss of Ngorninyi,

it

2115

seemedcertain that the base at Malangall would be their next objective .

Reinforcements from Iringa had joined Rodgers on 29 October, too late to help at
Ngominyi, but with a combined force of only 305 rifles it was greatly Outnumbered by the
enemy. AbI Winigens, having taken Ngominyi, now turned to deal with this force, attacking
heavily over the next three days. Faced with a shortage of food and concerned about the
situation at Iringa, led to the battered column's withdrawal on the night of 1/2 November,
206
from
For his part, Wahle continued to move south as an
the north.
ending all pressure
expanse of barren mountains prevented an immediate move to the east. As predicted,
Malangali was his next objective.

Belatedly, Smuts realised the true threat posed by Wahle. Norforce was in danger of
being overwhelmed, its supplies were threatened and 2nd Division was in no position to
provide further assistance. Poor communications and distance meant that the fight was out
of his hands while Northey was out of touch with a number of his own troops. Instead of
the early capture of Mahenge, he faced the prospect of a major defeat in the weSt.207

2111
CAB 44/4, pp, 17-20 WO 95/5329, War Diary Norforce, Entries 2
and 3-4 November 1916.
206CA13 44/4,
pp 19-20, The 7"' SAI suffered three dead, sixteen wounded and six missing.
2111
WO 95/5292. War Diary GHQ, I November 1916, Appendix 1, Telegrain F 5948, Smuts
to CIGS, I November.

63

Events elsewhere were also taking a turn for the worse. On 21 October, further
Murray's
itself
Hawthorn's
Kraut's
combined
and
opposite
south,
advanced guard revealed
driving
in
British,
he
day
Ruhuqje.
The
the
to
outposts
the
encircle
started
next
position on
and seizing and fortifying a feature to the west of their position.

The envelopment was

completed on 2') October and the track leading to Lupembe was cut.
Kraut had a sizeable force, equivalent to the defenders', and was under firrn orders to attack
208
British
However, as the enemy was cut off and isolated from the rest of Norforce,
the
.
Kraut considered that there was no immediate need to attack as he intended to starve tile
British into submission. He trusted to the strength of his encirclement ofthe position and to
his belief that the swamps to the north of their defences were impassable.209 What he did
not know was that Hawthorn had stockpiled a large quantity of supplies and was in wireless
communication with General Northey at Neu Langenburg.

After a week of isolation, with food running short and hearing of the WesurIII)I)e1l's
approach to Ngominyi, Hawthorn decided to break the Gen-nan siege. He realised that it
would be impossible to escape with his baggage column and precious wireless set without
driving the enemy off completely.

Laborious reconnaissances had I-ound several routes

through the swamps and a plan was devised.

Surprise was of the essence, and lie

determined that the Germans would be unprepared for a simultaneous attack from north and
south.

The night of 28/29 October was chosen for the break-out; the Rhodesians were to
strike the enemy to the SOLIth-east
while Hawthorn's troops attacked to the north-west. They
struck with considerable violence at dawn on 29 October, having infiltrated through tile
swamps to within 400 yards of the Germans, breaking-in and capturing a gull. The centre of
the German position was reached by 0830 hours followed shortly by the precipitate retreat
210
besiegers
the
bank
Ruhudje.
The action was a triumph for
to the eastern
of
of the
Hawthorn and, apart frorn inflicting some sharp casualties at relatively low cost, at 21
British to I 18 Germans, it freed up Norforce's only sizeable reserve. The victory and
2"' Boell, Die Operationen,
pp. 256-257.
MSS Afr. s. 1715 (300), Ilistory off KAR, pp. 160-161.
WO 95/5329, War Diary Norforce, Entry 30 October 1916; Boell, Die Operationen,
p. 257.

64

his
Northey
to
concentrate
Kraut's subsequentwithdrawal to the north-west now allowed
211
forcesaroundLupembein orderto dealwith the growing threatthere.
nd
Division
2
Deventer's
balance,
its
van
While Norforce was struggling to regain
headquarters
his
to
Having
in
fmally
shifted
able to move southward some strength.
was
Dodoma,van Deventertook commandof the Iringa areaon 30 Octoberand was orderedto
P
212
Brigade
Infantry
SA
The
it
was a shell
much reduced
reinforce asquickly aspossible.
75
Nearly
half
from
per
its
former
having
the
of
rations.
periods
prolonged
suffered
self,
of
battalions
four
down
the
had
were amalgamated
troops
with sicknessand
gone
cent of the
into a singleunit.213The mountedtroopswere little better,as I" SA Mounted Brigadecould
horses
by
October;
horses
in
245
and equipment
705
only pooling
early
men and
musterjust
nd
2
the
the
remnantsof
could a singlemountedregimentbe organised.Joining that unit with
SA Mounted Brigade, a total strength of about 900 rifles could be raised. This was to
diminish evenmore drasticallyasthey marchedthrough the tsetsefly-infested bushtowards
Iringa, losing over 500 horses.214In the meantime,a column under Colonel Taylor prepared
to follow up shortly from Dodoma. The much-reducedinfantry Brigadeheld Nyukwa's and
215
Kilosa.
the groundto the eastat Kidodi, therebysecuringthe routesto
Now that he had handedover Iringa and Kraut had been defeatednear Mkapira,
Northey was able to concentratesignificant forces. Although he had lost telegraphic
link
his
to
he
had
columns.
the
a
wireless
communications,
great advantageof possessing
This enabledhim to instruct Murray to march for Lupembeas soon as he had heardof the
Mynera
Column
Murray's
the
Kraut.
Leaving
2
November,
marched
up
victory over
on
Valley and thenceup the escarpmentto Lupembearriving a few days later. One company
holding
Malangali
immediately
100
forward
the
while the rest
to
recruits
sent
was
reinforce
216
defend
Lupembe.
of the battalion was sent forward on 8 November by motor cars to
Apart from being a major roadjunction, with one route leading north to Iringa.and another
211WO 95/5329,War Diary Norforce, Entry 2 November 1916.

212WO 95/5292, War Diary GHQ, I November

1916, Entry General Headquarters.
211CAB 44/4,
pp. 39-40.
214CAB 44/4,
K 651,2 Div to BGGS.
pp. 41-42; GSWA, Box 27,12 October 1916, Telegram
799 effectives while V SA Mounted Brigade had 604. Another
1,500 troops from P Division
2'5 WO 95/5292, War Diary GHQ, 7 November
9, Telegram OA 969, Genstaff
1916, Appendix
216CAB 44/4,
1916.
p. 27; WO 95/5329, War Diary Norforce, Entries 2,7 and 8 November

65

Yd SA Infantry

Brigade

could

muster

as reinforcements.
were proceeding
to CIGS, 7 November.

for
It
base.
the
key
road-head
was
supply
going easttowards Mkapira, Lupembe was a
functioned
Nyassa
from
Lake
by
and
being
delivered
north motor vehicles coming
supplies
217
for
It
the
Mahenge.
for
was
vital
base
the
against
the
operations
conductof
as
advanced
future
operations.
supportof currentand
The Germans surroundedthe garrison at Malangali on 8 November, cutting the
by
increased
day,
The
their
it
numbers
to
the
next
telegraphwires connecting
outsideworld.
Unable
bayonet
to
launched
three
total
At
Langenn,
attacks.
they
of
the arrival of
a
von
break into the camp, their artillery inflicted a great deal of damageand destroyedmuch of
but
held,
defenders
the situation
Considerably
the
food
the precious
outnumbered,
supplies.
wasvery precarious.
However, prior to the arrival of the Westtruppen, Northey had placed considerable
in
This
the crisis, as
into
improvement
off
tracks
paid
motor roads.
emphasis on the
of
Northey decided to risk 50 of his precious light cars in a daring relief move. Late on 8
November, he ordered Murray to take 130 men and 4 machine guns by vehicle to Malangali
via Ubena. The remaining 125 men and 6 machine guns marched off cross country with the
At
Wintgens
that
linking
At
time,
the
word
there
receiving
same
subsequently.
up
aim of
be
Hawthorn
to
Northey
headed
Lupembe,
told
ready to march
was
south-west towards
back there.218

Murray and his troops arrived to within two miles of Malangali unmolested,having
travelled 120 miles over some of the roughestroads possible. Debussingsafely, Murray
10
A
back
for
the
Buhora
the
to
of
on
morning
reconnaissance
sent
vehicles
supplies.
November showedthat the Germanswere too strong for him to attack and he decidedto
keep his position and maintain contact. Wahle's efforts against Malangali had cost him
heavily and the arrival of the Rhodesianscoincided with news of Kraut's defeat finther
219
He decidedto lift the siegeand move off south-easttowardsthe Lupemberoad to
south.
217MSS Mr. s.1715(300), History I KAR, pp. 161-162.
of
212WO 95/5329,War Diary Norforce, Entries 8 and 9 November 1916;CAB 44/4, pp. 29-30.
219WO 95/5292,War Diary GHQ, 17 November 1916,Appendix 23, TelegramNF 1498,Norforce to Genstaff, 16 November.
Norforce sufferedone killed and five wounded while reporting that the Germanslost eleven dead,26 wounded plus 40 casesof
ammunitionand stores.

66

join Wintgens. A two-company rearguard was detailed to maintain the blockade of

Malangali and to link up with HUbener'sdetachment,still not yet arrived from the west. As
the Wesuruppenslipped away on the night of 11/12November,Murray took the advantage
220
force,
it
to attackand overwhelmUs
causing seriouscasualties.
While theseeventswere taking place, Wintgens had reachedLupembeand had put
in an attack on 13 November, sending a detachmentto cut the road to Ubena. This
effectively cut off the garrisonof 250 KAR recruits,50 SouthAfricans, 4 machinegunsand
severalantiquatedfield pieces. It was well stockedwith food and ammunition, but it also
had the drawbackof having some300 wives and children of the KAR recruits,whosedepot
it was.221 An outwork was taken,the wirelessmastsdestroyed,but the garrisongrimly held
on. Heavy firing was kept up throughout the day and another determined,but ultimately
unsuccessfulattack,was put in at 0400hours on 14 November.Following a ceasefireto deal
222
further
Wintgens
day.
with the wounded,no
withdrew the next
attackswere pressedand
In the meantime,with Hawthorn still somedistanceto the east,Northey decidedto
redeployMurray by vehicle. His column drove back to Ubenaon 14 Novemberfrom where
foot patrols were sent out. They located the enemy roadblock about 30 miles west of
Lupembeand the column moved forward for action, spendingthe betterpart of two days in
reconnaissance. Striking on the evening of 16 November, he overwhelmed the two23
block,
Lupembe
company
clearingthe road to
The beleagueredgarrison was relieved by the arrival of Hawthorn's column on IS
November,sometimeafter Wintgens's departure,and following a slow and difficult j ourney
from Mkapira. This was timely for Wahle's main body appearedfrom the north-eastthe
next day. Knowing that Kraut had beenmoving westward,the British fearedthat he would

220CAB 44/4,pp. 28-30;WO 95/5329, War Diary Norforce, Entries II

and 12 November 1916. The British losseswere four wounded
and the Germanssuffered elevendeadand twenty-six taken prisoner as well as losing one machine gun, 72 carriers, 15 mules and
donkeysand 50 cattle; Boell, Die Operationen,pp. 293-294.
221CAB 44/4, pp. 31-32; MSS Afr. 1715,History I KAP, 163.
s.
of
p.
m WO 95/5329,War Diary Norforce, Entries 13,14
and 18 November 1916.
223WO 95/5329,War Diary Norforce, Entry 17 November 1916. The British lost
one killed and four wounded to the Germans' 13
deadand seventaken prisoner.

67

be
Ubena
to
Now
Lupembe.
that
for
safe,
Wahle
appeared
an attack on
combinewith
NortheyorderedMurray to move forward and rejoin Hawthorn at Lupembe.
The link-up duly took place on 18 November,with the two columns united for the
first time sincethe end of October. Togetherthey musteredan effective marching force of
224
blow.
about 1100rifles and had shown themselveswell able to give the enemy a sharp
However, Wahle had no intention of lingering near Lupernbe, as, after the failure of
Wintgens'ssiege,he realisedthat the British would reinforce it strongly and he decidedto
bypassit altogether. His main body moved unopposedto the north of Lupembefrom 17 to
19 November,moving via the Mynera Valley for the Ulanga lowlands. The march of the
Wesuruppenwas finally completedon 22 November,when Wahle met up with Kraut and
225
Tanganika.
headquarters
front
forces
Us
at
with
assumedcontrol of all
on
Freedfrom the main command,Kraut ultimately advancedup the easternedgeof the
Lupembe,
Mfrika,
and standingat an altitude
the
of
east
six
miles
seizing
of
plateau,
village
he
level.
Too
feet
to
6,000
feet,
1500
Mahenge
weak attack,
abovesea
of
with the
plains at
built
being
forward
somesix miles west at
there,
with a second
constructeda strong
position
Msalala. These fortifications, together with the thick bush and broken ground, made a
British attack a difficult proposition."'

Furthermore,both sides were exhaustedafter the

had
few
the
sufficient suppliesor carriersto
arduousmarching of
past
months and neither
light
in
An
further
the
of
coming
rains.
operational
sustain
offensiveoperations,particularly
pauseensuedandreorganisationwas setin train.
Almost unnoticed under the strain of the battle was the loss of Northey's
independenceas a field force commanderon 14 November. GeneralSmutshad requested
that Norforce be placedunderhis operationalcommand(althoughwithout responsibilityfor
fighting
in
7
November,
the
the Iringa-Mahenge
to
the
supply) on
citing
need co-ordinate
227
area. The changemadegood senseas it finally united all the British forcesin EastAffica
224WO 95/5329,War Diary Norforce, Entry 18 November 1916.
225Boell,Die Operationen,p. 294; CAB 44/4, 48.
p.
n6 WO 95/5329,War Diary Norforce, Entries 27-29 November 1916; MSS Mr. s.1715,History of I K44R,p. 163.
227WO 95/5292,War Diary GHQ, 7 November 1916,Appendix 9, TelegramOA 969, Genstaff, Morogoro to CIGS, London 7
Novemberand WO 95/5292, War Diary GHQ, 14 November 1916,Appendix 18, Telegram OA 24, Genstaff, Morogoro to Norforce
and Genstaff,Dar-es-Salaam,13 November.

68

Office,
having
Colonial
it
the
the
absurdityof
undera single commanderand also removed
War
Office
the
the
and
campaign
of
with no military staff or resources,controlling one part
the other. The move was welcomedby Northey on military grounds,althoughit seemsthat
he hadno high opinion of Smuts'stactical abilities.
While these actions were taking place in the north, von Grawert had been prodded
by von Lettow to deal with the enemy around Songea on 23 October. Advancing on 31
October, he sent orders to Falkenstein to advance on Songea and join him in the attack on 5
November. However, owing to delays and confusion he did not join up with von Grawert
force
British
Falkenstein
11
November.
Finally,
the
time
early on
on
moved
with
short,
until
12 November and without von Grawert's troops, part of whom had been left behind as
African
hostility
detachments,
the
the
population to
of
security
growing
a consequence of
228
Germans.
Despite a promising start, Falkenstein was killed leading the attack and
the
broke
Germans
further
began
Unable
the
to
to
off the
progress,
ammunition
make
run short.
battle, retiring to a position about three kilometres northeast of the boma. By 17 November
229
TES
forward
threatened
the
to
position.
more reinforcements were marching
reinforce
force was too strong for von Grawert who then prudently retired to the village of
Njamabengo.230

While theseeventswere underway,the almost forgotten rearguardunder Lieutenant

Colonel HUbenerremainedat large. With some300 rifles and a howitzer, it had madevery
slow eastwardprogress. It was located at Ilembule on 20 November, and, with Lupembe
secureand Wahle safely to the cast, Northey decided to eliminate this last threat to his
colunins. Again, using his superiorwireless communicationsand mobility, he sentMurray
with 450 men to Ubena by vehicle, whence it marchedoff to deal with HUbenerthe next
day. The German officer was not a great tactician nor overly energeticand allowed the
British to surroundhis position and securethe vital waterholesby 26 November. Despite
heavy firing by the howitzer, British casualtieswere slight and the lack of water made
22'Boell, Die Operationen, 255. At
p.
von Grawert and Falkenstein had about 500 men in total. During the battle only 18 Germans
and 271 askaris with 2 machineguns were deployedagainst26 British and 223 askaris. The German losseswere six killed, 13
woundedand three missing.
229WO 95/5292,War Diary GHQ 19 November 1916,Entry Nyasaland Force. Half
of 5 SAI had arrived and were immediately
pushedon to Songea;WO 95/5329, War Diary Norforce, Entry 17 November 1916.
' Boell, Die Operationen, 256.
p.

69

further defencehopeless. He duly surrendered,with sevenofficers, 48 Germansand 249

"
3
howitzer
10.5
into
Askarisgoing
and machineguns.
cm.
captivity togetherwith one
Therewere now no Germanforceswest of the line Neu Langenburg-Iringa.Despite
the local superiority of numbers,Wahle had failed to inflict any significant damageon
Norforce and had himself lost over 500 soldiers although he had successfullybrought his
force across 300 miles of desolateterritory. Both Murray and Hawthorn had shown
themselves to be determined and effective leaders who had given Wahle's troops some
nd
had
2
Division
been
have
handling.
Indeed,
exerted any
achieved
much more could
rough
General
The
to
but
from
threat
the
the
missed.
was
opportunity
north,
sort of pressure
Northey's force had passed. Indeed, a relieved General Smuts reported to London that some
71 German and 370 Askaris had been killed or captured during the withdrawal of the
Westtruppen.232

OPERATIONS NORTH OF THE LOWER RUFIJI

The captureof Dar-es-Salaarnon 4 Septemberled to a break in contactin the coastal
learned
British
It
September
the
that
of the
the
command
region.
was not until
end of
Kissangire.
Rufiji
lower
in
centre
of
the
administrative
presenceof an enemy company
Smuts decided to delegate the clearance of the area to his Inspector General of
Communications(IGC), a rather unusual choice given the severeadministrativeproblems
that the force was undergoing.As the IGC had neither an operationalheadquartersnor much
in the way of fighting troops, an attempt to clear the area on 4 October failed miserably
233
losses.
five
heavy
days
later
he
some
and returnedwith
This setbackspurred Smuts to send reinforcementswhich were in position by 21
October.234But, they arrived too late to be of use asAbt Stemmermannhad reinforcedthe
defenceswith four companies. While Smuts was concernedabout the possibility of raids
231WO 95/5292,War Diary GHQ, 29 November 1916,Entry NyasalandForce.
232WO 95/5292,War Diary GHQ, 25 November 1916,Appendix 35, TelegramX. 945, Smutsto CIGS, 25 November.
233CAB 44/5, pp. 4-6. The British lost 13 killed, 27 wounded and 8 missing while the German losseswere cited as being slight See
von Lettow,Reminiscences,
p. 160. He statesthat the defenderswere only the inexperiencedcompanyof the Kftgsberg crew.
234CAB 44/5, pp. 6-7.

70

for
Germans
there
have
the
he
does
that
a very
to
were
realised
not seem
againstthe railway,
differentpurpose."' They urgently requiredfood, and the lower Ruf1ji region providedrich
it
the
the
The
to
there
ripened,
crop
until
protect
was
garrison
and plentiful maize crops.
had
Lettow
its
harvest.
Von
1917,
date
March
to
for
then
oversee
and
earliest
which was
it
his
However,
instructions
task.
the
Sternmermann
value of
there with explicit
on
sent
left
his
the situation
failed
Smuts
to
and
aims
that
real
opponent's
appreciated
appears
unresolvedfor anothertwo months.
KIBATA

BATTLE
AND
THE
PRELUDE
-

As operationson the Mgeta wound down, von Lettow reconsideredhis options.

Apart from remainingon the defensive,he had the option of attackingthe now static enemy
forcesto his north. However,the groundwas difficult and the British were now well dug in.
Any attack was likely to result in heavy casualties and high levels of ammunition
in
hand,
Northey's
On
be
the
advance
other
expenditure,neither of which could afforded.
the westwas being dealt with by Kraut and the troops north of the Rufiji were protectingthe
him
Kilwa
the greatest
harvest.
lodgement
that
It
caused
the
at
ripening
was
growing
concem.
As soonas he had learnedof the fall of Kibata on 14 October,von Lettow resolved
to recoverthe situation. He orderedthe bulk of his troops to move there on the following
day, with At Schulzbeing despatchedon 17 October. In all, he took eight companiesplus
the bulk of his artillery south, leaving nine companiesand a gun to hold the line of the
Mgeta under the commandof Captain Otto. Von Lettow himself left the Mgeta front two
days later before moving to join his main body of troops opposite Kibata.236 However,
owing to the distanceand state of the tracks, this reinforcementwould take about two to
threeweeks. This delay, unavoidableas it was, enabledthe British to continuethe build-up
of their troopsunimpeded.
235WO 95/5292,War Diary GHQ, I November 1916,Appendix 1. TelegramF 5948, Smutsto CIGS, I November. He stated"Enemy
holds KISSENGIRE and MKAMBA on roads from RUFUI to DAR ES SALAAM and is patrolling very actively further North
apparentlyin expectationof attack from DAR ES SALAAM or as a threat to our railway communications."
216Boell, Die Operationen,p. 242; von Lettow, Reminiscences,
pp. 161-163; He statesthat he left Tafel in command,but Boell and
the courseof subsequentoperationsmakesit clear that Otto remainedin chargeof the Mgeta line.

71

-..

-----.

-------/

U
?eI-:p
.-

iz

I:..:.

I%.

-46

2p
.

It

ej

tI-,

IQ
--=i

C-,

CD

-.a

-All

-, P

-wI

21\

1w.

ca

cill

10-,

I=I

CN

o
_,

4 RLI
.1

C)
z

tL
ir
IIw

z
ci
Im-i

310,

its
to
threat
advanced
the
increasingly
growing
The Kilwa force was
aware of
A
in
the
Germans
series
region.
to
supplies
took
reduce
positionsand
a numberof measures
inflicted
lines
German
some
and
fighting
the
communication
of
of
patrols went out against
October
fell
29
Liwale
and
Mpotora,
the
on
importantsetbacks;the magazineat
road,
on
The
November.
these
2
destroyed
was
raids
Chini
of
Ngararnbi
outcome
on
was
anotherat
237
ffirther
north.
to forcethe Germansaway from the coastand
nd
2
the
Hannyngton
took some time as
In the meantime, reinforcement of
Brigade was in the process of reconstitution under Brigadier General O'Grady.
Reinforced by two newly raised KAR battalions and two more experiencedunits, the
brigadearrived in Kilwa over the last two weeksof November. The divisional commander,
Hoskins,realisedthat it would be very difficult to bottle up the Germansin such wild and
divided
he
forward.
Accordingly,
forces
he
neededstrong
mountainouscountry and that
his areaof operationsinto two; in the north, Hannyngton's 3rdEA Brigade held Kibata and
forward positions in the Mtumbei Hills. To the south and west, he deployed O'Grady's
force in the MatanduValley. A small reserve,consistingchiefly of the immobile artillery,
238
held
Kilwa.
at
was
British intelligence assessed,fairly accurately, that the Germans had about 10
by
in
the
increase
Kibata.
This
capture
confirmed
was
towards
strength
companiesmoving
239
first
Kibata
The
in
November.
documents
interrogation
on
probe
the
of prisoners
of
and
led
by
Captain
launched
7
November,
two
companyreconnaissance-in-force
was
on
with a
Schulz. It was repulsedby the British by 9 November and servedto increasethe volume of
their defensivemeasures.Despitethe rebuff, Schulz informed the Kommandeurthat, with
240
later
forecast
His
losseS.
heavy
Kibata
be
was
of
casualties
taken
artillery,
could
without
to prove sadlyinaccurate.

237Boell, Die Operationen,p. 246.

238CAB 44/5, p. 2 1.
239WO 95/5292,War Diary GHQ, 7 November1916,Appendix9. TelegramOA 969, GeneralStaff Morogoro to GeneralStaf&Dar-esSalaam,7 November.
240Boell, Die Operationen,p. 246; Moyse-Bartlet TheKing's African Rifles, p. 338.

72

defensive
ideal
from
far
position
Kibata,
an
The position at
although strong, was
for
hilltop,
high
fort
itself,
was obsolete modem warfare.
walls on a prominent
andthe
with
It was almost completely surroundedby a ring of small hills about one thousandyards
important
The
was
most
distantwith prominent river valleys running to the west and east.
PicquetHill, a long and narrow ridge to the north-westof the fort. Originally coveredwith
fire.
It
fields
improve
the
defenders
by
was
to
bush,
it
had
been
of
the
thom
cleared
it
had
its
importance,
two
in
fort
major
higher
than
the
of
and,
recognition
actually
dug-outs
lacked
No
2
Redoubts
they
known
I
sufficient
No
and
although
and
as
positions,
depth241 These featureswere in turn dominated from the north by a large ridge running
.
242
Kibata
the
high
loss
The
of
that
possession
make
to
ground would
of any of
west east.
fort difficult, but the retention of Picquet Hill was vital as it commandedthe roadsleading
243
in andout of Kibata aswell astwo of the three water sources.
In the first week of December,the British garrison consistedof about a battalion
the
two
by
held
hills
fort
lesser
each
of
as
were
company,
the
one
were
and a seriesof
with
36
the
battalion
Picquet
Hill.
The
with
away,
miles
some
was
nearestreserve
redoubtson
6
December
that
the
It
brigade
the
further
of
afternoon
the
to
the
on
southwest. was
restof
defendersfirst sighted columns advancing from the north, with the outposts driven in
line
first
Having
the
in
battle
of
secured
shortly thereafter;the
now commenced earnest.
defences,von Lettow used the night to bring his artillery and machine guns into position.
Using the samehuge gangs of African labourerswho had draggedthe guns all the way
from the Rufiji, he deployed one of the 4.1" K6nigsberg guns, a field howitzer and two
244
feature.
mountaingunsonto a commanding
With excellent gun positions ready, von Lettow then pushed two Atedungen
forward while a company was ordered to infiltrate around the defendersto cut the track
leadingback to Kilwa. The Germansnow numberednine companiesagainstthe defenders'

241Thatcher,W S, 7he Fourth Battalion,Duke of Connaught's Own 10' Baluch Regimentin the Great War, Cambridge:Cambridge
University Press,pp. 144-145. This history coversthe 129'bBaluchis.
242CAB 44/5, p. 22.

243WO 106/273,History of3 KAR, Plan of Kibata, p. 58a; WO 95/5330, History of 1`12"dKM
244CAB 44/5, pp. 23-27; Moyse-Bartlett, TheKing's African Rifles, p. 339.

73

pp. 5-6.

CL

ba
cis

5/ *.
.

CL

qu

CID

-jjas

Ct

C14

LO

t
4) rz

1= ---, ,,,\"I

cc

cir

i000-10z

w0
tL

wl
x

!!
Z

w=
;
.?:

CD

x-,,

7E

:2
0

iz:

>
0

's

ol

vI

E
o

-C
00

-Pf

lz

.0u

z
w

<

'fo

cc
E

1 .14 .-

-3ok-

245
Significantly,
their
defences.
artillery group
the
to
main
six and were ready move on
outrangedthe defenders'mountain gunsby a wide margin.
The next morning, the battle was began with a severe artillery bombardmentof
Picquet Hill. No 2 Redoubt suffered the worst and was nearly destroyed during the
byAbt
Schulz
dusk
A
hold
to
gaineda
assault
the
on.
morning although
garrisonmanaged
lodgementabout 80 yards from No 2 Redoubt. This now seriously threatenedthe British
246
With the
hold on Picquet Hill and prevented the use of one of the water sources.
defendersunableto expel the Germans,the night of 7/8 Decemberpassedwith both sides
in closeproximity to eachother. The British had sufferedheavily from the accurateenemy
artillery fire and poor camouflage.
On receiving news of the onslaught,Hannyngton committed his reservebattalion
in
34
36
forced
battle.
A
two
miles
march of some
and
gruelling
mountain guns to the
hours over razor-backridges separatedby deep gullies conductedin pouring rain brought
the relieversinto Kibata at 0200 hours on 9 December.On the sameday, GeneralHoskins
hard
it
but
nd
battalion
from
Brigade,
2
EA
too
some
would require
also releasedanother

defenders.
marchingto reachthehard-pressed
Von Lettow resumedthe attack in the morning of 8 December. Supportedby
artillery, the troops in the lodgementattemptedto rush the redoubtswhich came close to
isolate
day
the
failed,
Germans
The
having
the
to
took
the
the
of
rest
succeeding.
attack
hill from reinforcement. The heavy rain which markedthe day obscuredobservation,with
247
More shelling
Kbnigsberg
between
deadly
fire
the
positions.
gun alternating its
following the next day, but the Germanswere unable to launch a further infantry attack
failed
lodgement
In
the
British
to
the
turn,
the
and a
owing
an attack on
arrival of
reserves.
deadly stalemateensued. Hoskins was now seriously concerned about the situation,
P
EA Brigade the
sending more mountain guns on 9 December and remainder of
245Lettow,Reminiscences, 168-169.
pp.
246Boell, Die Operationen, 247. Von Lettow had
p.
about900 rifles and 15 machine gunsversusHannyngton's 930 rifles and 7
machineguns. However, the 4.1" Kbnigsberg gun, the howitzer and mountain guns outrangedthe two British mountain guns.
Thatcher,IOh Baluch Reginwnt,p. 146;
247CAB 44/5, pp. 25-26; WO 95/5330, History I"12"dKAR, 6.
p.
of

74

following day. Finally, on II Decemberhe placed General O'Grady in commandof the

battle.

In the meantime, von Lettow had lost his initial superiority of numbers as the
British rapidly built up their strength. After three days of relative quiet, they made an
attemptto drive the Germansback. The plan was simple; to seizea rocky and bare feature,
later known as Gold CoastHill, some3,000 yardsto the west of the Germanright flank and
separatedfrom Kibata by a deepvalley nearly. This feature provided a coveredapproach
from which a flanking attackcould be launchedonto the Germanpositions.
The two new battalions began their flanking move on 15 December, but quickly
hit
by
Hill,
Gold
Coast
fire
from
Now
they
were
came under
secure on
an enemy outpost.
heavy artillery fire that was closely followed by a strong infantry attack. A series of heavy
attacks ensued and the defenders' situation became desperate. With heavy casualties,
248
further
became

extricationratherthan

advance

essential.

While this maelstromwas raging to the west, the newly-arrived General O'Grady
was preparingto restorethe situation at Kibata. With him cametwo more mountain guns,
ammunition, including Mills bombs (heretoforeunknown in East Africa), and food. He
plannedto eliminate the lodgementon Piquet Hill as a pre-requisitefor the turning of the
Germanposition. A battalion was given the task of a night attack using the new Mills
bombs for shock action while the artillery spent most of 15 December registering the
German positions.249 After dark, a silent night attack drove the Germans out of their
trenches,and regainedPicquetHill. At small cost, they had won an important victory and
eliminatedthe most dangeroussalient. Ironically, it was von Lettow's earlier decision to
use his reserves for the attack onto Gold Coast Hill that left the main sector so
250
vulnerable. With it went his chancesfor victory at Kibata.

Clifford, Sir H C, Ae Gold CoastRegimentin theEastAfrica Campaign,pp. 51-55. The Regimentsuffered146casualtiesout of the 340
s
SeL
`reC 44/5, pp. 33-34; WO 9515330,History 1'012*dKAR,
p-7of
250Thatcher,I OkBaluch Regiment,
pp. 154-155. The Baluchis lost 15 killed or wounded and took 13 Germanprisoners;CAB 44/5, p.
35.

75

Neither side now had sufficient infantry to renew the offensive and activity was
Both
to
intermittent
regroup and
needed
sides
to
shelling.
reduced patrol encountersand
the situationremainedlargely unchangeduntil the end of the year.
The battle had drawn in the bulk of lst Division's fighting troops, while a large
inconclusive,
battle
The
tactically
force
Lettows
was
portion of von
was also engaged.
had
Von
Lettow
for
limited
both
heavy
certainly
gains.
casualties
with
sides suffering
blockeda British advancetowards Utete, but he had failed to gain the decisive victory that
he sought and had expendedlarge quantities of ammunition. In the circumstances,and
given his initial superiority, it is puzzling that he did not try cut off the position entirely.
The defenderswere on very short rations for over six weeks and their water supply was
very vulnerable. He appearsto have gone for set-pieceassaultsrather than more subtle
both
the
that
battle
to
However,
newly
sides
the
also
revealed
encirclementand starvation.
inexperienced
to
the
too
KAR
battalions,
full
meet
still
were
raised
although
of potential,
Schutztruppeheadon in battle.
THE FINAL PUSH TO THE RUFIJI
As the rains continued and he struggledto reorganisehis force, Smuts now had to
deal with seriousdissentboth in South Africa and in his own troops. In late October, one
of his own battalion commanding officers lodged a formal complaint, making strong
allegations of incompetence,indifference to soldiers' welfare and general negligence.
Matters were not helped by the harrowing stories beginning to emerge in the South
African press,notably the Rand Daily Mail and the SundayTimes. In mid-November, the
latter paper made accusationsof serious shortcomings and lack of planning (these are
covered in Chapter Six). A court of enquiry was establishedto deal with the military
into
forced
Minister
Prime
General
both
Governor
were
allegations, while
the
and
251
damage
limitation.
political

251Brown, King
and Kaiser, pp. 301-302; W033/858, TelegramsD 1, No. 1388,17 November 1916,TelegramOA84, Smutsto
CIGS.

76

Doubtlessthesemattersaddedto the urgency of ending the campaign. Smuts,too,

On
German
taking
his
surrender.
the
a
of
chances
retained
perennial optimism about
Norforceunder formal commandon 14 November,he laid out his plans to Northey at some
length. Viewing the Germans as operating in two discrete groups under the command of
keep
he
in
to
Lettow
in
them apart and
KrautfWahle
the
the
wanted
west,
von
east and
defeat each in turn.252

Thesethoughtswere formalisedon 30 November,when GeneralSmutsreportedhis

intentionsto the CIGS. He beganby estimatingthat von Lettow had about 4,000 rifles on
the Rufiji, while Kraut's 2,000 men had been joined by a maximum of 500 or 600 of
Wahle's force. With the Schuatruppehaving a maximum of 7,000 effectives,he proposed
to addthe newly arriving Nigerian Brigade:

"while on arrivalof NIGERLANSat front I shallendeavour

to cut off retreatof VON
LETTOW'S force and obtain decision on LOWER RUFIJI. Should enemyescape
Southwardfrom RUFIJI it will be reducedin numbersand morale...

253
,,

To van Deventer,he was blunter, giving his aim as follows:

"to drive the enemy everywhereover the Ulanga (Kilombero) before the end of
January,then to leave the Indian and African battalionsto hold the line of the river
, 254

during the rainy season,andto move the white troopsback to the railway'
.

He believedthat the combinedforcesof van DeventerandNorthey would be able to

deal with Kraut and Wahle separately. He was optimistic enoughto believe that Belgian
assistancewould probablynot be requiredastheir preparationswould take too long and they
would be a burden on his supply system. Finally, he consideredthat the expandedKAR
would be capableof dealingwith any forcesbeyondFebruary1917.255

212WO 95/5292,War Diary GHQ, 14 November 1916,Appendix IS. TelegramOA 24, Genstaff to Norforce, 13 November.
211WO 95/5292,War Diary GHQ, 30 November 1916,Appendix 42, TelegramOA 198, Smutsto CIGS, 30 November.
11 CAB 44n,
pp 1-2.

77

On 22 December,GeneralSmutsconsideredthat preparationsfor an advanceon all

frontswerecompleteand he andhis advancedGHQ left Morogoro on the sameday, moving
to Dutumi. While it was clear that the British situation was difficult, it did not seem
impossible;in the east 1" Division had stabilised the dangeroussituation at Kibata and
had
Hoskins's
Lettow
While
tired
had
back
troops
Rufiji.
were
and
to
the
von
pulled
suffered a number of casualties, they were still capable of offensive effort, although
limited by lack of transport. On the Mgeta front, Sheppard'sweakened I't EA Brigade
began
Brigade
Nigerian
to arrive
for
first
the
the
the
was readying
move as
elementsof
from Dar-es-Salaam. Echeloned to his rear, on the Ruwu-Tulo road, Beves's Force
Reservewas making similar preparations. In the west, van Deventer's 2ndDivision was
halted at Iringa, its fighting power largely dissipatedthrough disease,under-feedingand
it
in
Lacking
to
advance any strength, was
overwork.
sufficient porters or vehicles
force
Norforce
far
limited
forward
In
the
was
south,
capableof only
west and
movement.
in relatively good condition although it was still recovering from the exertions and losses
during the recent operationsagainstthe Wesuruppen.
In order to force a decision on the Rufiji, Smuts planned a three-pronged
Mgeta
River
German
the
Intelligence
had
the
that
on
positions
operation.
ascertained
had
body
fairly
The
light,
1,000.
withdrawn to the
than
main
were
not numbering more
Rufiji River area with the principal concentrationsat Kibambawe in the west and Utete
further east. Smuts's plan was to outflank and cut off the weak forces on the Mgeta by
sendinga column to the Rufiji at Mkalinso coupled with a rapid flanking march to seizea
crossingof the Rufiji at Kibambawe. Oncethe river had been crossed,the re-united force
would turn eastwardsand push along its southernbanks to Utete. This advancewould be
supportedby a flank column under Colonel Burne moving south from Kissangire to clear
the areanorth of the Rufiji, while Hoskins's I't Division pressednorth-west from Kibata
256
Utete.
Indeed,Smutshopedhis moveswould compelvon Lettow to surrenderthere.
to

215WO 95/5292,War Diary GHQ, 30 November 1916,Appendix 42, TelegramOA 198, Smutsto CIGS, 30 November.
236CAB 44/6, pp. 12-13;WO 95/5292, War Diary GHQ, 22 December 1916,Appendix 33, TelegramOA 355, Smuts to CIGS, 21
December."If the enemy's retreat is delayeduntil we have crossedthe RUFIJI near KIAGOBAWE [sic] and moved East so as to
effectjunction with HOSKINS this line of retreatwill be closed and enemy may be compelled to end on the LOWER RUFIJI."

78

The details were typically Smutsian - light frontal holding attacks in the centre
foes.
his
However,
in
be
to
to
encircle
were
coupledwith outflanking marches an attempt
lack
the
build-up
the
of stockpiled
even after the pause and
past several months
of
supplies in the forward areas, the poor health of the troops, and, above all, the
insufficiency of the transport were severely limiting factors in the execution of the plan.
It was a race, and in many ways a gamble, with fighting troops trying to cut off their
opponents,before the weather could do the sameto them.
The details of the advancewere as follows. Preliminary operations in the west
would be launchedto draw off the German reserves. Van Deventer's 2ndDivision was
orderedto clear the high escarpmenteast of Iringa and push eastwardsto Ifakara along
the Ruaha River, helping to close the trap from the west. Northey was to move east,
257
On
taking the position at Mfrika and driving Kraut's troops behind the Ulanga River.
the Mgeta front, a total of four columns would operateagainstthe forward positions. The
into
from
Dutumi
holding
launch
Nigerian
Brigade
to
attack
newly arrived
a
was ordered
the centre of the Mgeta line, while Sheppard's Is' EA Infantry Brigade would turn its
flank by moving from Dakawa to Wiransi in the west. Concurrently, a small, twobattalion column would make a similar march from Tulo to Tshimbe in east. Sheppard
would also detacha single battalion to conduct a deepflanking move againstthe enemy's
lines of communication stretching south-eastwardsto Behobeho. Key to the plan,
however, would be Beves's Force Reserve. It would start from Kissaki in the west,
making a wide sweep around all known enemy concentrationsto emerge on the north
bank of the Rufiji, near the village of Mkalinso. From there, it would march crosscountry to Luhembero, south of the river and astride the road running south from
Kimbambawe. Once the defendershad been rounded up, the columns would unite and
start the second-stagemove on Utete. At the same time, Hoskins's I" Division was to
prevent the eastward or southward move of the German main body, should it try to
258
in
directions.
escape those

257WO 95/5292,War Diary GHQ, I January 1917,Telegram OA 443, Smutsto Adminstaff, 31 December.
25'CAB 44/6, pp. 13-15.

79

Even if his army had not recoveredfrom its previous exertionsand the weatherwas
bad, Smutswas determinedto push on. Rain, although widely expected,was to upset his
23
December
Northey's
had
He
troops
and
on
moving
optimisticcalculations.
plannedon
theremainder2-3 dayslater, but his report of 20 Decemberwas hardly encouraging:
"Very heavyrain has fallen on all fronts and country Southof MGETA has assumed
appearanceof lakes. Weather now improving in RUFIJI area but heavy rains
continue to fall in areas covered by General NORTHEY and VAN
259
,,
DEVENTER...

Theseoptimistic prognosticationswere soon upsetby experience;on 23 December,

2ndDivision reportedthat rain had stoppedall transportfor 10 days,forcing its troops again
260
be
half
its
essential.
onto
operations would
rations, and that modifications to
Nevertheless,despite the continuing rain and mounting difficulties in movement, both
Northey and van Deventer began their advance as planned on 23 and 24 December
respectively.However,conditionswere lessfavourableon the Mgeta and north of Kilwa. It
was not until 29-31 Decemberthat the weatherclearedand not until New Year's Day 1917
that the main advancesetto start.
FIGHTING ON THE MIDDLE RUFIJI FRONT

Facing Smuts on the Rufiji, the main defenceforce, Abt Otto, with five companies
in the forward area, was supportedby Abt von Chappuis, with another three companies
261
lines
No other
followed
Kissaki-Behobeho
the
the
along
of communication that
route.
local reserveswere known to exist within the proposedbattle area.
On the RufiJi front, there were two major problemsfor the British commander. The
first wasthe absolutenecessityof seizingandmaintaininga crossingover the River Rufiji in
259WO 95/5292,War Diary GHQ, 20 December1916,Appendix 30, TelegramOA 341, Genstaff to CIGS, 20 December.
'" GSWA, Box 29,18 December1916,TelegramP32,2 Div to BGGS. Van Deventer
signalled "Have carefully reviewed supply
situation which is presentinggrave difficulties Only way out of difficulty is to move part of division back to railway
...
immediately..."; WO 95/5292, War Diary GHQ,
23 Dee 16, Entry 2d Division.
'6' MS Boell, N 14/30,26. Kapitel, 1904.Otto had
p.
eight companieswith von Chappuisas a subordinate;CA1344/6, p. 11.

80

!k

Tl I I-,

RUFIJI

n.

Plan for the Offensive

General Smuts's

-",

=m/

Kilo sa

d,
Morogorb'-

Mikese
/Mikese

e- NT

AUkkmafts,

# Kikumi
Kirengwe

Kidodi

V.0
Nlasenge
,

C0

)R'.U

Mfu; t

If k r-a-,4
-a

msanga

1,1110

9-

Duturni /
ati
Ki X
k11;

B
"o
'Behobeho
Kibanibawe
11!
!
k s7iikk,i
6

14

Nlitawe
It, v
Verl)
11
t3
f
A
Ma
Madaba
adabaA A,,

,--

Ahlv.s
21,

lo !2

fjV 141STORICAL
COMPILEP
(MILITARY
BRANCIO

20

SECTION

47-37

Miles
40

so

Ables
loo

NIAFIA
is y

(Uteteqj

Lines

of Communication

Posts

& Garrisons

General

-te
Mohoro

--------

Smuts
Ist

Proposed

WWI
Stage

2nd

do.

. -.

Stage

German
tAKT
Kibata
Div.
-Ast
Kilambi"
\',

'e,
-

mitole
mchemem

to I Inch
60

AUte

Nambanic

f)

Ngtalakida
%
Kikale

I Kibo

Utunge

Lill, -Ingo''
------------

seale

e,

Ndyimb, -,

Nanguwe
0
-.
Lukt&ro -, - Ngar,
Chim
Likwego

Jw

British

hUindaji

7
iji

Ungivam

C,

"iSi

REFERENCE

Koge.

Ole 0

I Mkindu
m
14ka inz(
Kibongr)
Ngwenibe
Alibab,

V-a

Kongo
11 kamba
Allst

,0

Malienge jj, w

jV11

eck

1(ida
1(idatu
Lof
ola

Maton ibo,

11
Nyn;

I,

KiseraNve

11

ULU&

a's

D.AR-ES-SALAAM
Irrz Vne

Ulcia

AREA

Detached

Forces

Posts

Kilwa

Kivinje

K,Jwa

Kisiwani

H-Q,, Westerr Force

I

Main Force (assurned position) C1

L. of

(as assumed)

----------

0"I'mom';

VIA

the face of strong defences. Assault craft were non-existent and a limited number of
vulnerablethree-manBerthon boats were his main crossing asset. These had to contend
with continuallyrising river watersand a strongcurrent. A daylight attemptto crossagainst
organisedoppositionwould almost certainly be disastrous. Any bridgeheadwould have to
be reinforced quickly against the inevitable counter-attacks,while ammunition and food
would also have to cross regularly. This meant that it would be essentialto widen the
lodgementas quickly as possible and to build a substantialbridge as soon as possible
thereafter.

The secondconstraintwas the needto contain and destroythe enemyforce before it

could slip off to the south. This would be bestachievedby achievingan undetectedcrossing
on the westernflank in conjunction with a move north-westwardby I" Division from the
Mturnbei Hills toward the Rufiji. If executedpromptly and vigorously enough,it offered
the chanceof cutting off von Lettow's main force from Kraut and Wahle in the west, and so
forcing him to end the campaign.262 Failure to achieveboth of these goals would leave a
weakenedBritish force in the depthsof the someof the wildest and most difficult country in
the world at the height of the tropical monsoon.
GeneralBeves's crossingof the Rufiji River at Mkalinso, some20 miles southwest
of Kibambawe,was critical to success. Starting on 31 Decemberand operating some 10
miles to the west of the main body, he made good progress,crossing the Mgeta on I
Januaryand having covered some 22 miles by nightfall on 2 January. Undetected,the
lead unit made the precarious crossing of the Rufiji near Mkalinso, with the far bank
securedby the morning of 3 January. Building rafts, Beves then brought his second
battalion over on the next day and was strongly positioned to move on his

next objective,

Kibambawe, the nexus of the German lines of communication. This had been
achieved
after a 30 hours'continuousmarchthroughvirgin bushandhad put the SouthAfricans about
24 hoursaheadof the plannedtimetable. Furthermore,the sole enemycompanyholding the
Mkalinso post remainedunawareof the imminent dangerto their position.

262Smuts,DesPatch111, 148.
p.

81

Early on 4 January,the leading battalion seizedMkalinso, dispersingthe garrison

information
first
his
this
Otto
about
the
telephone
received
and capturing
equipment.
force at midday and immediately despatchedanother company there. However, further
from
being
the
fears
west
outflanked
the
the
of
and
reportsof
enemy advance
strength of
led Otto to withdraw his entire Ateilung to the south bank of the Rufiji and to destroy
263
Kimbambawe.
the crossingsite at
Having achieveda major successand having turned the Mgeta position, the Force
It was true that the troops were exhausted by their
formidable exertions, but the urgency of the situation was well known and the chief
Claiming
in
the
located
Major
had
Pretorious,
area.
all enemypositions
scout,
personally

Reserve inexplicably halted.

264
his
little,
did
lack
from
GHQ,
Beves
advantage.
squandering
a
of explicit orders
now
Finally, on 6 January Smuts ordered him to attack north-east of Mkalinso, which he did
on the following day. The attack was not pressedhard and the garrison escapedthrough
the thick bush. Unknown to Beves, Otto had been forced to divide his force to protect the
line of communicationrunning from Njakissiku - Luhembero - Maba on 5 Januarywith
had
Pretorious
depredations
Furthermore,
to
the
already
the
of
only one company
east.
in
food
German
the
supplies
the
munitions and
captured entire
medical reserve,as well as
265
Otto
being
A
out at
the process of
catching
of
chance
real
very
evacuatedsouth.
266
initiative.
Kibambawewas lost through inaction and insufficient
On 31 December,Lyall's column was marching 12 miles eastof the main force on
the routeTulo to Kiruru. There,he crossedthe Mgeta and turned southto occupy Tshimbe
on the Kiderengwa-Behobehoroad. His task was to block the road in order to prevent the
Germansfrom escapingsouth;his patrolswere also to link up with the troops of Sheppard's
I st Brigade,ffirther to the west.267 By early on I January,he was south of the Mgeta and
reachedhis objective by early afternoon. Trenches were dug just as a strong enemy
force ran into the column. Successfully pushing the Germans north and capturing a
263MSBoell,N14/30,26. Kapitel,
pp. 1919-1919;W095/5292, War Diary GHQ, 4 January 1917,Appendix 4, TelegramOA518,
Smutsto CIGS, 3 January.
2" CAB 44/6, pp. 38-39.
265NIS Boell, N14/30,26. Kapitel, 1920.
p.
2" CAB 44/6, pp. 36-39.
267Smuts,DespatchIII,
p. 147.

82

THE
ADVAN'% '-,E TO TFM RLJFIJI
.
31s,t December 1916 -1-5th January 1917
Inch

MIAM
92
-.
d

/-. KITOHQ.

N_

Fre

3Q, =

KLsaki

5i
4f
.

"KIPEN10

PC"

2, '1
FUCAIr-,

/
/

2il

II-k<:.

2,11

311
HATAMSULO(

KIPAL(i_
Lake

2. )
Mhoulbi
MTUNDUSi,

sh
Sli
....

...

K il awbm"

Rapdx:

iapcuw

LUHVAaER0
4r,

REFERENCFN,

Z. d.

i
Shdpvard

*Glh saluchAs
eeve-s

----- 40-

L,Os of ;; traj,

0.
Do

Lyad
Crhffi

ez.,

Gcr-a

howitzer, Lyall then faced a strong counter-attack.Holding firm, he awaited the arrival of
the Nigerian Brigade, unawarethat Otto's troops had bypassedhim, having marched off
into the bush the night before.268
In the centre, the Nigerian Brigade made its holding attack on the forward
positions north of the Mgeta. Heavily supported by artillery and aerial observation a
battalion cleared south of Dutumi, finding a ford in the afternoon and reaching the
southern bank by early evening. The brigade consolidated its position throughout the
night, and resumedthe move south on the morning of 2 January, when it linked up with
Lyall. The acutelack of carriers meant that the Nigerians could take no further part in the
former
farther
to
their
they
the
were recalled
advance,and with
south,
enemy now much
269
fed
be
much more easily.
positions along the Mgeta where they could

Sheppard'sflanking battalion,aboutfive miles west of the main body,hadhada

it
Mgeta
difficult
from
Kissaki,
the
Starting
31
December
time.
on the
crossed
more
on
sameday andreachedthe Behobehoroadthe next morning. Thenturning to the northin
long
it
It
German
block
took
to
the
west,
withdrawal. was not
up position
expected
followed
by
hours
0730
the advanceguard
coming,with the leadpicquetsencountered
at
of At von Chappuis.A first attackwas held but then followed by a second,andmuch
heavier,effort. A desperatehand-to-handfight ensuedand the Indians were nearly
overcome. However,steadydefenceand the soundof Sheppard'sbrigadeforced the
Germansto breakoff the battle.270
The remainder of I" EA Brigade had marched south facing only light rearguards
covering the German withdrawal south. Linking up with the lead battalion on the
afternoonof I January, Shepparddecidedto send his main body south to the junction at
BehobehoChogwali in an attempt to cut off the enemy known to be further north while
also sending off another unit to take the village of Wiransi. The main body reached

26'CAB 44/6, pp. 22-24;

269CAB 44/6, pp. 25-27; Downes, With the Nigerians,
pp. 66-67.
270CAB 44/6,
pp. 29-30. The 130'bonly started454 all ranks and suffered some36 deadand 29 wounded, totalling 65. The Germans
had 12 killed and two taken prisoner.

83

Wiransi'on 2 January,but was unable to reach its objective of Behobeho Chogwali the
next day owing to intenseheat and difficult terrain.
In the meantime, Sheppardsent a battalion and a battery of guns to support Lyall
further to the eastin the areaof Behobehokwa Mahinda (north of Chogwali). Reaching
that place on 3 January, they linked up with Lyall's two battalions that night and the
I't
linking
Sheppard's
force
4
January,
the
up
with
combined
moved south on
morning of
EA Brigade on the main track in mid-afternoon. The next morning the advance on
Kibarnbawewas resumed, with Abt von Chappuis giving battle at about 1030 hours.
After driving off their opponentsinto the thick bush, Shepparddecided to push on to the
Rufiji at best speed.271
On 5 January,Is' EA Brigade,now with all its units back, reachedits objective,the
Kibambawebridge. Despitetheir fatigue and exertions,Otto had beatenthem there and the
crossinghad been destroyed. The failure of Beves to move on 4 Januarynow became
doubly important Not to be deterred,SmutsorderedSheppardto make a crossingduring
the night of 5/6 January.As the river was between 400 and 700 yards wide with a swift
found
Patrols
he
faced
a narrow
current and unfordable,
was
with a major problem.
point, some400 yards in width, about one and half miles from the bridge site and out of
direct enemyobservation. Severely limited in his crossingcapability, Sheppardhad only
sevenBerthon boats capableof carrying three men at a time. Despite the inadequacyof
his means,he beganferrying troops acrossthe river during the night of 5/6 January. Only
one companywith two machinegunsmanagedto crossbefore daylight and it had to spend
the entire day of 6 Januaryhiding amongstreedsand suffering intensely from the heat and
lack of food. Through good discipline and camouflage,it managedto remain undetected
despiteGermanpatrols. With darkness,Sheppardwas able to send acrossthe remainder
of the battalion plus another company, giving him 400 rifles and 4 machine guns on the
272
bank
by
first
light
7
January.
south
on

271WO 95/5292,War Diary GHQ, 5 January1917,Appendix 8, TelegramX 1133,Smutsto CIGS 4 January;CAB 44/6,
pp. 32-35.
The British suffered 16 KIA and 45 WIA including the famous FW Selous.
272WO 95/5292,War Diary GHQ, 7 January 1917,TelegramOA 567, Genstaff to CIGS, 7 January;CAB 44/6,
pp. 4244.

84

The British lodgementwas detecteda by patrol early that morning and, believing
them to be a weak detachment,a single company attacked it but was firmly repulsed.
The appearanceof the enemy on the south bank spurred Otto to move up a company to
273
from
Mpanga.
Four boats
Kunglio as reinforcementswhile also recalling two others
attemptingto land supplies during daylight were damagedby shrapnel and the resupply
effort was ended. A attack was then launched from the bridgehead, first with one, and
then two companies,but it was countered by strong German reinforcements. Facing
severeshortagesby afternoon, ammunition was rowed acrossthe river at great risk and
274
British
bridgehead
the
was stabilised.
275
lost
heavily
in
The Germansdid not counter-attackfurther, having
Europeans.
Sheppardusedthe night of 7/8 Januaryto ferry acrossthe rest of the secondbattalion and
an additional company, giving him a total strength of 600 rifles and 10 machine guns
acrossthe river. Early on 8 Januaryanotherattempt was made on the British lodgement,
but it was easily held off with the support of the mountain guns firing acrossthe river.
The arrival of two long-range naval guns on the north bank of the Rufiji assistedthe
forward troops and helped to suppressthe German guns. However, the combination of
insufficient strength and the high ground dominating the bridgehead meant that a
stalemateensuedover the next week.
On 7 January, the German positions were as follows: one company was at
Njakissiku, three were around Kunglio-Sud, two more at Mpanga, and one each at
Mkalinso and Mkindu. Otto was now faced with a threat from the west via Mkindu while
also having to prevent further expansion of the British bridgehead. Keeping three
companieson the river, he reinforced Mpanga back to three, with two in reserve,ready to
276
further
initiatives.
to
react

In the meantime, on 8 January Smuts had arrived at the Mkalinso crossing and
orderedBevesto re-concentratehis force at his original crossingsite. Two battalionsof the
m MS Boell, N 14/30,26. Kapitel,
pp. 1923-1924.
M WO 95/5292,War Diary GHQ, 7 January 1917,Appendix 11, Telegram OA567, Genstaff CIGS, 7 January.
to
275MS Boell, N14/30,26. Kapitel,
pp. 1923-1924. The Germanslost two kiled and six wounded.

85

NigerianBrigade were ordered from Dutumi, to move west then south and pass through
Beves'sexhaustedcolumn.
The loss of 3,000 kg of suppliesat Maba through aerial bombing was a blow and
German
The
ftu-ther
for
Otto
command was
to
south.
withdrawal
a
convinced
prepare
line
the
the
becoming
increasingly
of
main
safety
of
apprehensive about
also
by
Protected
Mahenge.
Luwegu
to
from
Rufiji
thence
to
the
and
communicationrunning
former
lay
the
detachment,
the
place and was
at
main ammunition reserve
only a weak
lines
to
headquarters
hurriedly
The
the
ordered
communication
was
of
of
evacuated.
now
lines
in
from
January
to
the
17
Madaba
Mlembwe
out of
to
re-align
an
attempt
on
move
277
British
rangeof the
advance.
While the fighting forces had made surprisingly good progressdespitetheir weak
had
British
transport
behind
basis
the
their
system
them the
survival,
condition,
of
collapsed. There had never been much of a margin of safety and now the administrative
in
have
he
to
the
numbers
that
reduce
would
staff warned the commander-in-chief
forward areassubstantially if actual starvation was to be avoided. During the previous
two weeks, with the combination of sicknessamongstthe drivers and the breakdown of
Full
its
force
had
10
rations
supplies.
the
essential
vehicles,
per cent of
received about
be
half
to
be
20
January
would
rations
could only
and a reduction
maintained until
278
if
Faced with this
necessaryby the end of the month, even sooner the rains came.
unmistakablewarning, Smutshad to amendhis plans
On 15 January 1917, on the eve of his departure from East Africa, he gave the
Nigerian Brigade a new role. Believing the enemy to be in small numbersat Kibambawe
and Mpanga's, a few miles to the east, he ordered Cunliffe to trace Beves's route to
Mkalinso, then to strike east towards Mkindu and Luhembero. Beves was to march

276MS Boell, N14/30,26. Kapitel, pp. 1924,CAB 44/6, pp. 4445.

2"MSBoell, N14/30,26. Kapitel,
pp. 1925-1927.
273CAB 44/6,
pp. 108-110. ReproducesMemorandum"Supply and Transport Situation", by AQMG 9 January 1917; GSWA, Box 29,
2 January1917,TelegramP 71,2 Div to Genstaff. 2d Division had beenon half rations since 24 December.

86

linkingMpanga's
up with
and
north-east, attacking the enemy grouping around
279
Sheppardwho was to break out south at Kibambawe.
It is noteworthy that his communiqu6sto the War Office were far more bland and
had
Sheppard
he
9
January,
that
his
On
stated merely
sanguinethan those of
staff.
crossedthe RufiJi at Kibambawe, while Beves had reachedMkalinso and was also across
the river, with Cunliffe marching to join him. No mention of the poor weather or
be
his
last
interestingly,
difficulties
to
this
was
official report
was made;
enormoussupply
final
handover
His
his
days,
just
for
to
of
command.
prior
on operations
another nine
been
had
Kibambawe
Rufiji
broadly
the
that
of
south
report was
positive and noted
Kibata
from
the
forces
had
the
Hoskins's
the
river
clearedand that
reached easternpart of
280
area. Again, no mention of the weatheror other problems was made.
It was now clear to the Germans that the main force at the Ruaha mouth was
British
for
detected
the
had
during
the
next
the
they
preparations
pushing east pause
became
Otto
build-up
Mkalinso.
the
concerned and telegraphed
move, notably
around
Lettow for instructions, being told that his main task was the protection of the lines of
he
decided
to
In
between
Madaba.
Maba
the
weaken
circumstances,
communication
and
his force along the Rufiji to only two companies,marching himself with four others to
Mkindu to face the main enemythreat.
After a clever feint operation, assistedby massedgunfire, l't EA Brigade was able
to break out of its perimeter in the early hours of 18 January. Kibambawe was cleared
the next day, the rearguard putting up only light resistance,while Beves's force made
slow, if unimpeded,progresstowards Mpanga's. The Nigerians operating to the south
had left the Mkalinso crossing site on 17 Januaryand promptly turned out the defenders
from Mkindu the next day. This left Otto with four companiesin Mkindu Sud with one at
Mabawith two further north, falling back from Nyakissiku.281

279CAB 44/6, p. 50; CAB 45/19, Nigeria RegimentRecord, 11. '

p.
280WO 95/5292,War Diary GHQ, 9 January 1917,Appendix 15, Telegram, OA 594, Smutsto CIGS, 9 January. The next telegram
was 18 January1917,Appendix 18, TelegramF 6339, Smutsto CIGS, 18 January.

87

However, the advance south across the Rufiji was now stalled; Sheppardhad
increasedthe size of his bridgehead,but lacked the offensive power to break-out. It was
lower
Rufiji
dinning
forces
becoming
the
Lettow's
area.
that
of
out
were
also
von
apparent
The gap betweenthe western limit of 1" Division's advancenear Ngarambi and Beves'
force.
Rufiji
the
the
Mpanga's
to
than
escape
of
permit
positionat
wide enough
was more
On 20 January,the Nigerian Brigade attackedOtto's positions. Unable to deal with such
strength,he had planned a fighting withdrawal to new positions further south near
Mkwembe. Deployed in depth around the important waterholesthere, he faced renewed
Heavy fighting with repeatedcounter-attacksby both sides
in
loss
by
day,
the
the
through
the
of much
resulting
carriers
continued
next
with panics

attackson 24 January.

by
battle
fine
heavy
by
It
and a
performance
equipmentand supplies the attackers. was a
282
Otto who had held off much superiorforcesin difficult conditions.
FIGHTING NORTH OF THE LOWER RUFIJI

The situation north of the lower Rufiji was largely unchanged by the end of
December1916. The defenderswere estimatedto have no more than 800 rifles and 10
Smuts
2
7
1,100
the
guns.
machine
guns
and
machine guns against
rifles,
attackers'
Given
Germans
the
his
lGC
force
the
the
the
to
river.
of
south
expected
and
scratch
push
inadequateresources, an attack on 3 January failed to take Kibesa.283 Finally, an
experiencedinfantry officer was placedin command. An attack on Mkamba on 9 January
fell on emptypositions,as did a secondon Kibesa. With only the major post of Kissengire
left, a flanking march was launched. The main column marchedsouth,linking up with 300
rifles who hadbeenlandedby the navy on the coaston 8 January.This move was eventually
successfuland by 18 Januarythe Germanshad beenforced out of Kissengire,withdrawing
acrossthe Rufiji. There was no further opposition sincethe river was in full flood and all
boatshad beenremovedto the far bank. The ill-effects of the now heavy rainfall were also
becomingvery apparent,asflooding impededthe march andthe sicknessrate rose.284
211MSBoell, N14/30,26. Kapitel,
pp. 1928-1931.
282MS Boell, N14/30,26. Kapitcl,
pp. 1936-1938. The Germanslost nine killed, 13 wounded and three missing out of some472
soldiers,making some5 per cent of strength. The British had 76 killed, 16 wounded and 15 taken prisoner out of about 1,000soldiers,
making some II per cent of strength.
233CAB 44/6, pp. 85-86. This
comprised 129Germans,630 Askaris, 10 machinegunsand3 guns.
2" CAB 44/6, pp. 87-91.

88

FIGHTING ON THE KILWA FRONT

Since the loss of their lodgementon Picquet Hill in mid-December,the Gen-nans
had remainedon the defensivein the Kibata area. Abt Schulz remained in the area with
four companiesand one 10.5cm.gun. Its orders were to mask the movement of the main
bodybackto the Ruf1ji andto preventany follow up by the enemy.
Smuts was keen for 1" Division to advance towards Utete, but the German
'which
in
lacked
transport
turn
the
to
too
artillery,
without
strong assault
positionswere
detected
had
I"
Division
the
By
January.
Kibata.
to
end,
year
until early
needed reach
P
it
thinning out of the Germanposition, and on 5 January ordered EA Brigade to advance
Brigade
Nigerian
the
linking
the
Hills
Mtumbei
on
the
the
with
up
and
with
aim of clearing
Rufiji. 2ndEA Brigade was held back in reserve,partially owing to a severeshortageof

carriers.
Using heavy artillery fire and aerial bombardment,the British were able to dislodge
P
the now-weakenedopposition with a seriesof limited flanking moves. On 7 January,
EA Brigadebeganthe clearanceof the areaaroundKibata, followed by subsequentmoves
flank,
Lettow
his
Concerned
ordered
the
von
about
and minor actions over
next week.
285
Schulzto hold the areaof Mbindia andto conducta vigorous opposition.

fall
back
Schulz
This wasnot easyin the circumstances,
to
slowly,
continued
and
destroyingthe Kbnigsberggun owing to the lack of ammunitionandthe utter exhaustion
of its carriersand crew. The British pressurecontinuedwith the captureof a magazine
and threateningthe supplies stockpiled at Pungutini. Now concernedabout the situation,
286

von Lettow despatcheda reinforcing companyto Utete.

'3'MS Boell,N14/30,26.Kapitel,pp. 1941-1942.

2'6MS Boell,N14/30,26.Kapitel, 1948-1952;
WO 95/5292,WarDiaryGHQ,7 January1917,Appendix11,TelegramOA 567,
pp.
Smutsto CIGS,7 January-,
WO95/5330,Historyof 1'12"dKAR,pp.8-9.

89

00, Yc'
lIP61

0.

-9z
)

II
5L
L:
<

r. Me
-*-0
45
4
cri,

'2 " -i

'at, "

m="

%',!Z

tI,

04
Z' a

Is

'.

10 40

C14

J-Q-

I
S
t

ut -C

03
I

0-

1 ZLz

">.

1611
'1

_wbl
Ir
it

z
0

LLI

a4

j'Iz

aIt
Li
LL)
LL

uj

aAp
CL

His headquarters had moved to Lake Utungi on 7 January with Abt von
Haxthausenand von Lieberman following behind. After an exhausting march, the
reinforcementswere concentratedat Ungwara, an important depot south of the lake, five
dayslater. Immediately on arrival, von Haxthausenwas pushedeastto defend Ngarambi
Chini on 13 Januaryto block the enemy advance. It was successfully occupied, but the
287
rd
into
later
day.
The British
troops ran
the same
the 3 EA Brigade at Ngarambi Juu
now moved on Ungwara on 17 January, so forcing the deployment of the sole reserve
companyto support von Haxthausen. At the sametime, Schulz was encounteringsevere
supplydifficulties and his reserverations ran for anotherfour days.
From the fighting, it was becoming clear that the British were heading for the
Lake Utungi and Rufiji River. The earlier attack on Kibambawe and north of the river,
indicated that they were make a large convergent movement. 288 The German supply
situation in the area was becoming increasingly severe as the area north of the river was
now lost, with only 7,000 kg of food left in Koge.

The main stocks of food, some

175,000 kg, were in transit around the junction of Mpanganja Logeloge Lake Utungi
and it was vital that they were safely moved south. Apart from the physical difficulties
in holding the British pincers apart, the supply situation made it imperative to move some
of the forward forces south of the river. Abt Tafel was ordered south of the Ruf1ji, with
Ungwara ordered to be evacuated on 12 January. The main body of troops, including Abt
289
Lieberman
UWngi.
Lake
von
was concentrated around

Further concentration of the Germans around Lake Utungi and Utete was set in
train on 21 January. At the same time, the I" Division resumed its pressure on Ngarambi
Juu, while Schulz received further reinforcements. Lack of water forced him to abandon
Kiwambe on 23 January, despite von Lettow ordering a stiff resistance. The use of KAR
mounted infantry to destroy ripening crops on 27 and 28 January was another blow to the
Germans.290

2"7MS Boell, N 14/30,26. Kapitel,

pp. 1953-1956.
23$MSBoell,N14/30,26. Kapitel,
pp. 1959-60;LettowReminiscences, pp. 174-175.
28')MS Boell, NI4/30,26. Kapitel, 1960.
p.

90

The Is' Division continued to press hard, with O'Grady pushed up to the Rufiji,
capturingMohoro on 18 January while Hannyngton cleared the country west of the
MtumbeiHills, north of Ngarambi.291Further west along the river, Utete was taken on 31
292
February.
Januaryandthe north bank of the Rufiji was clear by the beginning of
To the Germans,it appearedthat British main effort was in the Kilwa area and
293
The worsening of the weather
from
Rothe
thatAbt
the north.
neededreinforcements
on 25 January hindered both the attackers and defenders. On 1 February, von
Haxthausenfell ill and G6ring took command of his Abteilung. The deployment of the
forces in the east was as follows. Abt Tafel with five companies was in the area of
Mpanganjaalong the Rufiji river. Abt G6ring also with five companies was securing
Ungwara and its approahes. Headquarterswas on the Utungi-See with Abt Willman's
two companiesat Nanguwe. Further to the south, holding 2ndEA Brigade at bay, was
Abt Rothewith three companiesat Mpotora.294
The British held the general line Njinjo - Ngarambi Chini - Kiwambe - Utete.
3rdEA Brigade occupied the centre with two battalions forward, while 2ndEA Brigade
spreadits three battalions between Utete and Kibata. The increasing rains had made
motor supply to the forward areas impossible and two of the bridges over the River
Matanduwere washedaway; Hannyngton was forced to move his troops back to Kibata,
295
O'Grady
for
forward
began
back
Kilwa.
The
the
towards
to
while
rations
also
pull
296
troops were of the meagrest.
FIGHTING

IN THE WEST

The plan to draw off German strength through a diversion in the west was bedevilled

by difficultiesbeforeit evenstarted.VanDeventer'slinesof conununication

backto Kilossa
2' MS Boell, N14/30,26. Kapitel,
pp. 1966-1971.
2" WO 95/5292,War Diary GHQ, 18 January 1917,Appendix 3 1, Telegram F 6339, Smutsto CIGS, 18 January.
292WO 95/5292,War Diary GHQ, 31 January 1917,Appendix 63, TelegramOA 27, Genstaff Dutumi to Genstaff, Dar-es-Salaam,31
January;WO 106/273,History of 3 KAR, pp. 60-61.
293MS Boell, N 14/30,26. Kapitel, 1986.
p.
2" MS Boell, N14/30,26. Kapitel,
pp. 1977-1980.
2" WO 95/5292,War Diary GHQ, 18 February 1917,Appendix D, Telegram GO 687,1' Division to Genstaff, 18 February.
296Thatcher,TenthBaluch Regiment,
pp. 169-170.

91

had
begun
Ruaha
River
in
Dodoma
had
disintegrating
the
when
and
mid-December
started
to flood. By 19 December,the bridge over that river had beenwashedaway and the roads
in
fleet
half
his
MT
With
beginning
into
dissolve
to
thick
continually
over
were
mud.
impossible
build
it
became
to
food
up
workshops,
even more precariousand was
supplies
297
by
Any
seriously
was
weakened
potential
any substantivestocks.
remaining offensive
the necessityof placing the troops on half rations from 23 December. Further south,
GeneralNorthey had similar problemsand, as early as 13 December,he estimatedthat only
45 out of the 350 miles of motor roadswere usable. All suppliesnow had to be carriedby
298
before
forward.
from
Mountains
Lake
Nyassa
Livingstone
porter
going
over the
Regardlessof the determination of the troops, both forces began the operations under
immensehandicaps.

On the Germanside,their forcesremainedspreadthrough an arc basedon Mahenge.

GeneralWahle, now the Westbefehlshaber,was based there with a single company in
likely
British
detachments
deployed
the
to
the
against
reserve,while six
north and west
were
lines of approach. Abt Schoenfeld,with 3 companies,was based on Ifakara and faced
generallynorth, holding the areabetweenthe Ruahacrossingat Kidatu. Abt Lincke, with
two companies,was deployed in the highlands about 30 miles south-eastof Iringa and
blocking the track from Dabagato Muhangu. hmediately to his south was Abt Wintgens,
with four companiesand a battery of guns at Makua's. He was centrally positioned and
poisedto reinforce either to the north or south. Next was Abt von Langennwhich had four
companiesaroundMfirika. Abt Kraut was oppositeNorthey's main force at Ifinga, some20
miles distant.He was actually down in the valley of the Ruhuje River although still above
the flood plain. Finally, Abt Grawert was deployedopposite Colonel Byron's garrison at
Songeamuch fartherto the south.299
Food stockswere asmuch a problem asthey were for the British. Mahengewas the
centreof the supplysystemfor the forward troops. However,two disastrouslydry yearsand
lack of co-ordination meant that this formerly fertile area was unable to maintain these
2" WO 95/5292,War Diary GHQ, 19 December1916,TelegramP 32, GOC 2 Div to Genstaff, IS December.
2" WO 95/5329,War Diary Norforce, 13 December1916. The
mountainsreached9,000 feet in altitude and causedconsiderable
sicknessand deathamongsthis porters.

92

forces.300 This was confirmed by Wahle's order that directed all Ateilungen, other than
district
Songea
their
the
Lincke
Schoenfeld,
the
of
source
to
thoseof
north
make
and
301
freedom
the
tactical
the
commanders.
This
of
on
rations.
would causemajor restrictions
On the British side, the 2ndDivision was a shell of its former self and represented
horses
Its
like
division's
troops
and
almost
all
without
were
mounted
strength.
nothing
a
had
incorporated
Brigade
the
Mounted
SA
I"
infantry.
The
to
were required
act as
integral
its
nd
Brigade
Mounted
2
SA
units.
as
own
the
as
well
now-departed
remnantsof
However, Us still only provided four regiments, each fielding a single mounted troop;
for
As
1,500.
formation
than
fighting
numberedno more
strengthof this
overall,the total
800
battalions
nd
totalling
infantry,
to
two
SA
Infantry
Brigade
2
about
the
the
was reduced
rd
SA
3
disbanded
SAI,
the
I
O'h
joined
by
the
They
of
the
remnants
composite
soldiers.
were
Infantry Brigade, now only 550 strong. Somewhatstrongerwas an Indian battalion with
800, as well as a KAR unit with nearly 500 rifles. The artillery was limited to four 13for
difficult
but,
to
the
conditions movementand
owing
poundersand eight mountain guns,
"'
latter.
the
lack
to
two
horses
of
a
of
and mules,therewere only sufficient animals move
General van Deventer's pleas having been ignored by Smuts, he now made
C,
lettered
A
into
to
force
his
fulfil
his
divided
three
He
with
columns,
to
arrangements
role.
flanking
He
790
1,450,1,150
of
a
series
planned
respectively.
approximatestrengthsof
and
leading
Nussey
General
At
A
the mounted
Lincke,
Column
under
marchesagainst
while
troops on a southernRank march and Column C under Colonel Taylor making a similar
The aim was for the flanking columns to converge on
Muhanga and cut off Lincke on the Magomi Ridge before Column B under General
Berrangdlauncheda frontal attack from Dabaga on 25 December.303 Little was known

envelopmentfrom the north.

difficult.
it
be
the
terrain
to
that
traversed
and
about
was mountainous
except

'" CAB 44n, footnote to p. 4; CAB 45/72, Sketches74,83 and 84.

3' Schnee,Deutsch-Ostafrika,p. 274.
3" CAB 44n, p. 30. It was capturedby Hawthorn's troops on 5 January 1917.
302GSWA, Box 29,18 December1916,TelegramP 32,2 Div to BGGS, for lack of supplies and strength.
"3 CAB 44/7, pp. 7-9. A Column had 1450soldiers and eight machine guns; B Column had 1150 soldiers and eight machineguns; C
Column had 790 soldiers and eight machine guns and two mountain guns for a grand total of 3,390 soldiers,24 machine guns and two
mountain guns;GSWA, Box 29,22 December1916,Operation Order 24.

93

Good, sunnyweatheron 20 and 21 Decemberdeterioratedinto rain as the advance

began. Column C suffered terribly from the heavy rain and cold, particularly the nearly
desertion
from
A
brought
rapidly
the
and
of
sickness
plains. combination
nakedcarriers
up
left
be
had
heavy
to
tents
the
stores
and
column and all
reducedthe carrying capacity of
behind. The driving rain and slippery conditions soon exhaustedthe column and on 25
December,Colonel Taylor with a battalion pushed on alone to Muhanga. A short fight
drove out the small German detachmentthere and the road from Dabagawas effectively
its
its
forward
day
During
the
the
the
way
prepared
made
and
column
closed.
remainder of
defences.304

On the southernflank, A Columnconcentrated

on the easternbank of the Little
RuahaRiver. On 22 December,the flanking move was seriously delayed by heavy rain
days
day.
After
two
the
two
to
there
of climbing
next
with
units unable arrive
until
its
intermediate
had
Column
A
thick
objective
reached
mountainsand crossing
swamps,
southon the Magomi ridge and preparedto setup its block. Two regimentswere orderedto
holding
Taylor's
link
to
to
one
the
with
column,
up with
press on
north and east
Makungwa'sand the fourth in reserve. This plan requiredthe whole brigade,not more than
1,200 strong, to be spread out over a frontage of some 10 miles of the worst country
305
impossible.
possible,with tracks so difficult that passageon foot was almost
Lack of accuratemapsand guidesreducedthis plan to confusionasthe link-up force
spentmuch of 25 Decemberin a fruitless march looking for Muhanga, not arriving there
between
high
following
day.
Brigade
Headquarters
the
the
ground
until
was establishedon
that placeand Makungwa's, but communicationswere limited to runnersowing to the rain.
By 26 December,the trap appearedto be shut.306

In themeantime,the frontalattackplannedfor Berrangd'sB Columnwasunderway

on 25 December. Despitethe lack of guns,which had beendelayedby flooding rivers, the
attackwas reasonablysuccessfuland the troops dug-in within 250 yards of the Germanline
3" CAB 44/7, pp. 12- 14.
305CAB 44n, pp. 16-17; GSWA, Box 29,24 December 1916,Telegram D 1484,Brigzar [I SA Mounted Brigade] to 2 Div. The
brigadecommanderconsideredit "almost physically impossible cover distancesin time allowed."

94

:A..

.. z

1) J,-,

ci

rr,

: a

It

\,\ v,

Iv1.

li

LL:
UX

CL

z.

r- -JC,

-I.

-%

----.

as

vo

21
"1
,
I.

.--

oly

.". vow

cIt...

0-4
J-4

in

.2*

.-

-C

iL

CIO 9:

r -4:2---v

<zl
C13

w
11

Iz

(5)

."

AV

.;.

(' 0-+.

An

-4

--f
0

iz,

.1 "3

-n:- *-, .

...:

Ud
Z
0

UW

XQ-c
0

101,

-Z
r

91tc-

between
forces
fire-fight
intensive
in
the
two
few
losses.
day
The
with
was spent an
next
forward
27
but
infiltration,
followed
by
on
early
went
patrols
when
and was
a night
December, the trenches were empty, with the defenders having withdrawn to the south307
east.

Van Deventerhad hopedto captureLincke at Muhanga. Unfortunately for him, the

Germancommanderchoseto escapevia the village of Makungwa's, moving towardsAbt
Wintgens.The defenderswere now well dug-in, but the shifting of the reserveto Muhanga
Mission meant that the whole of Lincke's force, still nearly 500 strong, was able to
concentrateagainst about a third of A Column. Owing to the weather, the reserveonly
arrived at midday 28 December,by which time they were too tired to launch a night attack
late,
It
too
the
asAbt
next morning. was
againstan unreconnoitredenemyand waited until
308
Lincke hadhackeda path throughthe bushand escapedto the SoUth.

AlthoughMuhangahadbeencaptured,Abt Linckehad escapedlargelyunscathed.

Orders for a further advance were countermandedas the transport system had nearly
nd
2
December,
By
30
the
collapsedunder
strain of submergedroads and swollen rivers.
Division hadjust four days' rations for the Europeantroops and none for the Indians. Either
"'
follow.
Accordingly,
took
van
a rapid evacuationwould
would
place or starvation
Deventer ordered most of his South Africans back to the railway while leaving two
battalions,oneIndian and one African, to hold the line.310On 2 January1917,van Deventer
informedthe Commander-in-Chiefof the desperationof his situationand that he was forced
to abandonthe advancethrough lack of rations.311 This was virtually the last action of that
formation,for five dayslater GHQ announcedthat owing to the supply situation it would be
withdrawing it to Dodomaas a preliminary to returning van Deventerand his SouthAfrican
312
back
Union.
to the
whites

306CAB 44n, P. 18; GSWA, Box 29,28 December 1916,Telegram D 1491,Brigzar to 2 Div.
307CAB 44n, pp. 14-15.
"' CAB 44n, P. 19; GSWA, Box 29,28 December 1916,Telegram KT 422, Col Taylor to 2 Div; Telegram [n.ref,] Col Kirsten to 2
Div.
'O'GSWA, Box 29,27 December1916,Letter ADMS to GOC 2 Div. Over 40
casesof pneumoniahad beenreported in three days
and much more was feared. This outbreakwas put down to "(I) exposure(2) totally inadequatefood."
3'0CAB 44n, p. 22.
311WO 95/5202,War Diary GHQ, 2 January 1917,Appendix 3, Telegram P71, GOC 2 Div to GHQ, 2 January.

95

Colonel Taylor was left in command of the remaining two battalions and took up
highland
hinga
force.
The
the
Lincke's
high
overlooked
the
positions on
ground opposite
flooded valleys of the Ulanga valley that stretched for many miles below. As the month ran
had
it
the
decreasing
that
enemy
clear
on,
patrol encounters and skirmishes soon made
begun to thin out their forward positions in this area and were transferring troops to the still
313
ftirther

militarily active areas

south.

HOSKINS ASSUMES COMMAND

With General Smuts' departurefor the Imperial War Conferenceon 23 January,
Lieutenant General Hoskins now assumedchief command. Leaving Is' Division in the
His
his
but
he
had
little
the
to
plans of
predecessor.
carry on
midst of an offensive,
choice
forceswere too widely spreadand communicationswere too poor for him to make major
following
faced
do
He
in
in
to
the
the
troops
so.
condition
changesand any event
were no

situation.
In the middle Rufiji, Sheppard'sI't EA Brigade and Beves's2ndSA Infantry Brigade
held both banks of the river around Kibambawe while Cunliffe's Nigerian Brigade had
Hannyngton,
lower
the
On
Mkindu.
now
the
to
of
river,
part
pushed south and east
delta,
Mohoro
EA
Brigade
the
had
his
2nd
the
I't
Division,
the
on
edgeof
at
commanding
P
EA Brigade at Ngarambi on the fringe of the Mtumbei Hills. Both groupswere
with
trying to link up and trap von Lettow's main body, which was currently located between
Mkindu in the west and Utete in the east.314
In the west, van Deventer's attack had fizzled out several weeks earlier and his
division was strung out between Muhanga and the Central Railway. Further southwest,

Northeyhad clearedthe high groundand had driven Kraut eastof the RuhudjeRiver.
Songeawas secure,althougha furtheradvancetherewastemporarilyout of the question
312WO 95/5292,War Diary GHQ, 7 January 1917,Appendix 11. TelegramOA 567, GeneralStaff CIGS,7 January.
"' cAB 44n, pp. 22-23.
314Hoskins,Lt GenA R, Despatchfromthe GeneralOfficer Commanding-In-ChiefEast
Africa Relatingto Operationsin EastAfrica 20*
Januaryto 31,wMay 1917,dated30 May 1917,publishedin the LondonGazette'Supplement',No. 30447,27 December1917,pp. 156-157.
Henceforth,Hoskins,Despatch.

96

his
ftirther
flooding.
Wahle
to
the
and
position was
east
reserves
owing
retained strong
securefor the time being.
None of this reflected the disastroussupply situation and Smuts's optimism had
in
days
in
its
final
After
three
London
that
the
a mere
stages.
persuaded
campaignwas
forces
Indian
his
British
instructed
by
CIGS
Hoskins
to
the
and
reduce
command,
was
315
for
other theatres.
significantly in order to release shipping and manpower resources
Having just come from the front himself, Hoskins knew only too well the exhaustedand
worn-out state of his command. He was dealt a further blow on 25 January when the
rains struck even harder and the entire area became a swamp. He recognised the
dangerousstateof affairs, as he later stated:
"But the supply and transport situation was not all satisfactory. There was no
insufficient;
in
the
depots;
the
the
porters
was
of
number
reserve
advanced
drivers
dying
transport
in
the
the
transport
of
mechanical
and
animals
units were
line
in
front
falling
the
troops
that
the
could not
numbers of
were
sick so rapidly
be maintained there [the Ruf1ji]. ,316

By early February 1917,the end could be put off no further and Hoskins was forced
317
In that telegram, he outlined that the
Robertson
to signal
that the offensive was over.
Germansupply position was weak but that he simply lacked the transport and meansof
he
do
in
it.
Northey
to
to
the
as
still
able
operate
mobility
anything about
south-westwas
was on much higher ground,but nonethelessthe rainfall there was still substantial. Beaten
by the weather and the lack of adequatepreparations,the campaign would have to wait
until May.

3" WO 95/5292,War Diary GHQ, 25 January 1917,Appendix 46a, Repeat Telegram OA 846, CIGS to Hoskins, 22 January.
of
3'6Hoskins,Despatch, 157.
p,
317WO 95/5292,War Diary GHQ, 14 February 1917,Appendices,TelegramOA 261, Hoskins to CIGS, 14 February.

97

OPERATIONS IN THE FAR SOUTH AGAINST THE PORTUGUESE

Anglo-Portugueseco-operation proved to be more difficult in practice than in
theory. General Smuts feared the worst of their military abilities and wished the
Portugueseto do no more than move up to the line of the Rovuma in order to block a
German withdrawal south, although he later envisaged a limited advance towards
Liwale.318 Despite the British intention to control the coastline themselves,and under
decided
had
from
Gil
his
General
to march on the
already
pressure
own government,
319
inflicted
by
for
defeat
Seeking
the
Mikindani.
the
small coastal port of
revenge
Germansearlier in the year, his troops set out from the port of Palmas on 10 September.
The news of the British occupation of that place changed his mind and he altered his
in
delta.
land
This
Rovuma
Kionga
Triangle,
the
to
the
objective
a parcel of
undefended
had
He
honour
19
September
the choice of
was occupied on
now
was satisfied.
and
lay
just
Newala
harsh
Plateau,
Makonde
to
the
which
advancingover
or moving
and and
to the south-westof that upland, or remaining along the north bank of the Rovuma. In
320
just
inside
he
German
territory.
the end, choseto maintain his position along the river,
Although Smuts and the British had a very low opinion of their allies, diplomacy
forced a degreeof dialogue. The PortugueseGovernmentbelieved, from British sources,
that the campaign would be over in two to three months and wanted to obtain a
bargaining chip for the peace negotiations. This, together with Smuts's own
321
Newala.
Gil
inland
towards the nodal village of
recommendation, made
march
However, a rebellion againstPortugueserule broke out along the Zambezi and a number
of his troops were withdrawn from the southernend of Lake Nyassa. This limited Gil's
capabilities further, although Smuts expressedthe hope that they could be used to deny
Germanforaging parties accessto the southernpart of the Germancolony.322

31"W033/858, TelegramsD 1, No 1262,26 September1916,Telegram X757, Smutsto CIGS, 337.

p.
311Cann,John,"Mozambique
and GermanEastAfrica7, pp. 131-132.
320Hordem, Military Operations East Africa,
pp. 390-391.
311Ptlissier, Naissancede Mozambique,11, 692; Cann,"Mozambique", 133.
p.
p.
322WO 95/5292,War Diary GHQ, 22 October 1916,Entry PortugueseForce.

98

This activity causedthe Germans a degree of alarm as the Portugueseposed a

potential threat to their southernfood-growing areas. While they too had a low opinion
of their fighting abilities, the sheerlack of military resourcesin the region was worrying.
CaptainLooff, the newly appointedSadsbefehIshaber,
had only meagreforces, consisting
mainly of three companiesof young recruits. As this force was also responsiblefor the
containmentof the British enclaveat Lindi, it was clear that reinforcementswere necessary.
Assessingthe situation, von Lettow decided to send him 4bt Rothe consisting of two
323
K6nigsberg
deal
instructions
Portuguese
blow.
to
the
companiesand a
gun, and
a strong
While these reinforcements were marching south, the Portuguese launched a
probe towards Newala in mid-October. This was blocked, but a renewed attack forced
the weak defenders out and Gil had reached his objective. German security was also
threatenedby an uprising of the African population of the nearby Makonde Plateau.

Looff was now under orders first to eliminate the Portugueseand secondto
restorecolonial authority on the plateau. As soon asAbt Rothe reachedLindi, it was sent
out to attack the Portugueseat Newala. His small detachmentof 50 Germansand 300
Askaris faced nearly 800 Portuguese,who were dug in at Newala, as well as another750
holding Kivambo. Rothe reached Newala on 19 November and, after a few days in
reconnaissance,he surroundedthe garrison on 23 November, cutting off the vital water
supply. Despite their superiority of numbers,the Portugueseremained besiegedfor five
more days, when they decided a break out was essential. This took place on 28
November and a disorganised force made its way south to the Rovuma, losing four
324
large
mountain guns, sevenmachine guns and a
quantity of supplies.
While not a complete rout, the retreat was humiliating and all of the earlier
territorial gains were lost within two days. Captain Looff followed up in due courseand
eliminated all Portugueseposts north of the Rovuma; indeed, using his K6n!gsberg guns,
he shelled the border post at Nangade on I December, causing a major
panic and
abandonmentof the position. Gil himself was unwell, and lost his nerve completely,
323Lettow,Reminiscences, 165-166.
pp.

99

before
Massassi
had
2,000
Germans
attacking
that
around
concentrated
claiming
some
Newala.325 Ironically, the British had sent a warning several days beforehand,that Abt
Rothe, comprising one or two companies,was about to attack Newala, but word either
326
disregarded.
never got through or was
Having Red in disorder south of the Rovuma, rumours then abounded that the
Germanswere advancing on Palma. This was completely groundlessas Looff s troops
remainedsome 60 miles to the north, unwilling to be trapped by the now rising waters of
327
Rovuma.
Nevertheless,a British warship and a marine detachmentwere diverted
the
from their blockade duties to reassuretheir allies. After these panics, any remaining
Portuguesecredibility was destroyed. Consideringthat a well-armed and equipped force
of nearly 2,000 had been overwhelmedby less than 400 Germans,Smuts now soughtthe
War Office's permission to help defend PEA if required.328 Although a sensible
precaution,the combination of offensive operationsand the coming monsoon meant that
had
British
died
down
front.
By
the
the
ship
activity
mid-December,
on
southern
329
back
Mpotora.
to
returnedto its normal duties and,4bt Rothehad moved

324Pdlissier,Naissancede Mozambique,II,
pp. 692 -693.
325WO 95/5292,War Diary GHQ, 5 December1916,Entry PortugueseForce.
326WO 95/5292.War Diary GHQ, 7 December1916,Entry PortugueseForce. Smuts
noted "Above proves how little reliance must
be placedon fighting efficiency of GILS army as he grossly over estimatesenemy force."
'27P61issier,Naissancede Mozambique,IL
pp. 693-694.
328WO 95/5292,War Diary GHQ, 12 December1916,Appendix 17, TelegramOA 267, Smutsto War Office, II December.
329WO 95/5292,War Diary GHQ, 14 December1916,Appendix 21, Telegram F 6187, Smutsto CIGS, 14 December;Lettow,
Reminiscences,p. 166.

100

CHAPTER

REORGANISATION AND THE OPERATIONS OF JANUARY - MAY 1917

STRATEGY OF THE SECOND COALITION GOVERNMENT
As General Smuts had been making the final preparations for his offensive, greater
political events were underway in London. December had witnessed a major crisis in the
British government that had resulted in the fall of the Asquith administration.

Although

long-standing in cause, it had been precipitated by violent disagreement over conscription,

Ireland and the conduct of the war. 330 The result was a new coalition, under David Lloyd
George, who quickly instituted a smaller and more vigorous War Cabinet to direct the policy
331
of the war.

The new governmentretained its predecessor'sbroad strategic goals of retaining

Britain's statusasa greatpower togetherwith enhancedsecurityat the end of the war. More
than anything else it realisedthe need for continuedpublic supportif theseambitious aims
inconclusive
bloody
be
This
to
the
were
and
achieved.
meant achievingreal victories, not
fighting that the WesternFront seemedto offer. In practical terms, this led Lloyd George
and othersto supportoperationson the strategicperipherythat offered a chanceof success
332
hence
boost
to public morale. While EastAffica rankedsomeway below Italy,
and
a real
Palestineand Mesopotamia in order of strategic importance, it did show considerable
promiseasthe EAEF seemedon the vergeof victory.
From a political viewpoint, Smuts had conducted a rare campaign of genius
advancingmore in a day than the BEF had managedin months. Apart from capturinglarge
swathesof enemyterritory and the major centresof population,he appearedto be a daring
333
decisive
leader.
This impression had been reinforced by Smuts's own upbeat
and
330Turner, British Politics,
pp. 112-116.
331Turner, British Politics, 153; French,Lloyd George Coalition, 17.
p.
p.
332French,Lloyd GeorgeCoalition,
pp. 6-7.
333Riddell, Lord, Lord Riddell's War Dimy, 1914-1918,London: Ivor Nicholson & Watson, 1933,Entry 19 March 1917, 166.
p.
"... particularly Smuts,of whom L. O. spokevery highly. He said that Asquith held a very moderateopinion of Smuts at the time of
the SouthAfrican War. Asquith did not appreciatewhat a remarkableman he was"; MS Long, 947/547.29 March 1917, Letter Derby
to Long; MS Long, 947/602, It May 1917,Letter Buxton to Long; Roskill, Hankey, p. 388; Hughes,British Stratev in the Middle
East, p. 24. Lord Derby was so impressedwith Smutsthat he wanted to offer him a high command in Francebut was dissuadedon

101

had
hint
little
the
his
that
achievements
that
communiquds
emphasized successesand gave
beenattainedat a hugecost in diseaseand suffering.
Although his fmal offensive can only be describedas a military failure, it was not
initially viewed as such in London. On the surface,it appearedto have driven the Germans
into an increasingly tight position, trapped in the wild and undevelopedcentre of their
last
by
forces
from
In
the
this
gasp
effort represented
colony converging
all sides. reality,
incapable
ftirther
EAEF
that
of
offensive efforts
was
of an exhaustedand gravely weakened
without both substantialrest and reinforcement. However, this would not begin to emerge
until Hoskinsbeganto pick up the many pieces.
This unsatisfactory situation was worsened by Botha's and Smuts's public
334
left.
that the campaignwas more or less finished with only mopping up
pronouncements
In fact, nothing could have been further from the truth as the Gerinans had yet to be
defeatedin a defensivebattle and they had plenty of fight left in them. The damagefrom
Smuts'suntrue assertionswas to be considerable,as to politicians and senior generals,it
appearedthat EastAfrica could be removedfrom the list of theatresrequiring resourcesand
335
if
fighting
Hoskins
Unfortunately,
the
these
required
what
manpower.
were precisely
dismay
deal
be
brought
It
truly
to
to
of
anger
and
a
great
was
a successfulend. also caused
336
his
former
Africans.
South
troops and many
amongst
It seemsthat Smuts was aware that the campaign had not gone as smoothly as
publicly presented. In many ways, the summons to the Imperial Conference was a

the groundsthat he could not be sparedfrom South African service. The War Cabinet had subsequentlyoffered him commandof the
forces in Egypt and Palestinebefore Allenby, but that too had beendeclined.
334WO 33/858, TelegramsD 1, No. 1313,17 October 1917,Telegram D 280/1973, GeneralBotha to War Office, p. 355; Fendall, The
East African Force, pp. 87-88; CAB 44/9, p. 13, cites telegram Hoskins to CIGS 4 May 1917. It quotes a speechin Pretoria on 26
January 1917 in a Times article of 29 January. Smuts is also quoted as saying on 12 March 1917 "The campaign in German East
Africa may said to be over. What is delaying the absoluteend is the fact that March and April are the heavy rainy season. After April
the Germanswill have to surrenderor go into Portugueseterritory. They cannot hold out in East Africa, and the Portugueseare quite
preparedto deal with them. The GermanGovernor and his staff, the Commiander-in-Chieand other officers are still doing their best
to keepthings going, but it is merely the remnantof an army that is left, and not a formidable fighting force".
... WO 33/858, TelegramsD 1, No. 1516,20 January 1917,Telegram 28209, CIGS to Hoskins, p. 419. This telegramspecified
considerablereductionsin East Africa, largely basedon the information provided by GeneralSmuts; CO 691/19, Pike, Surg-GenWW
and Balfour, Col Andrew, Report ofMedical & Sanitary Matters in German East A#ica 1917, Nairobi: Swift Press,1918,p. 6.
Henceforth,Pike Report.
316MS Long, 947/545,22 April 1917,Letter Sir CharlesCrewe to Long; WO 95/5329, War Diary Norforce, Entry 28 February 1917;
Downes, With the Nigerians, p. 267.

102

337
it
himself
felt
Smuts
to
from
East
Africa.
honourable
necessary
convenientand
escape
draft a pressstatement,justifying his recall in terms of a military success:
"... The military situation in EastAfrica, is fortunately, such as to make a changein
indeed
the stepsthat
simple,
and
comparatively
command,and somereorganisation
demand
for
Smuts'
General
in
the
sudden
are now contemplated consequenceof
in
been
have
taken
those
any case very
which would
services elsewhere are
desertion
forces
in
His
[enemy]
are much
of
casualties
and
consequence
shortly...
reducedin strengtharid morale"338
BRITISH REORGANISATION

The failure of the late offensive left Hoskins in a very difficult position, both on the
in
had
forced
in
dealing
While
London.
the
operations,
a
pause
rains
ground and
with
increased
for
the pressureon the medical
they
also
giving a respite
rest and recuperation,
deteriorated
health
The
transport
troops
the
even more rapidly, while the
and
systems.
of
transportnetwork dissolved into mud or was submergedunder water. Drastic action was
first
but
The
distinct,
disaster,
he
faced
to
three
and
problems.
related
required avoid a
as
in
forward
to
the
troops
the
to
the
areas
the
of
most pressingwas
numbers
need reduce
levels that could be physically sustainedwhile concurrently stemming the rising losses
causedby diseaseand ill-health. In short,he had to maintain his army in being.
The secondchallengewas then to rest and reorganisethe entire force so that it was
included
This
dry
in
to
the
the
a complete
strong enough resume
season.
next
offensive
overhaul of the supply, transport and medical servicestogether with full replacementof
previous losses. Hoskins also recognised that far-reaching changes in the system of

$37
WO33/858,Telegrams
D 1,No. 1313,17October1916,Telegram280/1973,GeneralBothato WarOffice,p. 355. ".. -The
has
thecampaign
though,aswasexpected,
progress
of theEastAfricancampaign
to its inevitableendis continuingsatisfactorily
arrivedata stageatwhichtimeis necessary
to effectcertainregroupingandredistribution
of forces"; MS Smuts,Box 100,17
January1917,LetterBuxtonto Mrs Smuts;"But to dothis [hand-over
command]voluntarilyandwithoutanyobviousmomentor
reasonwasnoteasy,norperhaps
expedientandmightpossiblyhavegivenriseto some(quiteunjustified)criticism. But this 'call' to
Englanddisposes
absolutely
andcomplimentarily
of anypossiblecriticismsonthatscore".
338
WO33/858,Telegrams
D 1,No. 1500,11January1917,TelegramX 1182,Smutsto Secretary
of Statefor War,p. 412. In his
for Pressin connection
followingdraftof statement
privatepreamble
to thestatementSmutsalsostates"I submitfor consideration
herejustify thechangein command.
"
with myretirement,
asit shouldbemadeclearthatmilitaryresultswhichhasbeenaccomplished

103

commandand control, both tactical and administrative, were essential as most of the
previousyear's failures were linked to deficienciesin this area."'
The third, and from London's point of view the most important, question was the
needto deviseand put into place a strategythat would wrap up the campaignas quickly as
possiblein 1917. This called for a decisive and well-executed plan based on sufficient
numbersof fit and well suppliedtroops. Hoskins understoodthis very clearly, but it could
only be achievedoncethe first two objectiveshad beenachievedand neither was attainable
in the shortterm.
The situationwas not helpedby one of the wettest seasonsknown in EastAfrica for
many years. If Decemberand early Januaryhad been bad, the seasonalmonsoon would
makethings unimaginablyworse. Roadseverywhereceasedto exist and porterswere often
the only alternative,sometimeshaving to wade through waist-deepwater for miles. As the
wet seasonprogressed,the water rose everywherein the Rufiji-Ulanga-Ruaha basin. For
example,a level gaugehad been establishedon the normal dry seasonhigh water mark at
Kibambaweon the River Ruf1ji. By I April, the river was 147 inches (12.3 feet) over the
mark, nine days later the excesshad risen to 175 inches (14.4 feet), with the peak of 201
inches (16.8 feet) being reached on 14 April. Thereafter, the level stayed around 180
inches(15 feet) abovethe gaugeuntil the end of the month, when it again peakedat 201
inchesbefore diminishing in May. Consideringthat the valleys were generally broad and
340
the
shallow, volume of extra water was simply enormous.
Further west, the Dodoma-Iringa line of communicationscrossedthe Great Ruaha
River by an easyford in dry weather. During the wet season,there was not just a flooded
river to cross but a swamp six feet deep and six miles beyond either bank. Elsewhere
Kilwa and the lower Rufiji suffered heavily from flooding and raging torrents. Only in
the Iringa highlands, so far above the river basins,was movement possible, but even then
it was made very difficult by the heavy rain and the loss of most roads. Everywhere,
sicknesslevels escalatedrapidly, with malaria being particularly acute for Europeans
3'9WO 95/5292,War Diary GHQ, 24 February 1917,Entry P Division.

104

Animals
dysentery
from
Africans
could only
pneurnonia.
and
while
suffered more
341
death.
before
In
badly
deteriorated
the
few
their
survive a
condition
weeks and
for
from
the
there
either side.
misery
circumstances,
was no escape
THE REDUCTION

IN FORWARD

FORCES

The problem of reducing the forward forces to a manageable level was Hoskins's
first priority. Once 2ndDivision had ground to a halt in early January owing to lack of food
disbandment
ill
Central
Railway
health,
it
had
been
to
the
commenced
and
and
withdrawn
342
13
January.
In its place was a small two-battalion column of Indians and KAR under
on
Colonel Taylor. 343

344
1st
The thinning out of the forward areas began in earnest on 14 February.
Division had pulled its two brigadesback, with the 2ndEA Brigade ordered from Kitambi
back to Kibata by 19 February and 3rd EA Brigade also drawn back to the NgarambiNamatewaarea.345 The 0 EA Brigade, however, retainedthe unpleasanttask of holding
the Ruf1ji line until the now completeNigerian Brigade could move forward to relieve it.
The commander-in-chiefstill retained hope that the main supply route between Mikesse
andthe Rufiji would remain in reasonablecondition throughoutthe rainy season,otherwise
the forward positions would be untenable. In the west, the planned concentration of
Norforce around Lupembe and Songeawas taking place, but was soon to be disruptedby
346
Wintgens.
the appearanceofAbt
The Germanswere also forced to make a number of changesto their dispositions.
On the middle Rufiji, the forward troops were thinned out considerably, leaving only
major concentrationsoppositethe Nigerian Brigade at Mkindu and at Utenge. Even then,

conditionswere dangerous;the British took the surrenderof a hospitalcontaining200

30 WO 95/ 5293,War Diary GHQ, Entries 1-30 April 1917.
341HoskinsDespatch,p. 159.
'2 WO 95/5292,War Diary GHQ, 2 January 1917,Appendix 3, Telegram P71, GOC Div to BGGS, 2 January; War Diary GHQ, 13
January1917,Entry 2"dDivision.
343WO 95/5292,War Diary GHQ, 26 February 1917,Entry Nyasaland Force. 2ndDivision had left only the 17,h Infantry
and the
I"A"' KAR at Iringa under the commandof Colonel Taylor.
30 WO 95/5292, War Diary GHQ, 14 February 1917, Appendix
of Communiques, Telegram OA 261, Hoskins to CIGS, 14 February.
345CAB 44/9,
p. 17.

105

soldiers that had been cut off by the rising waters and were threatened with either
forced
River
Ulanga
drowning.
In
the
an evacuationof
the
the
rise of
west,
starvationor
Sickness there was tremendous; Abt
347
for
duty.
On
Braunschweigreportedthat over 75 per cent of the Europeanswere unfit

the west bank and a climb to drier ground.

the lower Rufiji, life was equally miserable.

Although these reductions prevented starvation, conditions remained deplorable.

Rationsremained grossly insufficient owing to a severelack of transport. For example,
the Nigerian Brigade was put on half rations on 30 January as all of their porters were
brigade
This
the
keep
lines
virtually
to
the
rendered
of communication open.
required
immobile and dependenton the vulnerablelifeline to the north. Remaining in the areainto
This
by
its
duly
the
February,
was
of
rations.
paucity
early
mounted,aggravated
sick rate
for
had
500
28
Brigade
in
from
disease;
the
men unfit
officers and
reflected the casualties
duty, or approximately22 per cent of the officers and 21 per cent of the African ranks that
had startedat the end of December."' These sufferings were not unique to the Nigerian
Brigade, as others endured similar shortages. The Gold Coast Regiment, serving in I"
Division and stationedsouth of Utete, had the misfortune not only to be on half rations for
had
deal
that
but
the
January-February,
arrived
the
supplies
of
most of
also a great
period
Many
for
human
of the soldiers were very emaciated
rotted and were unfit
consumption.
349
had
80
for
to be hospitalised starvation.
and
Despitea brief improvementin the food situation in early March, it was decidedto
left
This
for
I"
EA
Brigade
the
only the
withdraw
rest and recuperation.
remainder of
Nigerian Brigade along the middle Rufiji position, and by 17 March it held a line of nearly
50 miles, running from Nyangandu to Nyakisiku. However, the Rufiji continued to rise
andthe supply situationworsenedappreciably. From 28 March to 1 May all ranks were on
half rations, supplementedonly by shooting of game. However, even this sourcebecame
depletedthrough flooding and constantdepredationsby hungry soldiers. To cap things off,
346WO 9515292,War Diary GHQ, 22 February 1917,Appendix of Communiques,Telegram G218, Hoskins to CIGS, 22 February.
347Boell, Die Operationen, 253.
p.
348CAB 45/19, Nigeria RegimentRecord,
pp. 3 and 18. The brigadestartedwith 125British Officers, 70 British NCOs and2402 African
NCOs andprivates,WO 106/273,History of3 KAR, p. 55.3 KAR had beenon half rationsfrom leaving Mbuyuni on 22 May until 18
September1916.
349Clifford, Vie Gold Coast Regiment, 68-69.
pp.

106

the flying bridge was swept away and the entire force was cut off for three complete
days.350

Finally, despiteall human efforts, on 8 April the Nigerian Brigade was compelled
to sendhalf its troops back to the railway line owing to lack of food and clothing. General
Cunliffe's words underlinedthe needfor more of everything:
"... lack of supplies in addition to nullifying all possible military initiative is
gradually sapping stamina of officers and men who by transfer to healthier
051

for
conditionswould return recuperated eventualadvance.

By mid-April, the food situation was getting worse even with half the brigadeback
to Morogoro. The returning units were so weakenedwith hunger that during the 10 day
marchto Mikesse,large numbersfell out, incapableof completing the trek. The road was a
loosedescriptionas in many placesit was waist deepin water, and littered with the bodies
It
down
dead
broken
was a
carts.
vehicles and
of
animals and even porters as well as
depressingsight - eventhe patrols had to be conductedin canoes. Finally, in mid-May the
rains beganto abateand the supply situation improved slowly, although active operations
352
fit
for
useby MT.
would not be possibleuntil the MikesseRoadhad beenmade
REST AND REORGANISATION

The worn-out and weakened forward troops required much more than simple
withdrawal to the Central Railway, itself far from healthy or comfortable. They needed
completerest in a dry climate with a proper diet and replacementof their raggeduniforms.
This systemhad already been instigated in early 1916, when a British battalion had been
sentto South Africa in an attemptto restoreits health and fighting efficiency. One of the
assumptionsbehind the evacuationof most of the South Africans was that a smaller, fitter
contingentwould return after appropriaterest and recovery. To this end, GeneralHoskins
3" CAB 45/19, Nigeria RegimentRecord, 19.
p.
351WO 95/5293,War Diary GHQ, 8 April 1917,Appendix 66, TelegramN 715, GeneralCunliffe to BGGS, 8 April.
352CAB 45/19, Nigeria RegimentRecord, 20-22.
p.

107

for
Another
in
back
battalions
the
SAI
time
offensive.
next
three
was counting on seeing
British unit was then sent to South Africa for recovery, while the four longest serving
Indianunits were rotatedto India in exchangefor fresh units as shipping becameavailable.
Obtaining and maintaining sufficient fit soldiers was always one of the biggest
in
infantry
dramatic
by
facing
EAEF.
Matters
the
the
reduction
were worsened
problems
By
December,
1916.
October
the
began
the
the
that
end
of
of
repatriations
manpower
with
loss of nearly 12,000 South Africans had only been partially offset by the arrival of the
Nigerian Brigade, some 4,500 strong. On the other hand, the indigenous King's African
Rifles had proven their effectivenessin battle as well their resistanceto disease,but with
decisive
be
force
to
too
battalions,
brand
at
they
two
them
small
a
were
only eight
of
new,
the operationallevel. Although an expansionprogrammehad been agreedsome months
before,the recruiting, equipping and training of thesenew units would take many months.
Hoskinstook a particular interestin the KAR, but his efforts were hamperedby the lack of
difficult
British
it
to
trained
officers and
get
suitably
particularly
was
personnel,while
353
NCOs who could actually speakthe local languages.
The CIGS's unexpectedorder to withdraw eight infantry battalions on top of the
departingSouthAfricans, as well as five artillery batteries,three armouredcar batteries,the
354
blow.
in
RFC squadronand various smaller units to join the forces Egypt was a major
The loss of these units seriously weakened Hoskins's offensive capability and would
impedeplans to wind up the campaignin 1917. Hoskins replied the sameday, citing that
the alreadyplannedloss of 3,500 infantry from 2ndDivision plus the 4,000 extrasrequired
from his other formationswould reducehis numbersto below the bare minimum neededto
conductoperations. In its stead,he offered to releasehis South African field batteriesin
return for the retention of two Indian infantry battalions, the net decreasebeing reduced
from 7,500 to 6,100.355Robertsonwas quickly won over and on 25 Januaryagreedto let
Hoskinskeep as many troops as he consideredessentialto fight the campaign. Indeed, it
353Hoskins,Despatch,p 163;Moyse-BartlM King's African Rifles,p. 701. On I July 1916,the regimentwas over 8,000strong;six months
lateron I January1917,it numberedover 15,000.
354WO 33/858, TelegramsD 1, No. 1516,20 January 1917,Telegram28209, CIGS to Hoskins, p. 419 gives the reductions; WO
95/5292,War Diary GHQ, 25 January 1917,Appendix 46a, Telegram OA 846, Hoskins to GeneralEwart, 22 January,gives
Hoskins's response.
355WO 95/5292,War Diary GHQ, 25 January1917,Appendix 46b, TelegramOA 847, Hoskins to CIGS, 22 January,

108

in
Middle
for
the
in
than
reinforcements
an urgent need
wasthe reduction shipping rather
Eastthat inspiredthe transfers.
It emergedthat one of the reasonsfor the scaleof the reductions was that London
believedthat the Germanswere considerablyweaker than did Hoskins. The War Office
lacking
6,000
600
had
Lettow
either
and
askaris
that
whites
of
a maximum
estimated
von
356
by
Hoskins's
This
bases
assertionthat the
countered
was
organised
or communications.
Germansactually had about 1,500 whites, of whom 1,150 were considered effective,
togetherwith some7,300 askaris,making a total of 8,450- the differencewas 1850men or
28 per cent more than the War Office estimate. Ammunition was plentiful, although food
However,
Utete.
Mohoro
the
loss
in
following
the
and
of
was short supply especially
357

districts of Lindi, Rovuma and Mahenge all promised to yield good levels of food.

Although London accepted the larger figure and the misunderstanding seems to have arisen
from Smuts's own over-optimistic reports of only 6-7000 remaining effectives, it was not
358
an auspicious start.

Robertsoncalculatedthat with the re-entry of the Belgians into the campaign,the

in
South
Indian
Army
the
units
recuperating
of
return
rotation of worn-out
units, and
Africa, Hoskins would have a significant advantageover the Germans,with some 50,000
This was followed several days later by confirmation of
troops to counter 8
'000.359
renewedSouth African support in the form of two newly raised infantry battalions, plus
360
These
for
MT
the
were welcome additions,
units.
replacements
and other specialist
disputed
Hoskins
Furthermore,
the
be
the
althoughmost of
soldiers would
new recruits.
figures and placed his "effective" infantry currently at 16,000, with the rotation of tired
units and the new South Africans adding another 3,700 and the Belgians a further 3,000.
He calculatedhis effectives at being 23,000,of whom at least 20 per cent would be unable

356WO 95/5292,War Diary GHQ, 25 January 1917,Appendix 46a. Telegram OA 846, Hoskins to GeneralEwart, 22 January.
357WO 95/5292,War Diary GHQ, 27 January 1917,Appendix 54, Telegram OA 936, Genstaff to CIGS, 27 January.
338Meinertzhagen,Army Diary,
p. 193, diary entry for 20 July 1916.As far back as July 1916,Meinertzhagenwas noting that Smuts
believedthe Germansto be weakerthan his Intelligence estimatesand that the Askaris had lost their morale. This trait appearsto have
colouredhis reportsto CIGS.
359WO 95/5293,War Diary GHQ, 19 April 1917,Appendix A120, Telegram 32999, CIGS to Hoskins, IS April.
360WO 95/5293,War Diary GHQ, 28 April 1917,Appendix A200, Telegram H 3875, CSO Pretoria to Hoskins, 27 April.

109

to participateowing to sickness. The new KAR battalionspromisedto bring another2,500

361
low
for
many monthsto come.
men,but the quality of theseunits would be
SUPPLY AND TRANSPORT

Smuts's failure adequately to consider supply and transport when making his
operationalplans had led to many of the problemsthat Hoskins was now trying to resolve.
The first and most obvious weakness had been the failure to brief or consult his
administrativestaff; as a divisional commander,Hoskins had seen and experiencedthe
limitations of GHQ's planning. However, on his assumingthe chief command, formal
evidenceof the problem was supplied in a memorandumby the Director of Supply and
Transportto the principal administrativeofficer. It was an indirect, but withering, critique
of Smuts'smethodsas it depictedthe gapbetweenthe staff branches:

"Withoutanydesireto critisize[sic] themethodsof directionof the [campaign],the

first difficulty I asD. of T. hadto contendwith wasthe absence
of informationasto
the plans of operations,which shouldnormally in my opinion be conveyedin
in
time
to
sufficient
allow of preparationof reserves animal and mechanical
transport.
Had this informationbeenavailableto me, someopportunitywould have
beenaffordedto point out mostof the salienttransportdifficultiesregardingthese
plansandto suggestto you meansof meetingsuchdifficultieswherethey couldbe
overcome.
Speakinggenerallyof the conductof the campaign,I must with due
deferenceinsist that from a transportpoint of view my resourceswere overstrained
from the fact that long advanceswere carried out without halts at suitable intervals
in order to allow transport to catch with the troops and enable supplies to be
accumulatedbefore the next move forward [author's footnote: Two halts were

361WO 95/5293,War Diary GHQ, 27 April 1917,Appendix B21, Telegram G 165, Hoskins to CIGS, 27 April.

110

Railway,
Central
Same
the
which never were
on
and one
promised, one at
given]. 9362

The reasonsfor short feeding were due to beginning the advance without any
transportreserves,the continual and unexpectedchangesin forward troop strengths,new
long
lead-time
for
integral
the
transport,
replacement
and
units arriving without any
vehicles. The note endedwith a plaintive plea:
"... Finally that an opportunity be given to all Administrative Servicesto discussthe
063
General
Staff.
proposalsof the
For a commander of the right temperament and training, the problem of coordination was relatively easy to solve. However, the resolution of supply and transport
difficulties posed much more of a challenge. Food, ammunition and equipment could
generallybe shippedto the seaports of Dar-es-Salaam,Mombasaand Kilwa without major
problems;it was moving them to the forward troops in a timely and efficient mannerthat
hadprovento be sucha nightmare.

harbourof
Portswerea major bottleneckandit took time to clearDar-es-Salaarn
the sunkenfloating dock and scuttled ships. In the meantime,an engineercompanywas
employed full-time on the building of wharves and installing steam cranes for the
364
Furthermore, as the expeditionary force's main base, it
unloading of ships there.
required the construction of extensivevehicle workshops and storage facilities. Further
south,Kilwa had an excellent natural anchorage,but ships were unable to come close to
shoreand unloading had to be carried out with lighters. It was almost totally undeveloped
when the British occupied it in Septemberand a great deal of further developmentwas
requiredthere. Finally, Lindi which was destinedto take on a key role in 1917,had only
362WO 95/5293,War Diary GHQ, April 1917. The War Diary
enclosed"Printed Notes from War Diaries, Part CCLXXXV, EAEF
Force"B", GeneralStaff, Army Headquarters,India, April 1917,Confidential Print ofKey Extractsfor Army Departments. See
Director Supply and Transport, EastAfrican Force, 18 to 31 January 1917,Appendix A to DA&QMG, General Headquarters,No. ST
34 dated 18 January1917,p. 15. Although not addressedto Hoskins personally,the existenceof a printed note by Army Headquarters
India, and subsequentlycirculated to all staff branchesmeansthat it would have almost certainly come to his attention. Given the
embarrassingnatureof ale contents,it seemsunlikely to have beenreprinted without his permission.
'63WO 95/5292,War Diary GHQ, April 1917,Confidential Print
ofKey Extractsfor Army Departments,p. 16.
3" Woodhouse,Capt H L, "Notes
on Railway Work in EastAfrica, 1914 - 1918", TheRoyal EngineersJournal, XXK (1923), pp.

III

in
facilities
basic
dock
the most
which would require significant enhancement the coming
troops
technical
trained
by
Hoskins
faced
The
of
shortage
a
were
months.
main problems
to
between
the
materials
and
machinery
the
time
necessary
and
ordering and receiving
EastAfrica.
The most efficient means of moving personnel and stores forward remained the
Central Railway. German demolitions had been makeshift and were not particularly
difficult to overcome,but they did take time to repair. Despite Smuts's claims that it would
take only a couple of weeks to restore operations,it took the in-theatre railway battalion
two monthsof flat-out work just to restorethe line to a basic level of function betweenthe
locomotives
destruction
Tabora.
The
and rolling stock was another major
of
coast and
handicapthat was solvedby resourcefulnessand improvisation. Engineersfirst openedthe
tracksto hand-operatedtrolleys followed by 30 hundredweightlorries mountedon railway
6
October
in
Dodoma
first
However,
and
tractor
only
on
the
arriving
wheels.
such motor
build
let
daily
fraction
up
the
alone
this service could only meet a
requirements,
of
drawing
locomotives
full
before
December
It
service of
a
reserves. was not until early
railway freight carscould be restored.
Hoskinshad attemptedto use the Rufiji to move suppliesand the navy conducteda
365
from
in
innovative
late
to
January.
An
supplies
ships to motor
attempt off-load
survey
boatswas defeatedby the strengthof the flooding river's current which simply outmatched
the power of the small engines. This meant that the maintenanceof troops in the area
366
would haveto rely on lessefficient means.
The biggestchallengewas to createan effective transportsystemforward of the
feasible
(AT)
The
transport
option
traditional
system
was
not
a
ports andrailheads.
animal
had
been
However,
draught
horses
the
already
proven.
as mortality of
animals
and other
despitethesemajor limitations, it still representeda large proportion of the overall effort, if
only becauseof lack of alternatives. The best solution was clearly the developingsystem
42-44.
'63WO 95/5292,War Diary GHQ, 22 January 1917,Appendix 38, Telegram OA 835, Hoskins to CIGS, 21 January.
'" WO 95/5292,War Diary GHQ, 27 January1917,Appendix 52, Telegram OA 947, Hoskins to CIGS, 27 January,

112

of MT; in one day a single Ford car of 300 lbs cargo capacitycould match the efforts of
300carriers. However,the almost completelack of roadsin the southernpart of the colony
bad
to
insufficient
their
weather
vulnerability
coupledwith
numbersof vehiclesand
This
form
from
becoming
MT
transport.
the
meantthat considerable
of
precluded
main
in
difficult
be
terrain or
human
transport
areas
of
relianceon
would essentialparticularly
whereroadshad not yet beenconstructed.
Hoskins and his staff put considerableeffort into calculating the minimum levels of
transport resourcesrequired to support an offensive. The numbers were huge; plans
worked out in February called for 160,000 carriers together with provision of 16,000
had
By
the
made
conditions
poor
working
replacementsper month.
neglectand
mid-1916,
service in the Carrier Corps extremely unattractive regardlessof the rates of pay, and
recruiting was becomingvery difficult. Even with coercion and the subsequentpassingof
for
Africans
in
Compulsory
in
Service
1917,
Act
March
the
massconscription of
a
which
like
levels
impossible
labour
it
the
to
reach
anything
carrier or
servicewas authorised, was
for
African
Special
to
the
more
required.
various colonial governments
appealswere sent
"'
insufficient.
but
healthy
labour
the
manpower,
was
pool of
The Commander-in-Chief launched his first plea for help on 6 February with
368

requeststo the Governorsof Ugandaand Nigeria for 5,000 additional carrierseach.

The

Governorof British East Africa was also approachedand demurredon the needto sustain
369
Both territories were heavily pressed,but, caught between the need to
recruiting.
had
by
1917
July
they
the
to
the
support
war effort and
anxiety prevent a popular revolt,
only supplied67,799 and 10,934men respectively. Even West Africa was approachedfor
370
5,000
The Belgians and Portuguese
some
carriers, which were eventually provided.
cameunder considerablepressureto open up their colonies to British recruiters,but these
efforts cameto little.
367Hodges,77ieCarrier Corps,
pp. 99-102; Hodges,"Military Labour in East Africa7, in Page,Melvin (ed), Africa and the First
World War, London: Macmillan Press,1987,p. 139.
36'WO 95/5292,War Diary GHQ, 6 February 1917,Appendix A, TelegramOA 109, Hoskins to Genstaff covering OA 56 Hoskins to
GovernorUganda,5 February; WO 95/5292, War Diary GHQ, 6 February 1917,Appendix B, Telegram OA 110, Hoskins to Genstaff
covering OA 54 Hoskins to GovernorNigeria, 5 February.
369WO 95/5292,War Diary GHQ, 9 February 1917,Appendix D. Telegram Governor,Nairobi to Hoskins, 9 Februarycovering
TelegramGovernor,Nigeria to Hoskins.
3' Killingray, David
and Matthews, James, "Beastsof Burden: British West African Carriers in the First World War", Canadian

113

Carriers, although important, were far from the only demand on civil manpower.
The existing lines of communicationhad to be maintained,new roads had to be cut and
improved,while defensivepositions and camps had to be built along their length. Ports
and basesneededlabourers,while many worked as personal servantsto the military and
371
large
had
In
that
numbers of soldiers
already provided
civil adminstrations.
a region
followed by regular trawls for labourersand carriers,first by the Germansand now by the
British, the pool of manpowerhad begunto run dry.
Realising the urgency of his requirements,General Hoskins bombardedthe War
Office with requests for more materiel and specialist manpower. Despite its lack of
in
deficiencies
best
transport,
importance,
his
did
the
Robertson
to
remedy
strategic
lorries
light
from
British
484
By
April,
total
sources
of
armamentsand personnel.
a
early
togetherwith 607 drivers and artificers were underway at sea while a fin-ther 563 men
400
300
berths.
drafts
Included
these
signals
personnel,
were an additional
awaited
with
372
Lewis guns,twelve of the new 2.75" mountain gun, and twenty Stokesmortars. Perhaps
tiny by Western Front standards,these shipments added substantially to East Africa's
technicalresources.

MEDICINE
The medical serviceshad been in a poor stateof organisationprior to the advance
south from Morogoro. Despite the overwhelming importance of health in the campaign,
the General Staff failed to appreciate the need to treat and evacuate casualties.
Unfortunately,the principal defectsof insufficient transport, lack of consideredplanning,
373
and non-existentbriefings to the staffs and units concernedremaineduncorrected. One
Journal ofAfrican Studies,XIII, No. 1, (1979), pp. 8-11. By 1918,9,391hadbeensupplied with the majority employed as carriers.
371CO 533/216,EastAfrica Protectorate,Volume 11,1919, Letter Acting Governor to Secretaryof State for the Colonies, 31
December1919. This coversReport by Lt Col 0F WatkinsCUG, CBE, DSO, Director of Military Labour to the BEA Expeditionary
Force: On theperiod August 4*, 1914 to September15", 1919, pp. 451469. Henceforth, WatkinsReport.
3'2WO 95/5293,War Diary GHQ, 9 April 1917,Appendix A78, Telegram 32324, CIGS to Hoskins. Quoteshis 31195 of 18 March
and 31391 of 31 March 1917.
373Macpherson,Maj Gen Sir WG
and Mitchell, Maj T J, History ofthe Great War Basedon Official Documents- Medical Services
General History Volume4. Medical ServicesDuring the Operations on the Gallipoli Peninsula; in Macedonia; in Mesopotamiaand
North-WestPersia; In East Africa; in the Aden Protectorate and in North Russia, London: HMSO, 1924,pp. 452-453. Henceforth,
Medical ServicesVolume4.

114

in
late
1916.
South
Africans
12,000
the
the
grounds
examplewas
on
medical
evacuationof
Smutsmade the decision suddenly on 25 September,but before an evacuationplan had
beendrawn up or the various staff branchesconsulted. The result was a muddle and it took
374
be
for
the organisationof routes,transportand campsto organised.
sometime
The medical services were severely overstretched. Already lacking sufficient
transport for true tactical mobility, the rains halted the movement of the motor ambulance
convoys (MAC).

This meant that the sick and wounded had to be evacuated by carriers

which was naturally much slower. It also caused casualties to remain away from treatment
for longer, placing more work on the remaining fit men, and sapped the vitality of the
carriers themselves. Furthermore, in those areas where MT was still feasible, sickness
amongst the drivers, particularly the newly-arrived, was a major problem. Seldom could
all usable vehicles be driven owing to sickness. The case of a section of a field ambulance
at Kilossa is instructive: equipped for a maximum of 50 patients, it had to hold, treat and
375
feed300 for
in November 1916.

a considerabletime

Reorganisationof the medical serviceswas brought about with equipment being

made easier to carry and stretcher-bearersadded to establishments. Two extra field
ambulanceswere brought out from India, and the systemof evacuationand treatmentwas
rationalised. Individual columnswere now allocatedtheir own field ambulancesectionsas
well aselementsof the bearercompaniesand sanitarysections. Forward clearing hospitals
were setup on eachline of advancelinked by rest stationsand supportedby a forward field
depotof medical supplies. Behind them were the stationaryhospitalswhich then fed back
patientsto the generalhospitals in Dar-es-Salaam. Greateruse of vehicles in the MACs
was made and they delivered the casualtiesback to the railway where special ambulance
trains made the final journey to Dar-es-Salaam. The system of convalescentcampswas
extendedfurther forward; previously clusteredaroundNairobi, new builds were erectedat
Morogoro and Dodoma. In the caseof those consideredas being too seriousto recover in

74Macphersonand Mitchell, Medical Services Volume4,

pp. 458459.
75Macpherson
and Mitchell, Medical Services Volume4, p. 468.

115

EastAfrica, three hospital ships were employedin the transportof patientsto India, Egypt
376
South
Africa.
and
The carriers, on whom so much depended,received much more attention than
previously. Numbering over 150,000strong,they were allocatedthe use of some28 mixed
377
hospitals.
five
hospitals
However,these
reststations,
and sevencarrier
casualtyclearing
changes,overdueasthey were, were too late to easethe immediateproblemscausedby the
tropical monsoon. In the period 8 Januaryto 5 May, over 38,000 were admittedto hospital
with an admissionrate of 206 per thousandper month. Of these,some23,000 were due to
malaria, making an admission rate of 126 per thousand,while dysentery accountedfor
2,800 casesand 7.5 per cent respectively. 'Mese two diseasesrepresentedthe bulk of
hospitaladmissions,with malaria accountingfor over 60 per cent and dysenteryfor 7.5 per
cent of all admissions. Equally important they respectivelycaused26 per cent and 23 per
cent of all hospital deaths. Kilwa had the unhappy record for sicknesswith admissions
reaching41.3 per centmonthly.
The levels of sicknesswere not just a temporaryconcernas many of the caseswere
permanentlylost to the campaign. Of the 38,333 soldiers admitted during this four month
period, some 10,436were invalided overseas,43 were dischargedfrom military service,
and 642 died - this representeda permanentloss rate of 5.9 per cent. The carriers also
sufferedheavily too, as 33,169were admittedto hospital,with 4,168 dying comparedto the
642 soldierswho succumbed. The sick rate reached7.1 per cent per month, with malaria
accountingfor 9,629 or 2.0 per cent and pneumonia2,342 or 0.5 per cent monthly. Malaria
represented29 per cent, dysentery 18 per cent and pneumonia.7 per cent of all admissions.
Permanentlosses were also high, with 15,845 being invalided out of service, 4,435
desertingand 4,168 dying.378 Sadly, matters did not improve in the subsequentperiod,
with a peakof 7,500troops and 2,000 followers in hospital by the first week of December.

376Macphersonand Mitchell, Medical Services Volume4,

pp. 475-477.
377Macphersonand Mitchell, Medical Services Volume4, 477.
p.
378Macpherson
and Mitchell, Medical Services Volume4, pp. 478-479

116

COMMAND AND CONTROL

The previous year had exposed a number of weaknessesin the system of

operational command and control. While the divisional structure was well proven
elsewhere,it required the constituentbrigadesto operaterelatively closely to one another.
This enabledthe commanderto supporthis fighting formations with artillery and engineers
as required,while divisional supply and transport units maintainedthe link betweenthem
and the lines of communication troops. The system ran best with the combination of a
centralisedrear areasupportinga number of similarly organisedformations, all backedup
by an efficient rail and road network. To be really effective, it also required a large and
well-trained administrativestaff under the control of a single officer. Unfortunately, none
of theseconditionswas presentin EastAfrica.
Reorganisationhad left only the I" Division, plus a number of independent
formations directly under the control of the commander-in-chief. By early 1917 it was
clear that the divisional system was not working well in the difficult conditions of the
African bush. Quite simply, the system could not cope with the problems of supplying
widely spreadbrigades at great distancesfrom their railheads together with insufficient
transport. The answerwas the expandeduse of the independentcolumn. Thiswasnothing
new, as it had been the mainstay of past colonial operations,and had been used most
successfullyby Norforce. In this system,eachcolumn was basedon a varying number of
fighting troops,usually about 3-4 battalionsstrong,dependingon the task.
Transport units had been designedto support a brigade of fixed composition and
could not easily cope with increases. The solution was centralise all resourcesand then
allocate them to the columns as necessary. By pooling junior commandersas well as
vehicles,the systemcould becomemuch more flexible. A force headquarters,its size also
dependenton the mission at hand, supporteda varying number of columns and maintained
the lines of communications. The system was more flexible and could be adjusted
accordingto the supply or tactical situation. While it could not resolve the problems of
acquiring sufficient transport or moving supplies over difficult terrain, it did give tactical
117

it
the
that
to
It
freedom.
number
ensure
easier
also
made
commandersgreateroperational
haunting
local
the
the
forward
to
supply capability;
were equivalent
of troops sent
forgotten.
However,
Rufiji
littered
the
divisions
were not
around
memoriesof semi-starved
it was not possible to overhaul the entire command apparatus at once and the decision to
for
implemented
fully
be
late
February,
in
several
taken
would not
move to columns,
379
months to come.

PLANNING

FOR THE DRY SEASON OFFENSIVE

IN 1917

The halt imposed by the rains also provided an opportunity to formulate the
The
late
June.
before
begin
1917
dry
for
the
which would not
seasonof
operationalplans
difficulties of trying to advanceand sustainoperationsacrossthe Mgeta and Rufiji rivers
had already been amply demonstrated. Fighting at the end of elongated lines of
in
British
transport
jungle
in
while
the
weakness
magnified
communication primeval
base
by
Kilwa
However,
in
few
of operations,
as
a
using
tangible advantages return.
giving
better
to
British
advantagewhile also
the
could utilize their maritime supremacy much
by
the
lines.
Movement
rapid and
possible
also
made
the
ship
shortening overland supply
largescaleredeploymentof troops,if only along the littoral.
Thesefactors,coupledwith the realisationthat von Lettow was likely to retreatinto
future
base
Lindi
led
German
East
Hoskins
Africa,
to
of operations.
as
a
select
southern
However, the landings in Septemberhad left only the immediate area of the town and
high
dominating
to
the
holding
in
hands,
the
British
ground
the
enemy
anchorage
with
it
be
base
take
development
Before
would
the
place,
could
port and
of
west.
any serious
forward
to
from
direct
secure
establish
to
observation and
necessary clear the area
General
Brigadier
Hoskins
To
selected
this
task
out,
positions.
was carried
vital
ensure
O'Grady, an aggressiveand energeticofficer, to assumethe commandthere. He and his
brigadewere to reinforce the small garrison with the initial task of removing the threat to
Lindi andthen preparingfor a subsequentadvance.380

379WO 95/5292,War Diary GHQ, 24 February 1917,Entry P Dtvision, Telegram Hoskins to Hannyngton,24 February.

118

Hoskins set out his thoughtsto the CIGS at the beginning of March. He envisaged
into
heartland
four
from
the
of
converging
a
pronged advance
widely separatedpositions
the Germancolony. In the north, a column would push south from the Rufiji River towards
Madabaand Luwegu, supportedin the west by anothercolumn advancing from Iringa to
Mahenge;from the coast, a third column would move westwards from Kilwa towards
Liwale and a fourth would move north-westwardsfrom Lindi in an attempt to cut off any
escapeto the south. Norforce was to move east,blocking any German attemptsto move
into either Rhodesia or Nyasaland. Fundamentalto the successof this plan was the
drivers,
signallers
and
as well as many
provision of substantialreplacements,particularly
moreporters. Hoskins stated:
"the speedy provision of light lorries and porters is the crux of the whole question;
without these operations cannot commence and unless numbers are adequate and
increased progressing as the advance is made hostilities cannot be brought to a
381
"
conclusion.

The difficult question of numbersalso arose. Of the some 24,000 infantry in the
main command,and deducting 1,200 for the lines of communications,only 8,600 were
assessed
as being efficient for operations. Northey had some 4000 rifles, of whom about
1,900could be placed in the field making a total striking force of 10,500;provided that the
promisedthree South African infantry battalions,due to return after rest at home, arrived,
Hoskinsbelievedhe had sufficient troops to make his plan work. However, he emphasized
the needfor frequentdrafts of reinforcementsfor all troops,especiallyhis West Africans.
The commander-in-chief also informed the War Office that he considered it
impossible to stop the Germans from entering Portuguese East Africa if they wished to do
so. More ominously, he noted that should he be required to continue the campaign in
Portuguese territory, then operations would be very lengthy. It was to be a prescient, if
382
forecast.
unwelcome,
320WO 95/5292,War Diary GHQ, 22 February 1917,Telegram G 218, Hoskins to CIGS, 22 February.
3" WO 95/5292,War Diary GHQ, I March 1917,Appendix A, Telegram, G 337, Hoskins to CIGS, I March.
382WO 95/5292,War Diary GHQ, I March 1917,Appendix A. Telegram, G 337, Hoskins to CIGS, I March.

119

GeneralRobertsontook only a few days to make a decision, informing Hoskins on

6 March that he agreedwith the latter's assessmentof German intentions. He also stated
insufficient
Lettow
to
it
impossible
be
of
a
combination
to
owing
that should
surroundvon
forces and size of the country, the EAEF should concentrateon the most dangerousor
Nyasaland
The
to
first,
defeating
time.
threat
them
via
at
a
one
vulnerableenemycolumns
PEA was seen as less serious since British control of Lake Nyassa would allow the
On
be
German
before
thrust
forces
by
a
could
effective.
made
any
steamer
switching of
lesspositive note, he deemedNorthey's forces to be overly dispersedand suggestedthat
On the vexed question of supplying more carriers,
former
Portuguese,
the
Belgians
being
the
with
and
conducted with
negotiationswere
having promised to supply some 5,000 recruits from their occupied territory. The CIGS

they be concentratedpromptly.

in
for
had
the
War
Cabinet
a campaign
potential
considered
also announcedthat the
PortugueseEast Africa and had decidedthat if the Germanswere to break into that colony
383

in force,then they were to be pursueduntil they surrenderedor were defeated.

Given the go-ahead in principle, Hoskins quickly ran into problems with his
Based
South
African
on prior assurancesof renewed
assumptionsabout
manpower.
his
the
assumed
plans
convalescence,
and
support after a suitable period of refitting
inability
to provide
first
the
infantry
brigade.
The
trouble
was
sign
of
provision of an
384
followed
by
This
for
in
East
Africa.
the unwelcome
the
was
replacements
units still
The
battalions
to
that
the
news
promised
would not now return owing recruiting problems.
PrimeMinister, Botha, becamepersonallyinvolved in the decision:
"... GeneralBOTHA greatly regretsthat his apprehensionalready conveyedto you
for
is
insufficient
leave
the
it
is
that
quite
months
one
are confirmed and
now clear
bulk of the men who have come back, and that it is impracticableby any meansto
385
fit
',
battalions
trained
those
to
men.
and
of
strength
make
up any respectable

... WO 95/5292,War Diary GHQ, 7 March 1917,Appendix A4, Telegram 30514, CIGS to Hoskins, 6 March.
3MWO 95/5292,War Diary GHQ, 9 March 1917,Appendix A7, Telegram G 447 Genstaff to Norforce, 9 March. This telegram
quotesthe signal to UNECA (South Africa) on 26 Februaryand the reply (no date given).

120

Hoskins promptly made London aware of the magnitude of the shortfall with the
corollary that the campaign would be prolonged even further if the shortage of troops
386
persisted.

While this dispute was underway, Hoskins was discussing his plans with Northey
telegraphically, having been prevented by the weather from a personal visit. In response,
Northey demurred, stating that there were insufficient troops to round up the enemy over
such a vast area. He also thought that Wintgens was making for Tabora and that the
follow-up would be difficult owing to the country and the lack of supplies in his wake.
Given the German foraging parties already in Portuguese territory, he expected to have to
commence operations around the southern end of Lake Nyassa before long. In response to
a comment, Northey replied that he had been unaware that the Belgian forces had ceased
387
before.
Hoskins was astonished to learn that Northey had
active operations some months
not been informed of such major developments and briefed him on the latest situation. He
went on to outline his strategy that boiled down to two essentials: beating the Germans in
the field and depriving them of their food centres. Envisaging a commencement of the
advance in early May as the ground began to dry out, he reiterated his plans for a
converging attack by all his columns towards the centre of the south-eastern portion of the
388
colony.

As March continued, relations between Hoskins and Robertson began to

deteriorate,with the CIGS's requestfor more detailedplans receiving a rather vaguereply.
The commander-in-chiefcounteredthat he would be unable to formulate his exact plans
until he receiveddetails of the transport that would be available and merely reiteratedhis
idea of a concentricadvance,albeit with reducedeffectivenessand range.389This, and the
unco-ordinatednature of his requests,irritated Robertson considerably and he ordered
Hoskinsto submit a single consolidatedand up-to-datelist of requirementsfor the future.
395W095/5292, War Diary GHQ, 11 March 1917,Appendix A12, Telegram H3359, UNECA, Pretoria to Hoskins, 10March.
386WO 95/5292,War Diary GHQ, II March 1917,Appendix A13, Telegram G 487, Hoskins to CIGS, II March.
317WO 95/5292,War Diary GHQ, 18 March 1917,Appendix A28, TelegramNF 2541, Norforce to Hoskins, 17 March.
3111
WO 95/5292, War Diary GHQ, 19 March 1917,Appendix A34, Telegram G 601 GeneralHoskins to Norforce, 19 March.
319WO 33/858, European War Telegrams,SeriesD, Volume2,1' March, 1917, 14' April, 1919, London: War Office, 1919, 27,
to
p.
No. 1696,21 March 1917,Telegram31391, CIGS to Hoskins. Henceforth, TelegramsD 2; WO 95/5292, War Diary GHQ, 29 March
1917,Appendix A41, Telegram G 702, Hoskins to CIGS, 29 March.

121

Oncethis had been received, the War Office would inform him of what personnel and
have
dates.
Hoskins
to make
then
be
would
materialcould supplied and as well as arrival
390
his plansaroundthe availableresources.
Confirmation of additional transport and specialistsupport for East Africa came in
391
April
Now armedwith firmer details, Hoskins signalled his final plans at the end
early
.
having
forces,
Germans
thinned out
In
the
the
terms
assessed
as
were
of
month.
of enemy
their troops in the Rufiji area, with detachmentsnear Mpotora, Makangaga,Ukuli and
Lindi. Kraut's main body was in the Tunduru area with Wintgens still moving west and
dependent
by
issued
Hoskins
disclaimer
that
the
then
on the
plans
were
stating
north.
a
MT
the
the
and
vehicles, and the
of
porters
supply
weather,
arrival of reinforcements,
preparationsof the Belgians. However, he remained committed to the multi-column
concentric advance,with the Nigerian Brigade advancing south from the River Ruf1ji
towardsMadaba,with the Kilwa force of two brigadesmoving south-westto Liwale, and
the Lindi brigademoving either westwardstoward Massassior north-castto the Mbemkuru
River as necessary. The Iringa Column would clear the area between the Ruaha and
Ulanga Rivers while the newly arriving Belgians would take on responsibility for
containing any attemptedbreakouts to the west and moving on Mahenge. This would
Liwale
be
Northey
further
to
to
or
to
either
enable
prepared move on
concentrate
south and
Tunduru or both as required. Finally, a generalreservewould be kept back at Morogoro.
The Portugueseforces were consideredvaluelessand were more likely to provide supplies
to the enemythan effective opposition.392
ANGLO-DELGIAN

OPERATIONS

The re-entry of the Belgians into the campaign was spurred by the urgency of

rounding up AN Wintgensand the need for more troops in the main area of operations.

With political agreementreachedin London,Hoskinsandhis Belgianoppositenumber,

Colonel Huyghd, met in Ujiji from 18 19 April, and modified the troop levels to a total
390WO 95/5293,War Diary GHQ, I April 1917,Appendix A6, Telegram 31916, CIGS to Hoskins, I April.
391WO 95/5293,War Diary GHQ, 9 April 1917,Appendix A78, Telegram 32324, CIGS to Hoskins, 8 April.
392WO 95/5293,War Diary GHQ, 27 April 1917,Appendix B21, Telegram G 165, Hoskins to CIGS, 27 April.

122

force of 3,000 soldiers divided into two mobile columns of 1,200 rifles, each supported
by an operationalreservebattalion of 600 rifles, plus a further two battalions of 500 men
eachto provide individual replacementsfor casualties.
The first Belgian column was to be ready immediately to take up the pursuit of
Wintgens while the second was to be organised and equipped for the main offensive.
Once the raiders were dealt with, the two columns were to unite in the Iringa area
preparatory for an advance across the Ulanga River towards Mahenge.

They would

operate between Norforce in the Songea-Ubena area and the British detachment at Iringa.
Liaison officers were exchanged at both general headquarters as well as with the various
column headquarters, while the Belgian base was fixed at Dodoma on the Central
Railway. 393

A major consideration for both officers was time; for if the conquest of East
Africa were to be made before the start of the rainy season(mid-November to early
December),it was vital that the offensive commencedby the end of July or mid-August
at the latest. Failing that, a new campaign in 1918 would be necessaryunder even less
favourable conditions. It would exposethe pursuersto all the problems encounteredin
the past rainy seasonand would allow the Germans a chance to regroup and, more
importantly, cultivate new food supplies.394
Fortunately, Colonel Huygh6 had taken a number of precautions beforehandthat
were now to pay useful dividends. Despite not being required to raise extra forces in the
April agreement,he had ordered the re-mobilisation of three additional battalions in case
they were needed in the new circumstances. In May, he had organised his available
forcesinto two major columns, one basedon the old Brigade Sud, armed with the Mauser
rifle and the other basedon the Brigade Nord, armed with the Gras rifle. Each brigade
also had a battalion dedicatedto the provision of battle casualtyreplacementsas well as a
battery of six 70mm St Chamond mountain guns and a pioneer-bridging company. An

393CampagnesColoniales Belges,111, 34-35.

pp.
394CampagnesColoniales Belges,111, 131.
p.

123

formed
battalions
d'Operation
RO)
(Reserve
two
strength
was
of
operationalreserve
or
while two othersprovided security for the lines of communication.
As dealing with. 4bt Naumannwas the first priority, Belgian troops were launched
in pursuit early in May. Three battalion-sized columns were formed with the aim of
preventing interference with the vital Central Railway. Since the allies were unable to
launched
line,
Belgian
from
breaking
Naumann
the
that
columns
were
prevent
north of
north in hot pursuit, following the raiders all the way to Lake Victoria. Their task then
becameone of preventing an incursion into British East Africa. "'
The regrouping of Belgian troops took place throughout June and July. GHQ

battalions
from
June,
Ujiji
Dodoma
29
the
to
various
made the
while
moved
on
preliminary moves noted above. Thereafter, further time would be needed for
concentration
and preparationin the forward areasbeforeall was in readiness.On the
in
battalion
Katanga,was
less
Sud,
Brigade
the
the
still
eve of
one
operations,
Brigade
The
Kigoma.
Lake
Tanganyika
the
the
organisation
of
concentrated
of
at
port
Nord was less satisfactoryas it had two battalionsstill involved in the pursuit of Abt
Naumann,with anotherstill guardingthe rear areassouth-westof Taboraandthe fourth
396
for
Iringa.
The artillery was moving forward to join the brigadeswhile the
en-route
pioneerswere preparingto start work on the new baseat Dodoma. The rear services
operatedon the existingline Stanleyville-Albertvillewith new stagesof Dodoma-Iringa
andKilosa-Ruaha.By the end of June,the Belgianswere slowly moving into position
397
for thenextphaseof the campaign.
ANGLO-PORTUGUESE

OPERATIONS

Relations between the British and Portuguese were not so easily resolved
although both sides did make efforts to work together. With the recall of General Gil in
1916,and the plannedsendingof anotherexpeditionary force in early 1917,the Governor
39sCampagmsColonialesBelges,111, 132
pp
and 309-310. Orders are given at Annexe No. 16, Tabora, 30 May 1917.
396CampagnesColoniales Belges,III,
pp. 133-134.
397CampagnesColonialesBelges, III,
pp.135-136.

124

General, Dr Alvaro de Castro, assumedthe role of Commander-in-Chief. Consciousof

in
late
1916.
Smuts
General
for
he
had
his
the weaknessof
asked a meeting with
position,
However, this was not possible for a number of reasonsand it was not until 25 January,
de
Smuts
in
Africa,
East
that
had
days
he
with
met
two
relinquished command
after
Castro in PortugueseEast Africa. It seemshighly likely that General Hoskins was too
busy with taking over his new command and conducting the offensive to make the
journey and Smutswas conveniently passingby.
If the renewed Belgian assistancewas welcomed, General Northey faced further
by
from
being
border.
Apart
in
Nyasaland-Portuguese
the
weakened
problems the southon
the pursuit of Wintgens,he also neededtrained troops to deal with the developing German
in
held
Portuguese
Although
by
Stuemer.
Abt
the
generally
were
threat posedthere
von
low regard, GeneralHoskins did try to involve them in support of both his and Northey's
"'
operations.
At the meeting with de Castro on 9 May the subject of military cooperationwas
high on the agenda. The Portuguesewere always slightly suspiciousof British intentions
To
Hoskins's
into
Portuguese
forces
territory.
to
their
and were very reluctant
allow
dismay, de Castro proposedto send a force of over 700 rifles from the coast to Chinde,
thence up the Zambezi River into southernNyasaland and Lake Nyasa, where it would
land at Mtengula which was believed to be von Stuemer's objective. He suggestedthat
this movement would take place between 15 and 20 May, but this was hopelessly
diplomatically,
facilities.
Although
the
Portuguese
transport
received
optimistic given
in
likely
Northey's
it
to
get
seen
as
was
as
suggestionwas consideredmost unwelcome
it
be
British
From
the
would
perspective,
way and would causemajor supply problems.
far more useful for the Portugueseto establisha strong blocking force along the Rovuma
as well as protecting the vulnerable line of communicationrunning up the Zambesi River.
Hoskins put his suggestionsas tactfully as possible, but, unable to change da Silva's
399
he
left
the matter unresolved.
mind,

WO 95/5293,War Diary GHQ, 27 April 1917,Appendix B21, Telegram G 165, Hoskins to CIGS, 27 April.

125

This was as well, for the Portuguesetroops only reached Chinde, at the head of
the Zambezi, on 25 May and von Stuemer's move on Mtengula forestalled their further
local
Portuguese
it
Instead,
that
the
commander would meet
advance.
was agreed
Northey in Zomba to arrangea revised deployment. This was ultimately decided as the
reinforcementof the existing Portugueseboma at Mlanje, on the other side of the BritishPortugueseborder, and was forward of both the railway line and River Zambezi. It
provided much-neededprotection of each force's lines of communications and, as the
400 With the outbreak of a major antiBritish desired, kept the two annies separated.
Portugueserevolt in the districts of Tete and Barue, to the south of the areaof operations,
and the need to reinforce the beleagueredgarrisons there, it was probably as much as
could be obtained. This would remain a maj or considerationfor the Portugueseuntil the
401
June.
beginning
the
rebellion cameunder control at
of
GERMAN REORGANISATION

The Schutztruppefaced similar problems to their opponents, particularly from

insufficient transport,food and medical supplies. However, they did have the advantages
of being on interior lines in friendly country and could retire on routes of their own
choosing. They were also able to live off the country, having few scruplesabout clearing
out local food stocks.
On 1 April 1917, the total strength of the 0suruppen was some 6,534, consisting
402
1,423
Germans
fight.
5,111
Askaris,
4,419
of
and
of whom
were considered able to
The numbersfor the Westtruppenwere 2,854, with 535 Germansand 2,319 Askari (Abt
Wintgenshad another554 soldiers).403With a total of over 9,000 mouths to feed and the
loss of the lower Rufiji, the question of food supply was increasingly severe. Stocks had
been consumedfar faster than expected,with levels held on the lines of communication
some 125,000kg less than the forecast 300,000. As the forward force consumedover
3" CAB 44/9, pp. 39-40.
400CAB 44/9, pp. 40-41.
401WO 95/5329,War Diary Norforce, Entries 25
and 29 May 1917;Ranger,Terence0,
402MS Boell, N14/30,26. Kapitel, 2032; "Revolt in PortugueseEastAffica: the MakombeRising 1917", in St Antony's Papers,(ed) K
p.
of
Kirkwood,Number 15Affican Affairs No 2, London:Chatto& Windus, 1963,pp. 62-64.
403MS Boell, N14/30,26. Kapitel, 2044.
p.

126

404

The struggle to provide adequate

levels.
As
to
drastic
minimal
nourishmentrequired
measuresand rations were reduced
10,000kg daily, this was a significant shortfaJI.

thehistorianof the campaignpointed out:

"Neverthelessthe German forces found themselvesin a more critical situation
than the British, the most critical that troops can face, that of existenceor nonit
for
time
The
the
and
would
second
existence.
spectreof starvationthreatened
not be easily overcomeby the reoccupationof the southernportion of the Dar-esSalaamdistrict."405
OPERATIONS ON THE RUFTJI AND THE FAST

Operationson the middle and lower Rufiji during the period February to May
1917largely consistedof thinning out the forward positions and patrol actions owing to
the rain and flooding of the countryside. Sporadic fighting and occasionalraids took
food,
insufficient
by
both
place,although,with
sides seriously weakened sicknessand
suchencounterswere usually indecisive.
The Germanshad notedthe British redeploymentand the reductionson the Rufiji
line. Sufferingfrom similar problemsto his opponents,von Lettow decidedto reducehis
forward forcesand to begin the slow withdrawal to the south. He wantedto be ready to
fend off any subsequentBritish moves against his lines of communicationas well as
reinforcehis troopsalong the coast.406 Concernsaboutrear areasecurity and the loyalty
of the African population were also apparent,especially in the area of the Makonde
highlands.The British had beentrying to inducethe inhabitantsto rebel againstGerman
rule sincethe previousautumnand had provided somematerial support. The disaffected
tribes began to disrupt food gathering and made the links to the west much more
dangerous
than previously.

MS Boelt NI4/30,26. Kapitct

p. 1990.
MS Boell, N 14/30,26. Kapilci.
p. 1987.
80c11,Die OPerationen.
p. 319.

127

If the main force was suffering badly from the climate, Norforce, although better
off, neededa rest after the long advanceto Iringa and the subsequentfight againstthe
Westtruppen.The columnswere tired and requiredreorganisationwhile the long lines of
communicationwere now struggling to function in the wet weather. However, further
unexpectedeventswould shortly makethis requirementevenmore troublesome.
Although the heavy rains had brought operationsin the Rufiji areato a complete
halt, the precipitation further south %N-as
initially.
least
As has already
at
not as severe,
beendescribed,the bulk of the lVesttruppenhad brokenthrough Northey's slenderforced
and had linked up with AN Kraut in the Ulanga valley. Although they had suffered a
significantnumberof casualties,the now-united force was much strongerthan Norforce.
While the low ground was flooding heavily, two enemy groupsremainedon the edgeof
the greatescarpmentand in close proximity to Northey's troops. AN Kraut remainedin
its strongdefensiveposition at Mfrika watchedby Hawthorn's column, but AN Wintgens
was not fixed. From captureddocuments,Northey was aware of the supply problems
facedby the Germansin the Mahenge
both
instructions
Wahle's
ordering
areaas well as
407
Kraut andWintgensto operatein the Wiedhafen-Songea
for
long
aspossible.
area as
The threat to Norforce was very real, for on 3 February both Ateilungen arrived
at Gumbiro and surrounded its garrison. On receiving Wahle's orders of 29 January, the
two commanders had a major dispute. Kraut, the senior officer, insisted that they move
south in accordance with their instructions, but Wintgens, a strong-willed and difficult
personality, wanted to go north. It is unclear why he disobeyed such explicit orders,
although it has been suggested that his askaris wanted to return home to Tabora and that
he did not get on
well with von Lettow. Whatever the reason, on 6 February he lifted the
408
unsuccessful
It
Kraut took his troops
to be the

siegeand movednorth while

south.
was
startof an epic movementthat would only end someeight monthsand many thousandsof
mileslater.

Boell,Die OPerationen. 31S. At Schoenkk Li"cke

Langennweredrawing on the Mahengewhile At Kraut and
P.
and
von
Wintgenswereordered
by
to
Wahleon 29 January.
move
408
Boell,Die OPErationen. 325,16ISS
Aft. c. 1713(300), HWopyof I X4P, p. 165.
p.

128

Norforce was widely stretchedover a vast area of country; the distance from his
headquartersin Ubena to the forward troops was 130 miles alone. The columns were
split between Likuyu, Songea,Kitanda, Alt-Langenburg and Lupembe. As the month
progressed,the rains became increasingly heavy and the broad lowlands east of the
escarpmentcontinued to flood. Although on higher and healthier ground than Hoskins's
main body, the troops of Norforce were still suffering from the climate and disease.
Numbershad been reducedto less than half normal strength and the South African units
409
hard-hit
from
were particularly
sickness.
On 8 February, intelligence indicated that Kraut was moving southwards;this was
confirmed two days later when he appeared on the Wiedhafen-Songearoad making
severalunsuccessfulattacks on isolated British posts. He then continued south the next
day with Hawthorn now in hot pursuit. Reportedly in a discontentedstate from lack of
supplies,Kraut's askaris moved quickly and, despite some heavy knocks by Hawthorn's
column, they managedto make the Portugueseborder. On 24 February, Northey was
informed that a large enemy force was assemblingat Likuyu and he recalled Hawthorn
back to Songea.410 No longer being pursued, Kraut was now to forage in the southern
portion of GermanEast Africa and poseda potential threat to the Portuguese.
Wintgens was concernedabout the increasingly constricted areaof operationsand
the need to obtain food. In order to regain his freedom of movement, he intended to
conduct minor operations on the line Songea-Lupembe and then to break northwestwardstowards the northern end of Lake Nyassa, to Tandala. Leaving his sick and
woundedbehind at Milow Mission on 15 February,he ran into a combined South African
and KAR column the next day. The British column was forced back that night, having
lost its commander and a number of others killed. This fight confirmed the renewed
threat to Northey's lines of communication running betweenAlt-Langenburg and Ubena,
while his forces were too small to prevent a breakthroughto the west. Acting rapidly, he
recalled Murray's Column from the Ruhudje, and it promptly marchedback to Tandala,
at the northern headof Lake Nyassa.
40 CAB 44/9, p. 33, Sketch96.

129

KETCH

84,

llembule

Mpan
AfPQ(l

Cl
Ernmuberg

Rgangule
3onlours approximate
AfoightsIn metres

Kidna
4-

- -

2,
0100
I 1 UPCM
-L

-\Dl

"M

to

2nd February 1917

16
12/4(, Vl,
A

---A ,

OPERATIONS OF
NORTHEYS FORCE
19th December 1916

Twangv
I
-) 16/1

K8H

7 4/
Q
im

it

'Pitu

II ewe
J5
.. Ibn a,
sladaba

":At
a eti U

Lilowa
Sadane

A,

LAKE
NYASA

Cpp.
Cpi

<
w

Lukamanon
31
%K
Imputa
putla

J;L

114/1
4/1
K, tagda

GumbiruO
fus

REFERENCE
/0

27,1'
0 "'0

in

'x
rl *C;.
i rthy

WAtens. Krout
K.
DfundedPositions
rl
171111
P'$ 1,
--CAI &ICYIL-N
_
OMAkc.

R No."'r

1*

3 -1

Likuy

Makngandu
L, I

1. - -.
'0
All"

PIN
*'I 11.)
- 9 .

Iq

Wic
;v

tHWIAPY

t,

Bagaya

R.

ilyamabengpIL
Plrraml

C'...

1, wile JO 3
sorlea8
Cwlsbl

F-".

npadu

Plain

Nyamtumbo
Scale-28-77.
1.0

210

Miles to I Inch
3,0

4.0
0, "

50 Milos
R. -

I",

Two days later, a KAR company near Tandala was nearly surrounded and cut off
by the Germans who then besieged the garrison. Murray's Column timely arrival on 22
February forced Wintgens to abandon the siege, leaving behind a destroyed 3.7 cm gun.
Heading northwards, Wintgens had about 520 soldiers, II machine guns and two field
41
1
As soon as the relieved garrison was sorted out, Murray set off in hot pursuit on
guns.
25 February. 412

Now faced by serious opposition, Wintgens intended to use his head start and
in
best
He
to
to
superior mobility
planned make a series of short marches a
advantage.
leading
lines
British
direction
he
the
of
communication
old
north-westerly
until
reached
from Neu Langenburg to Iringa. From there, he would feint towards the Rhodesian
border while actually moving towards Lake Rukwa. He expectedto find more food in
that region, which so far had been largely undisturbedby the war, while shaking off his
he
lost
direct
These
as
ground contact was
pursuers.
measureswere successful and
turned south-westnearNeu Utengule. British aerial reconnaissancesubsequentlylocated
his camp and then successfullybombed it on 26 February. However, it was not until II
March, when the Abteilung had reached the Igale Pass, south of Alt Utengule, that
Murray's patrols were first able to restore contact. The continued pursuit convinced
Wintgensto move away from that areaand to make for St Moritz further north.413
Murray was a determined and capable officer who could be thoroughly relied
upon. However, even his best efforts could not force Wintgens into pitched battle,
chiefly for reasonsof supply. The countryside was lightly cultivated and the Germans
stripped all available food in their wake, while Murray had to have his food brought
forward by carriers with all the inefficiencies and delays that this method brought. As
well, there was considerableuncertainty as to Wintgens's intentions; and arrangements
had to be madeto cover moves either to the south or west.414
4'0 Hoskins, Despatch,pp. 159-160.
411Boell, Die Operationen,p. 325. The Germanslost two killed and sevenwounded,while the British suffered nine deadand 13 taken
.soner. Wintgens left Gumbiro with 59 Europeans,464 Askaris, II machine guns and two guns.
F-r"Hoskins,
Despatch,p. 160.
413Boell, Die Operationen,p. 325; WO 95/5320, War Diary Norforce, Entries 24 February, 10 and II March 1917.
414Wo 95/5329, War Diary Norforce, Entry 8 March 1917;Hoskins, Despatch, 168.
p.

130

Murray's Column had reachedAlt-Utengule on 13 March and St Moritz Mission

five days later. He had also tried to securethe Northern Rhodesianfrontier by sending a
battalion to garrison Bismarckburg, while another,detachedfrom Hawthorn, was ordered
to protect Alt Utengule. A further patrol action took place on 17 March with the
Germansbeing scattered. Murray now decided to move on St Moritz, believing that he
had Wintgens trapped against the swollen Lupa and Songwe Rivers, which were only
415
by
bridges.
crossable
However, the patrol victory was reversedon 20 March, when Wintgens launched
his main body in an attack. The battalion suffered heavy casualtiesand was forced from
the battlefield, moving to a position some 6 krn to the south-west. They were not left at
rest, for two days later, their camp was surroundedby the enemy and they were cut-off.
A breakout attempt on 25 March was rebuffed by the besiegers, but again Murray
marchedto the rescue, arriving on 26 March. Once again, Abt Wintgensbroke off the
bridgehead
headed
for
St
Moritz
position was taken up.
siegeand
where a strong
After some reorganisation the bulk of Murray's troops continued the pursuit
towards St Moritz.

During this time, Wintgens had been busy devising an escape over

the rivers hemming his troops in. Constructing some improvised rafts out of crates and
local material, the first crossing of the Lupa began on I April with the entire Abteilung
being evacuated by the P.
Murray, delayed by supply problems, arrived just as the
enemy completed its escape. Lacking rafting or bridging material and suffering from a
severe lack of food, he was unable to cross the river and the follow-up was again broken
416
off.

The emergenceof these five companies into the previously secure rear areas
causeddisruption out of proportion to their fighting strength. The thickness of the bush,
the sheersize of the country being traversedand the lack of any infrastructure meant that
substantial pursuing forces were required. In order to support Murray, at the end of
4'sBoell, Die Operationen, 326.
p.

131

March Hoskins sent an Indian battalion to take over as the garrison of Bismarkburg,
while a still-forming KAR battalion was diverted from training and sent to block any
keeping
4"
The
British
Tabora
the
troops
and
problems
of
approacheson
shortage
of
.
them supplied now led to a formal requestfor the resumption of hostilities by the Belgian
Force Publique. As this would entail the re-mobilisation and re-deployment of the bulk
of this force, under similar supply difficulties as the British faced, it would be not be fast
enoughto catch Wintgens quickly.
Wintgens's plans on crossing the Lupa River were to make for Kipembawe,

movingin the oppositedirectionon the routehe hadusedthe previousOctober. He too,

wasconcernedaboutthe ration situationandsplit off a columnto relievethe pressurein
the local areaaswell asto seta falsescentby marchingin variousdirections,with orders
to rejoin him at Mdabulo. He launcheda feint move on Bismarckburgand spentthe
418
remainderof thetime collectingsupplies.
In the meantime,the KAR battalionhadreachedTaboraby rail andmovedsouth
on foot to Kitunda Mission, reaching it on 5 April and remaining there for a number of
days. The Germans,having rested for five days, began the move north again, this time
towards Kipembawe. In late April after a clash of patrols, the British commander,who
was outnumbered,took the prudent course and withdrew back to Sikonge. This left
Kitunda Mission open, and it was occupied by the Gennans on 4 May. Here Wintgens
decided on a major halt as he and many of the Europeanswere suffering from typhus.
However, Murray was once more in pursuit and his leading elementscame into contact
from the south on 6 May. It took a number of days for the main body to catch up and it
was not until 14 May that the column was close to Kitunda, by which time the enemyhad
moved north.
By 21 May, Wintgens's condition had worsened seriously and he determined to
surrenderto the enemy. Commandpassedto Lieutenant Naumann, who was enjoined to
416Boell, Die Operationen, 326. WO 95/5329,War Diary Norforce, Entries 22-26 March 1917.
p.
417WO 33/953, TelegramsD 2, No. 1727,31 March 1917,TelegramG 738, Hoskins to CIGS, 38.
p.
418Boell, Die Operationen, p. 326; WO 95/5329, War Diary Norforce, 24 May 1917.

132

ll
.

4t
z

141

%iA
Itt,
W-

IL
tv

Ac

41r

4t

41

T.

&%

*,

Ad

.
%%

%A*.

.-:!

-As%-

A-1

tus .

I
-t
es
.
I-

..........

xI

Ul

4r
J-L

-W

'$

!; 6

't

%6.
Icl

)c
-Z
all

a,

so.

14C

Im IQ
ll
Al X

qr

fJ,
(32-c-

link up with the Wesuruppen,back in the Mahenge area. In the circumstances,with

Murray pressing from the south, this was impossible, and Naumann decided to march
4
19
Despite
it
the
Mkalama.
the
Central
Railway
to
pressure,
to
the
at
cross
and
north-east
fight.
full
they
of
remained
askaris'morale remainedgood and
In the meantime,the Allies were gathering further forces to deal with the raiders.
The Belgians had taken over responsibility for Bismarkburg and Sikonge while the
British moved reinforcements up the Central Railway to co-operate with Murray.
Brigadier General Edwards, the Inspector General of Communications, was placed in
half
battalions
had
three
total
and one
the
and
a
of
a
chargeof
operations and was now
420
deal
Naumann.
squadronto
with
The withdrawal of the main body of the Osttruppe began in early March as soon
its
from
forward
back
had
it
became
I"
Division
that
the
positions towards
pulled
as
clear
Kibata and Kilwa. Leaving only a relatively weak force on the Rufiji, von Lettow moved
later.
In
days
five
Mpotora
the
headquarters
his
12
March,
at
south on
new
establishing
in
Sfidbefehlshaber
forces
he
the
the
order to meet the
of
meantime, also strengthened
increasingenemy strength in Sudi and Lindi. At Rothe, with three companies,arrived
421
forces
in
defending
there mid-March, strengtheningthe
considerably.
South of the Rufiji, the Germans still held Lake Utungi and the Ligonya River
battalionformidable
A
to
of
small
number
valley which was now a
obstacle movement.
level operationsin the secondhalf of April near Utete resulted in local British successes,
but fighting petered out by early May as both sides were now in the processof shifting
422

their forces southwardtowards Kilwa and Lindi .

Thesemoves enabled I" Division to move both its brigadessouth to the Matandu
River. The brigade structure was now dropped, and on 2 April Colonel Grant (the
419Boell, Die Operationen,p. 327. Naumann took over 48 Europeans,42IAskaris, II machine guns and two guns.
420WO 95/5293,War Diary GHQ, 9 May 1917,Appendix A62, Telegram P 19, Edforce to Genstaff, 9 May. The new force was
entitled "Edforce".
421Boell, Die Operationen,pp. 319-320. Looff had 65 Germans,500 Askaris, nine machineguns and two guns. There were also 100
auxiliaries who had beenenrolled as recruits.
422CAB 44/9, pp. 23-34.

133

former 2nd EA Brigade) was given command of all troops north of the Matandu and
Colonel Rose (the former P EA Brigade) was given command of all troops south of the
423
In early April, patrols noted activity in the area of Kimamba Hill and Rumbo.
river.
A camp was set-up in the bush around Makangaga; on 12 April it was estimated that
Rirther
three
the
to
ten
a
main
road
with
some eight
companies were coming up
424
Mchakama.
General
Kimamba
further
via
companies moving
east towards
Hannyngton,the GOC 1" Division, was concernedthat the presenceof this force would
disrupt his preparations for the general advance and decided to take pre-emptive
425
In mid-April, he instructed Rose to drive the enemy out of its positions.
action.
Accordingly, a column consisting of a reinforced battalion, numbering some 540 rifles,
left Rumbo on the morning of 18 April to attackAbt von Lieberman, which was over 300
strong.
The journey was not easy as the column had to cross the steeply-sided and
swollen Ngaura River before reaching the enemy defences, and it poured with rain
426
day.
The going was slow and difficult with the column soon running
throughout the
into heavy fire. Facedby three companiesin good positions surroundedby tall elephant
grass and dense bush, visibility was very poor and communications difficult for the
attackers. A flanking move was unable to make sufficient progress and it was decided
that the enemy was too strong to defeat and a withdrawal was ordered. However, von
Lieberman reacted vigorously and threw in a strong counter-attack that nearly
overwhelmedthe column's rear guard, losing three machine guns after their teams had
beenkilled. Furthermore,a panic amongstthe carriers resulted in the loss of a great deal
of ammunition and baggage. The situation was saved by a courageousand determined
427
improve
did
little
defeat
by
to
that
the rear guard,but it was a stinging
effort
morale.

123CAB 44/9, p. 18. Mnasi was 22 miles WSW of Kilwa Kivinje, near the
main road to Liwale.
424CAB 44/9, pp. 19-20.
425Moyse-Bartlett, TheKing's. 4frican Rifles,
pp. 360-361. On the surfacethis action seemsunwise as the column mustereda
maximum of 5 companieswith the aim of locating and beating a dug-in enemy of superior numbers. There was no generalreserveto
back up the column.
426CAB 44/9, p. 20. The 4e Pathanssupplied400 rifles and V/2"d KAR 140 men.
427Clifford, Gold Coast Regiment,pp. 82-86; CAB 44/9, pp. 20-22. Facedby the At
von Debermann with 300 rifles and four
machineguns,the Germanslost three deadand 22 wounded as comparedto the British who were much more heavily hit with 36
killed, 70 woundedand 43 wounded,totalling 152.

134

428
but it was unlikely to
failure,
Hoskins later tried to put a positive gloss on the
have worried the Germans. Indeed, the whole point of the raid was highly questionable
from the outset, as the sending of the equivalent of five companiesagainst a detachment
forward
defences,
in
least
300
have
the
8-13
to
was almost
estimated
companies,with at
dangers
it
fail.
As
the
to
the
of sending weak
re-emphasized
certain
attack merely
was,
and unsupportedcolumns againstunreconnoitreddefences.
This event marked the end of serious offensive efforts by 1" Division for the
in
losses,
British
the
heavy
The
the
and
officers,
particularly
remainderof
rainy season.
drop in morale were serious, but the lack of carriers really precluded any resumption of
the offensive.429It also showedthat the new KAR battalions still had a long way to go in
terms of training and tactics before they were able to take on the more experienced
Schutztruppeon equal terms.
It was the combination of the Makonde rising and the British build-up at Lindi
that was now the main concern for von Lettow.

In mid-March, he ordered At

three companies strong, to leave Tunduru for the port of Mikandini.

Krfiger,

Initially given orders

to crush the rebellion in the Makonde plateau, KrUger was soon drawn into Looff s
operations against the lodgement in Lindi. Indeed, the Kommandeur believed it was time
to strike before the British could reinforce further and to drive them "out of the
430
continent".

Meanwhile, O'Grady's brigade launchedan attack on At Rothe, taking a forward

company position at Njangwani, some 12 km north-west of the town. Eager to redress
the loss and to use his newly arrived reinforcements,Looff decidedto strike a week later.
Leaving At Krager opposite Lindi, he sent At Hinrichs, with three companies,to take
the small port of Sudi. The position was a strong one, with the defendersprotected by
'28Hoskins, Despatch,p. 166. "Although this attack failed in its immediateobject, it had the effect of relieving the Kilwa force of any
further pressurefrom this direction."
429WO 95/5293,War Diary GHQ, 19 April 1917,Appendix Al 15. Telegram0 402,1 Div to Hoskins, 18 April; War Diary GHQ, 19
April, Appendix 116,Telegram0405,1 Div to Genstaff, 19 April. In this signal, the enemy was estimatedat being between I-2,000
strong.
4'OBoell,DIeCperafionen, p. 32l. AbtKraut had beenbroken up and the bulk of his troops had passedto Kroger. Krautwasordered
to move to Lindi and to assumethe appointmentof SadbefehIshaberfrom Looff. He assumedcommandon 5 May.

135

lacked
24
April,
the
Launched
artillery, were
who
attackers,
on
wire and stakeobstacles.
battle
The
defences
was
break
through the
and suffered a number of casualties.
unable to
43
1
The next day Abt
British.
broken off and a single company was left to contain the
Rothe gained its revenge for its earlier setback and seized the lost forward position,
432
into
Lindi.
back
defenders
driving the Indian
Despite von Lettow's stated aim of destroying the British coastal enclaves,the
tactically
Sudi
and
on
an
operationally
soldiers
scarce
of
waste
attack at
was an expensive
for
decisive
declared
he
3
May,
time
that
the
later,
few
days
A
on
unimportant enclave.
face
the
to
to
he
had
two
Lindi
march
north
companies
measuresat
run out and ordered
Dissatisfied with the conduct of operations there and
having clashedrepeatedlywith Captain Looff, he dismissedhim as SadbefehIshaberon 5
May, leaving him in commandof an Ateilung. Major Kraut took his place and quickly
growing threat from Kilwa.

in
Ngurumahamba
the
four
defences
strong
the
placed
companies
with
reorganised
Majani.
to
the
Lindi;
two
at
south
and
west
positioned
position west of
and with another
On 5 May, the Germans caused consternation when they opened fire with a
from
harbour
Kisiwani
Kilwa
the
lying
Portuguese
the
off
ships
at
captured
gun
mountain
for
little
than
This
the
to
other
a minor clearance
result
achieved
mangroveswamps
west.
Kivinje
Kilwa
Further
the
in
the
north,
near
areas.
rear
operation and renewedvigilance
Germans continued to pull back from the Matandu river and reinforce their existing
positions on the Ngaura River. By 20 May, Mpotora had been completely evacuatedwith
the garrison splitting betweenLiwale and Likawage.
The move of the I" Division south of the Matandu was marked by the end of its
formal existence. It becameknown as Hanforce after the name of its commander,Major
GeneralHannyngton, despite repeatedchangesdue to illness. The changesin structure
were also reflected at brigade level which was now formally abolished. The new No I
Column, now led by Colonel GM Orr, arose from the old 3rd EA Brigade, and No 2
411Boell, Die Operalionen,pp. 320-321. The Germanshad a total of 333 soldiers and eight machineguns. Hinrichs lost II killed and
29 woundedout of his detachmentof 265 men and six machine guns.
412Boell, Die Operalionen,p. 321. Rothe had 3 companiestotalling 280 men and 6 machine guns, losing four killed, 20 wounded as

136

Grant, was formed from the old 2ndEA Brigade. The troops
long
25
in
deployed
Column
No
I
mile
follows;
an approximately
was
were deployed as
River
line
followed
front
the
the
Its
Kilwa
of
to
port.
arc, covering the approaches
Column, led by Col HFL

Ngaura to the deep inlet known as Beaver Haven, with a total of five forward postsbeing
held. No 2 Column was more concentratedand occupied Mchemera, on the Matandu
433
It was poised on the main route leading
River, less a few detachmentsto the north.
in
heavy
However,
the
interior.
from
Mpotora
Kilwa
the
rain
continued
towards
and
west
by
followed
drying
it
May
out period
and then required a
coastalareasuntil the middle of
intensive repairs on damagedroads and tracks before the supply system could become
fAilly operationaland effective.434
OPERATIONS

IN THE SOUTH AND WEST

Northey could do little during the wettest months of March and April save hold
the defensiveand recover. From his headquartersin Ubena, on the junction of the roads
leading north and east, his force was strung out south-eastwardson the line LikuyuKitanda-Lupembe,roughly paralleling Lake Nyasa. With the loss of Murray's column,
his forces had been reduced by over three battalions; if At Wintgenshad passedout of
his immediate area, 4bt Kraut posed another threat further to the south either around
.
Songeaor in PortugueseEast Africa. 435 One welcome piece of news in March was
Hoskins's decision to send a KAR battalion south from Iringa as reinforcements for
Northey's weakenedcolumns.

At vonLangennwhich wason the Ruhujeto the eastof LupembeandAt Lincke

Songea-Liwale
River
Likuyu.
Likuyu.
the
(the
road)wereworrisome,
the
on
at
crossingof
Hawthorn
Colonel
Songea.
50
latter
the
of
especially
which was about miles north-east
River
Likuyu
He
the
drive
Lincke
in
from
Songea
March
near
to
arrived
away.
wassent
he
later,
four
days
to
the
7
inconclusive
April
cross
attempted
skirmish
on
and,after an
floodedriver. This took sometime and Hawthorn'sforce was successfullyover the
comparedto British lossesof 22 killed and one wounded.
433CAB 44/9, p. 70, Sketch94; Off, "From Rumbo to the Rovurna", pp 109-110..
434Hoskins, Despatch,p. 167.
435WO 95/5329, War Diary Norforce, Entry 31 March 1917,"Summary of Situation at End of March"; CAB 44/9, pp. 33-34.

137

obstacleon 23 April and was somesix miles north of the suspectedGermanencampment.

He occupied it unopposed on 27 April, with Lincke having evacuated some days
beforehandand moved further east.436
While Hawthorn was struggling through the sodden bush, Kraut's force had
reachedTunduru.and was believed to be making enquiries about accessto the Rovuma
River and the areasto its south. There was considerabledoubt as to the veracity of these
reports, but these were dispelled on 20 April, when a captured document revealed that
Kraut had split his force into two columns, one under himself and the other under Major
437
Stuemer
had
Portuguese
This was unpleasant
territory.
the mission of raiding
von
who
news for Northey, who now faced yet anotherdemandon his much diminished force.
A week later, reports confirmed that von Stuemer, who had left in advance of
Kraut, had reachedthe area of Mwembe, some 80 miles south of the Rovuma River in a
fertile tract of land betweenthe Lujenda River and Lake Nyassa. His force was estimated
to contain 37 Europeans,200 Askaris and a large number of ruga-ruga (armed African
auxiliaries). It appearedthat the local Portugueseauthorities had been taken completely
by surpriseand that a large amount of food had been seized.438Kraut's force was still on
the move and unlocated,but it was presumedto be heading south for the Rovuma.
Whilst PortugueseEast Africa had relatively little strategic value in itself, the
presenceof a German force there was important. Firstly, it could plunder the relatively
rich food areasfor the supplies so badly neededby the main force further north more or
less at will.

Secondly, the extreme antipathy felt by the indigenous Africans for their

Portuguesecolonial overlords meant that recruiting for the German force was both
possible and likely.

Thirdly, Northey's long and vulnerable lines of communication

along the Zambezi River to the Indian Oceanwould be at seriousrisk from such a move.
If these were interrupted, then his entire force would no longer be capable of serious

436WO 95/5329,War Diary Norforce, Entrics II

and 27 April 1917; CAB 44/9, pp. 34-35.
437WO 95/5329, War Diary Norforce, Entry 22 April 1917.
432CAB 44/9, pp. 35-36.

138

offensive operationsas the land links through the Rhodesiaswere too undevelopedto be
of practical use.
Northey reacted at once and sent his only available troops, a South African
battalion, to Fort Johnston on 30 April. This unit, which had been badly affected by
illness and had been were recuperating while carrying out lines of communication
protection duties betweenWiedhafen and Songea,was his only formed reserve. Even so,
it was far from fully trained and barely equalled the numbers of Germans in the area; it
could only be supplemented by KAR
Zomba. 439

recruits diverted from essential training in

The situation was becoming critical as, with the pursuit of Wintgens, Northey's

brigade-sized
force was spreadout over 450 miles of wild bush,mountainand swamp.
He now riskedbeingdefeatedin detail whilst his relianceon lengthysupplylines meant
thathe couldnot respondasquickly ashe would havewished. On I May, hetelegraphed
GeneralHoskinsaskingfor the return of the detachedtroops in order to rest and refit
themprior to futureoperations.This requestwasturneddown in light of the pressureto
hunt down WintgensandNorthey was instructedto keepHawthornfrom going too far
east. Accordingly,Hawthornwastold to stayin the vicinity of Likuyu wherehe had so
440
recentlyoccupiedthe campandcrossingsite.
While these deliberations were taking place, the threat to Northey's southern flank
continued to increase. Von Stuemer was now known to have over 400 soldiers plus a
substantial ruga-ruga contingent.

Furthermore, it was clear that the local population

welcomed the German presence and were actively supporting their activities.

An

advance party from his column had already reached as far south as Mtonia, a village only
15 miles from Lake Nyassa and the British border. His main body was believed to be
heading for Mtengula which lay north-west of Mtonia on the lake shore. Abt Kraut was
also identified in Portuguese territory, about 70-80 miles to the east of von Stuemer, and
in Mwiriti

on the Msalu River. Apart from the above mentioned supply and recruiting

439
CAB44/9,
CAB45n3,Sketch
V June
98"TheStrategic
Situation
1917".
pp.35-36;

139

opportunitiesthat this offered, Mwiriti lay on a motorable road (with telegraph line) that
led to the coastand Port Amelia.441
Having been denied the return of Murray's column, General Northey made
severalchangesto counter the mounting threat to Nyasaland. On 6 May, he moved his
new KAR battalion to Fort Johnstonwhile one of the detachedbattalions was marching
back to rejoin the main body. At the sametime, he shifted his headquartersfrom Ubena
(north and east of Lake Nyasa) to Zomba (south of Lake Nyasa and Fort Johnston) in
442
be
order to near the greatestthreat.
Whilst these further deployments were getting underway, the South Africans had
arrived at Fort Johnston with barely 270 effective soldiers.

Promptly reinforced by

several hundred KAR soldiers from the depot, the unit left immediately for Mangoche, a
few miles to the north and east, where they arrived on 8 May.

Their speed was needed,

for on the same evening an advanced detachment of von Stuemer's force took, the
Portuguese boma at Mandimba, only 6 miles from the border and close to Mangoche.
The Germans actually crossed into Nyasaland the next day, but were deterred from a
further advance by the presence of the newly arrived battalion and withdrew
Mandimba.

to

The British force then shifted to Namwera, a few miles north-east of

Mangoche, where it commanded the main road between Nyasaland and Portuguese East
Africa.

Neither side was strong enough to force the other and a local stalemate
443
prevailed.

Events continued to move rapidly.

General Northey reached Zomba on 14 May

and on the same day half of the KAR battalion arrived in Fort Johnston.

Given the

seriousness of the situation in the south and the fact that the Songea operations were now
halted owing to rain, Northey decided to bring Colonel Hawthorn down to command the
force in Nyasaland. Three days later, on 17 May, the Germans occupied Mtengula on
Lake Nyassa while intelligence showed that they might shift their attentions further to the
440
WO
95/5329,
War
Diary
Norforce,
Entry
2
May
1917.
"' CAB 44/9,

pp. 36-37.
442WO 95/5329,War Diary Norforce, Entry 6 May 1917.

140

southand towards the key port of Quelimane. This would be a direct threat to Portuguese
444
be
neutralised.
operationsas well asNorthey's communications,and neededto
However, it soon appearedthat von Stuemer had encounteredmore opposition
than he had expectedand was unlikely to pressinto Nyasaland,given the relative balance
of forces. A detachmentoccupied Mtonia on 26 May, but patrols and scoutswere unable
to locate any Germanssouth of that place. However, presenceof any enemy was both a
threat and a drain on available resources. Accordingly, Northey resolved to drive them
into
Rovuma
Portuguese
the developing envelopment
the
territory
out of
and north of
being led by the main forces.445
On 25 May, the remaining half of the KAR arrived at Fort Johnston, giving
Northey a local numerical advantage. This enabledhim to order patrols into Mtonia the
next day, followed by its capture on I June. The Germansappearedto be thinning out
and it was decidedto reinforce the advantagesouth of Lake Nyasa. On 3 June,Northey
moved his KAR battalion forward to support the South Africans, while withdrawing his
details of trained KAR soldiers back to Zomba where they resumed the vital task of
446
for
training the new recruits
the forthcoming operations.
HOSKINS'S DISMISSAL

General Hoskins had expendeda great deal of effort in the reorganisationand

refurbishmentof the EAEF. He had taken over a force in the worst of condition and had
done much to restore its fighting effectiveness. Much remained to be done, but in the
circumstancesof an exceptionally wet season,the prevalenceof diseaseand the physical
weakness of the troops under his command, he had been successful. However,
unbeknownstto him, problems loomed on the home front. Although, he was clearly an
energeticand efficient officer, his communicationswith the War Office had not met with
the CIGS's favour. He had issued a large number of Piecemeal and sometimes
44'WO

95/5329,War Diary Norforce, Entry 9 May 1917;CAB 44/9, pp. 38-39.

44'WO 95/5329,War Diary Norforce, Entries 14
and 18 May 1917;CAB 44/9, p. 40.
"s CAB 44/9, pp. 41-42.

141

uncoordinatedrequests while not always answering General Robertson's questions in

sufficient detail. Perhaps more importantly, his efforts belied his predecessor's
pronouncementsof the campaign's imminent end, while Wintgens's breakout to the west
looked preventableon a large-scalemap.
The CIGS began to entertain doubts about Hoskins's suitability to carry out the
duties of an independentcommander-in-chief, While the matter remains shrouded in
mystery and the views of the War Cabinet,of which Smutswas a member,are not known,
Robertsondid consult Smutson the Hoskins's performance. A glimpse of this comesin a
personalletter written to the CIGS by the SouthAfrican in mid-April:
feel confident that the stepswhich you have taken to supply Hoskins with

him
bring
to
the campaign
personnelandmaterial,shouldwhencompleted,enable
to an early conclusion...""'
Despitethis, discussionsin the War Cabinet led to the decision to replaceHoskins
in late April.

His replacement was a surprise. General van Deventer, formerly

commandingthe 2ndDivision, was chosendespitethe fact that the EAEF was now largely
an Imperial rather than a South African force. Smuts supportedhis selectionstrongly and
provided the necessarypolitical backing to the appointment. Whatever the background
negotiations,the changewas agreedby the War Cabinet on 23 April with the following
comments:
"The Chief of the Imperial General Staff statedthat for some time he had not felt
that the British operationsin East Africa were being carried out as satisfactorily as
could be desired. He had discussedthe matter on more than one occasion with
General Smuts, who while holding a high opinion of the Officer in Command,
Major-General A. R. Hoskins, had agreedthat apparently he had lost grip of the
operationsand perhapshad becometired."""
4" WO 95/5329, War Diary Norforce, Entry I June 1917;CAB 44/9, 42.
p.
MS Smuts,Box 100,13 April 1917,Letter Smutsto Robertson.
CAB 23/2, War Cabinet Meeting 124,23 April 1917,paragraph4.

142

The reasonfor the decision is unclear, and certainly by early May, Smutshad changedhis
opinion of Hoskins sharply:
"The progressof Wintgens in the direction of Taboramakesit clear to me that there
, A49

is no firm handling of the situationat presentin that theatre.

While he may well havehad genuinedoubtsaboutHoskins's performance,personal

in
Chapter
be
Six, the results of a
As
have
seen
will
animosity may also
played a part.
Court of Enquiry set up by Smuts in November 1916 were forwarded by Hoskins to the
War Office."" The findings reflected adversely on van Deventer, but also implicitly
is
little
Smuts's
There
force
the
the
command.
under
criticised
entire administration of
doubtthat Smutswas aggrievedby the findings while the whole tenor of Hoskins's various
have
finished
former's
directly
to
the
claim
off the campaign
requests
undermined
successfully.He later claimed:
"Further, in my opinion, the report is one-sidedand does not sufficiently consider
the campaignas a whole and the unforeseenphysical difficulties encounterednor
, A51

is there any allowance given for the results achievedby forced marches...

Hoskins himself receivedofficial notification in late April. After the manifold, but
asyet largely unpublicized,failures of Smuts'stenure and his own hard work, it was a very
bitter blow. While Hoskins acceptedhis supercessionprofessionally, the change was
widely resentedin East Africa, and many felt that an unsavoury political deal had been
behindit. As one seniorofficer put it:
"This seemsa more probable explanation- that it was the work of the politicians
and that the War Office was merely the conveyor of the order. Surely it can only
449MS Robertson,1133/4
8,2 May 1917,Letter Smutsto Robertson.
4" CO 551/101,Union of South Africa, 1917,Volume 8, Folio 38195,31 July 1917, 473. The report was formally signed off on 17
p.
May 1917,but the outline conclusionswere telegraphedto the War Office on 2 May 1917;CAB 23/2,2 May 1917,War Cabinet
Meeting128. The CIGS announcedthat they were still awaiting the findings of the board. Smutswas formally approachedat the end
of July, but it is not known whether there was any early and unofficial knowledge of the findings.
431CO 5511101,9August 1917,Letter Smutsto Sir GeorgeFiddes, 234.
p.

143

have been so, for such an action might well seemas nothing to the politician when
out to pleaseone of their own kind, though a graveinjustice to a regular soldier."452
Interestingly, even the Germanhistorian of the campaignsympathisedwith
Hoskins's ill-fortune.

453

452Fendall, TheEast African Force, 101.

p.
453MS Boell, N 14/32,28. Kapitel, 2245.
p.

144

CHAPTER 5- SUPPLY AND TRANSPORTATION

GENERAL

If there was one consideration that influenced the conduct of the East African
campaign above all others it was supply and transport, part of what is now called
logistics. The need to keep the troops, followers and animals of the respective forces
adequatelywatered, fed, armed and equippedwas paramount. Yet, while the tribulations
of each are amply described in the various memoirs and the official accounts, the
secondary literature tends to ignore this unglamorous but essential facet of warfare.
Indeed, many of the tactical and operational movements can only be understoodwhen
in
desire
hunger
is
desperate
it
to
thirst
the
and,
a
prevent
or
supply considered; was often
decisions.
led
bay,
instances,
hold
that
to
to
commanders'
number of
starvation at
Ironically, despite having the full support of the respective civil governments and the
settler populations, both sides suffered tremendously from insufficient diet and lack of
water. Hunger was constant and malnutrition was commonplace for much of the
both
followers
the
the
sides.
of
campaign,while
suffering was probably worst amongst
Certainly, the health of many of those who participated in the fighting was adversely
affectedby the conditions under which they served.
Supply and transportation in East Africa were extremely difficult compared to
other theatres. The first factor was that the areasof conflict were largely undeveloped
and often thinly populated. The farming was limited to subsistencelevel for most of the
African population, while the relatively few European-ownedplantations were geared
towards producing cash crops suitable for export. It was not possible to support the prewar settler population without imports, especially of Europeangoods, and the influx of a
large number of soldiers, together with the disruption and destruction causedby the war,
reducedthis ability even further. This was exacerbatedby the need to concentratelarge
forces,often in areasof little food production, and to maintain them for extendedperiods.
The armies and followers literally ate all of the local areas' food in short order and a
systemof supply depotsand magazineswas necessary;operationswere frequently based
145

454
It
depot.
from
be
force
the
days'
was a problem
the
would
marches
on the numbersof
very different from the Western Front and one that any eighteenth century commander
would have instantly recognized.
With a number of notable exceptions, the land being fought over was not
the
lack
fertile.
to
The
with
crops
coupled
cultivate
machinery
of
particularly rich or
from
farms
be
to
the
food
that
only
moved
often
could
meant
primitive network of roads
depotson the headsof porters. Even then, this took considerabletime and wastagewas
forward
depots,
From
the
to
the
the
in
move
considerable,especially the rainy seasons.
troops had to be organised and more porters would be required. To compound the
lines
the
food
troops
those
the
on
of communications
the
and
problem,
porters carrying
had to live off the suppliesbeing brought forward for the fighting units. The strengthof
the front line was based on a careful balance between having sufficient supply staff to
bring up the food and not having too many non-combatantsforward and consuming the

samestocks.
The situation was worsenedby rinderpest and the encroachmentof the tsetsefly
throughout large portions of the Germancolony. This vector, and the tryponasmiasisthat
it carried, quickly causedthe death of all domestic farm animals, especially the cattle so
important to local wealth 455This influx had beentaking place over a considerableperiod
.
less
had
to
inhabitants
time
migrated away
of such affected areas
of
and many of the
hostile places and their farms quickly reverted to bush. Depopulation also had a more
human causeas large numbersof the indigenous locals had been killed or starved during
the Maji-Maji rebellion of 1905-06and its aftermath. Germanferocity in suppressingthe
in
farms
the
destruction
particularly
and villages,
of many
uprising and the subsequent
least
75,000 people; certainly
death
had
the
the
south of
or migration of at
colony,
seen
456
in
10 yearslater the many of the affected areashad not recovered any meaningful way.

"4 Lettow, Reminiscences,makesfrequent referencesto theseconsiderations. Similar quotationscan be found throughout the
unpublishedBritish war diaries in the narrative chapters.
455Blenkinsop and Rainey, Veterinary Services,pp. 407-408.
4-4Koponen,Developmentfor Exploitation, pp. 597-599. Someestimatesgo as high as 250,000 to 300,000.

146

If lack of sufficient food supplies was a major problem, then obtaining sufficient
high
having
Despite
averagerainfall, most
a
water was often an even greaternightmare.
in
the
for
torrential
dry
time;
the
German
East
Africa
wet
rainfall
of
much
was very
of
the
dry.
While
in
is
followed
by
the
could
supply
major
rivers
marked aridity
seasons
holes
Water
forces,
large
they
suitable
away.
moved
as
problems aroseas soon
needsof
for villages of 50 people or so were quickly exhaustedwhen a force of 1,000 or more
together with pack animals descendedupon them. These troops would generally be
desperate
in
hot
the
labour
hard
from
hours
the
often
was
sun;
physical
parched
many
best way of describing their condition. Animals required even larger quantities of water
frequently
before
had
be
looked
the
troops;
to
supplies were unpalatable or
and
after
brackish which did little for health.457 Furthermore, lack of specialist personnel for the
458
did
All
to
the
survive on
could and
treatmentand purification of water added
problem.
fatal
invariably
lack
but
for
the
within
was
of water
reducedrations
considerableperiods,
a very short time.
In contrast, on the Western Front, the forces were highly concentrated and
links.
There,
by
intricate
an army was seldommore
supported an
network of road and rail
than ten miles distant from the nearest railhead, and light railways were frequently
"9
in
fought
divisions
brigades
Corps,
forward
close
to
and
pushed up
gun positions.
Being
thirty
than
to
over
miles.
out
more
spread
proximity
each other and were seldom
in the main theatre,they also receivedthe highest priority for storesand equipmentwhile
460

level.
highest
dealt
failings
in
1915
1916
the
political
at
with
administrative
where
and
Even in a campaign not noted for its standardof administration, such as Mesopotamia,
the advancefollowed a navigable river back to a seaport. In Egypt and Palestine,the
advancewas basedon the Suez Canal and Port Said and an enormouseffort was made to
build the water pipelines and railways necessaryto support a large force in the desert.461

'" Blenkinsop and Rainey, VeterinaryServices,p. 4 10.

4511
CO 691/19,Pike Report, p. 49.
459Brown, Logistics on the WesternFront, pp. 158-159and 162.
460Adams,RJQ, Arms and the Wizard Lloyd Georgeandthe Ministry ofMunitions 1915-1916,London: Cassell& Co, 1978,pp. 3537; Brown, Logistics on the WesternFront, p. 142. Lloyd George's appointmentof Sir Eric Geddesto run the BEF's supply system
testifies to this.

147

SUPPLY

All armies of the period had distinct supply problems related to the race or
backgroundof the forces involved. The bulk of soldiers and followers were Africans
NCOs
homes,
the
diets
to
their
and civil
to
officers,
while
who were used
particular
food
largely
European.
This
that
separate
was essential, as
meant
government were
food
European
the
style
necessary.
other's unless absolutely
neither group would eat
but
first
from
imported
Germans
to
the
the
suffer
shortfalls,
were
so
was almost all
basis.
There
British
was
that
the
short
on
a
regular
also went
shipping restrictions meant
had
feeding:
British
to
forces
the
that
supply
the
needed
also
complication of outside
Indian troops with food appropriate to their religion and caste, West Africans were
different from East Africans, and the South Africans were used to different things. The
followers presentedespecialproblems as they were often pressedmen, unusedto travel
drawn
from
The
the tropical areasof
to
to
men
and unable adapt changedconditions.
Uganda and around the Lake Victoria area lived largely on a banana-baseddiet and
462
diet.
Understanding
from
disorders
millet-based
or
maize
a
suffered gastro-intestinal
of diet was frequently poor and the British, in particular, made a number of major errors
in setting up the rations of those who fought for them. Settler prejudice and ignorance
inability
African
diet
to
in
the
troops
the
of
while
of some
played a significant role
poor
supply the correct types of food or in sufficient quantities was linked to the weak system
463
of supply. However, there were deeperproblems, as SurgeonGeneralPike pointed out
in his report:
"Let it be clearly understoodthat we are not now speakingof the failure of food to
initial
but
due
the
to
to
transport,
the
troops
reach
or carriers, or any shortcomings
deficient
Supply
faulty
in
laying
down
the
or
error
scales physiologically
...
Department should not operate without scientific advice. If this were the rule
scale would not be issued showing substitutes which physiologically are not
461Murray, LieutenantGeneralSir A J, Sir Archibald Murray's Despatches,London: JM Dent& Sons, 1920,pp. 188-205. These
annexesdetail the huge logistical effort, for example,on the Kantara East line alone, over 1,300enginesand waggonswere supplied
by Britain and Egypt.
462CO 691/19,Pike Report, pp 56-57.
46'Hodges,The Carrier Corps, pp. 119-123.

148

from
dietaries
be
drawn
up
which are excluded
substitutes at all, nor would
life
health...
9,464
to
and
substancesessential
Specific points included the issueof variable quality meal to the carriers coupled with the
fact that it was frequently impossible for them to carry both their full load and their
authorisedrations. It was also noted that the Cape"Boy" (mixed race) rations had no fats
and insufficient vegetableswhile the West African and West Indian Regiment rations had
465
no vegetablesat all.
TRANSPORTATION - GENERAL
The poor state of tracks and roads made the problems of supply much more
difficult when the campaign moved into the more remote areas. The British had solved
much of the water problem through the construction of pipelines and railway cars whilst
still back in their own colony, but this was never a realistic option once the advance
began. Shortages of materials, trained labourers and the means of moving heavy
materials precluded such construction in all but the most developedbases. Even where
this was attempted, such as at Kilwa or Lindi, the movement of the columns quickly
outstripped their means of transporting goods. The Germans,cut off from the outside
world, never had that alternative and relied on mobility and an extensivesupply systemto
keep forces in the field. However, it is also clear that they suffered heavily from the
466
British
sameproblems as the
and their allies.

ColonelOrr, formerly Hoskins'sDA&QMG at Vt Division, and subsequently

a
laterwrote:
columncommander
of somesuccess,
"... The transport had broken down by the end of May. Throughout the June
advanceand the August advanceand even during the July halt, except for a few
days, the transport had not been able to deliver more than half the needsof the
4" CO 691/19,Pike Report, p. 56.
463CO 691/19,Pike Report, pp 59-60.

149

long
before
had
have
it
lost
the troops
had
The
transport
troops...
any elasticity may
A67
Central
Railway.
reachedthe
This was not an isolatedview and variantswere widespreadamongstthe GHQ staff
in
Colonel
CP
Fendall,
General,
fighting
formations.
Adjutant
The
Assistant
the
noted
and
his diary that:
"... In the conduct of the campaignthe fault has been movement without sufficient
blame.
been
They
has
to
The
were
much
not
administrative staff
preparation.
invariably
for
faced
called
more transport
which
constantly
with new conditions
than they had been able to anticipate from previous information. There had been
for
both
by
keeping
breakdowns,
to
the
them
men
causing great privations
serious
long periods on reduced rations, and depriving them of proper hospital
A68
it
because
therewas no transportto take up to them.
accommodation
SHIPPING

The Allied forces did have one major advantage in that the sea lines of
latter
This
the
the
throughout
to
them
part
of
campaign.
communication were open
meantthat troops, food, vehicles and storescould be moved from Europe, India or South
Africa relatively quickly and in bulk. Importantly, shipping could move unimpeded to
East Africa from either the Cape of Good Hope or the Suez Canal routes; external
resupply was largely denied to the Germansapart from two small blockade runners. The
chief limiting factors were the shortageof ships, particularly from January 1917 onward,
and insufficient developed ports at which to unload. Dar-es-Salaarnwas the principal
baseand by far the most developed,while Lindi and Kilwa were both relatively primitive
at the time of their capture by the British. Furthermore, the ports were often some
distancefrom the fighting and resupply had to be conductedover the difficult conditions
' MS Boell, N14/32,26. Kapitel, pp. 1990-1997;Boell, Die Operationen,pp. 294,299,318; Schnee,Deutch 0staftika, pp. 238 and
268-280; and Lettow, Reminiscences,pp. 173-179.
467Orr, Colonel G M. "Smuts v. Lettow: A Critical Phasein EastAfrica: August to September1916", TheArnW Quarterly, LXX,
(1925), p. 294.
4" CAB 45/44, Diary of Colonel CP Fendall, Entry for 4 February 1917,p. 70.

150

describedabove. Still, despiteall thesehandicaps,the possessionof seacommunications

by the Allies and the correspondinglack of external sourcesof resupply for the Germans
madethe ultimate captureof the colony possible.
RAILWAYS

The railways, both British and German,were the most efficient meansof moving
men and materials within the theatre of war. The Northern Railway connected the
Kilimanjaro area with the port of Tanga while the Central Railway linked the capital of
Dar-es-Salaamwith the interior right up to Lake Tanganyika."' While small and having
limited rolling stock, these lines made the conduct of large-scale operations possible.
Unfortunately, the paucity of connecting communications limited the supply value to a
few miles' radius from the major stations. Furthermore, while both made good use of
this medium in the early stages,the Germans lost theirs to the British advanceby the
summer of 1916 and General Smuts never appeared to have understood fully their
importance. Obsessedwith surrounding and cutting off his foe, he failed to appreciate
the needto restore first the Northern and later the Central Railways as a objective of the
utmost urgency after capturing their termini. His troops remained reliant for too long on
lengthy, slow and inefficient lines of communication rather than switching to a more
flexible systemof rail linked to seatravel. While it is true that the subsequentoperations
south of the Central Railway were also dependenton porter and light vehicular traffic, the
establishmentof a fully operational and effective railway would have greatly assistedthe
building up of the stocksessentialto a successfuloffensive. On other hand, van Deventer
extendedlight railways from Kilwa and Lindi as far as his resourceswould permit.
ANIMAL TRANSPORT

Like their medical counterparts,the veterinary services encounteredsignificant

problems through lack of planning and general ignorance of animal management
procedures.The South African mountedregimentscamein for particular criticism as they
"9 Woodhouse,Capt H I, "East Africa", Journal the UniledService Institution
of
oflndia, XLVL (1917), pp. 330 and 334.

151

lackedcentralisedproceduresand paid little heedto the veterinary officers and their advice.
This resultedin enormouswastagerates:for the period March 1916to January1917nearly
100%of animals employeddied. One causewas similar to that faced by humans;lack of
sufficient food in the forward areas. But, the overwhelming reasonwas trypanosomiasisas
carried by the notorious tsetse fly. VvUle generally fatal to animals, the infection and
effects of this diseasecould be amelioratedby careful proceduresand attention to local
conditions. However, this good advice was frequently ignored at great cost to subsequent
military operations. One exampleillustratesthe consequences
of good care:
"During October, 1916,1,720 horsesand mules were taken over at remount depots
and conductedforward by regimental personnel. The distancewas approximately
200 miles. One hundred and fifty-five animals were lost en route. [9 per cent]
During the same period, 2,528 horses and mules were conducted forward by
loss
the
the
remountpersonnelover
of a single animal.9A70
sameroute without
The losses with the mounted troops were enormous - in the period August to
October 1916,2 nd Division was issued with 4,470 horses and 1,831 mules while P
Division was given 2,168 horsesand 1,032 mules.471 Totalling some 9,501 animals, all
would perish before the end of the year. )While nothing could be done to cure
trypanosomiasis,proper managementand feeding could have extendedthe working lives of
the animals by as much as 50 per cent. As with the other administrative services,the
veterinarycorps had little influence at GeneralHeadquarterswith most of its advice being
ignored:

"during this period of campaign [March 1916 to January 1917] the commander...
in-chief and generalstaff were not in intimate touch with the Veterinary Services.
There was

no

veterinary staff

officer

in

the

field

with

general

headquarters combatingdiseasewas only half the battle. Bad animal management

...

47'BlenkinsopandRainey, VeterinaryServices, 412.

p.
471Blenkinsop and Rainey, Veterinary Services, 417-419
pp.
and 425.

152

Veterinary
but
the
be
the
of
power
were out of
represented
and starvation could
Service to remedy. -A72

MECHANICAL

TRANSPORT

Owing to their physical isolation from the homeland,mechanizationwas never an

option for the Germans. However, the British used the earliest cars and trucks to good
300
light
Ford
carrying
of
pounds of cargo,
cars, capable
advantagewhenever possible.
had an immenseadvantageover porters as they required only a driver and were capable
breakdown
the
did
to
The
on
subject
were
petrol,
require
cars
of much greater effort.
disease
fatigue
drivers
to
difficult
as the
the
and
tracks,
vulnerable
as
were
very
and
incapable
the
in
Furthermore,
the
negotiating
the
were
of
vehicles
season
porters.
rainy
flooded swampsor the surrounding mud and in the dry seasonthe collapse of the roads
into fine dust caused considerable difficulties.

However, the sheer efficiency of

mechanicalversus human transport soon made the motor vehicle an essentialpart of the
British supply system; it has been estimated that one lorry was the equivalent of 30
473
be
the
tracks
limits
The
the
cut,
their
could
which
usable
speed
at
porters.
use were
on
keeping
drivers,
the system working under the
and
provision of sufficient vehicles and
baleful influence of the climate and pestilence. In the final analysis, the motor vehicle
was a significant factor in keeping the British advance going, particularly when the
supply of porters withered in late 1917.474
As with the other areasof administration, there is considerableevidencethat the
British did not make best use of the available resources. From the onset of the advancein
EAEF,
1916,
MT
to
the
one was allocated to a naval
early
out of six
companiesprovided
kite balloon section, four were attachedto the artillery and only one was actually under
the control of the Director of Supply and Transport. The problem of ever-expanding
lines of communications also impacted heavily on the MT, as divisional or column

472Blenkinsop and Rainey, Veterinary Services,pp. 417418. The figure of 50 per cent was an estimategiven by veterinary officers.
473Hodges,The Carrier Corps, p. 57.
474WO 106/1490,MTHistory, Part III, SectionIV, p. 6. At least eight supply routes in excessof 100 miles were servicedby MT;
CAB 44/9, p. 67, footnote 1. By 20 May 1917,over 900 vehicles were in use in East Africa with a further 555 on order.

153

be
by
having
difference
to
fixed
the
had
met
transport generally
operating radius with
a
huge:
475
frequently
distances
the
This
were
the supporttransport.
was a major problem as
Route

Distance

Voi - Morogoro
Korogwe - Morogoro

500 miles

Kondoa Irangi - Kaijado

300 miles

Dodoma- Dabaga

170 miles

Kilossa - Mahenge

130 miles

Lindi - Newala
Lindi - Maundi

150 miles

Kilwa - Liwale

150 miles

150 miles

150 miles

.f10

It was not until Hoskins assumedcommandthat a thorough reorganisationof the

MT was undertaken. Recognising the inflexibility of the system, all vehicles were
the
to
the
column
of
size
according
allocated
and
control
gradually placed under central
to be supported. By Augus 1917, there was a single MT unit that supplied all aspectsof
the East African force and it was to be the mainstay of the organisation for the remainder
of the war.""
CARRIERS

If the railways provided the best meansof transport, local conditions dictated that
the armies had to depend on the least efficient method, human porterage. It was the
food
hiring
to
in
long-standing
tribes
carry
out
men
tradition
of
certain
adaptationof a
and storeson their headsfor safaris. The urgency of the war and the problems of moving
food and ammunition saw a radical and ruthless expansion of the labour force that was
length
the
Initially
the
of
unprecedented.
conditions of service,
supplied with volunteers,
The
led
large-scale
horrors
to
the
the
problem with
compulsion.
absenceand
of
unknown
475WO 106/1490,MT History Part 111,SectionIV, pp. 29-30 and p. 39. Theseroutes were not all in use simultaneously.
'76WO 106/1490,UT History Part III, SectionIV, pp. 6-7.
477WO 106/1490,MTHislory Part Iff, SectionIV, pp. 28-30.

154

food
in
labour
hard
the
involved
it
conditions,
that
arduous
physical
very
porteragewas
being
diseases
insufficient,
there
a
chance
of
a
very
real
was
were rife and
was often
battle casualty. The same diseasesand maladies that affected the soldiers were several
times more likely for the followers through a combination of poor diet, overwork, lack of
basic hygiene and ignorance.478 The question of porters becamea dominating factor for
the British and Belgians while it never lost its importancewith the Germans.
If getting the correct amount of manpower was essential, then keeping it fit and
healthy was equally important.

By the end of 1916, it was calculated 160,000 men had

been recruited in the northern British territories as well as in the German colony, of
479
died
field.
had
in
Some
5,349
63,000
the
while an
whom only some
remained working
been
had
30,000
deserters
the
26,318
remaining
or missing. astonishing
were counted as
discharged in line with their engagement 480 However, on taking stock of the situation
.
his
160,000
Hoskins
General
that
the
staff
calculated
and
after assuming
chief command,
carriers would be needed to keep the force going on the road; in recruiting terms this
meant an extra 16,000 carriers per month to keep up with the estimated wastage rate of
15 per cent;481 These enormous numbers also had to be found in direct competition for
carriers with the Belgians who needed 18,000, as well as recruits for the rapidly
expanding King's African Rifles.

Despite the best efforts of the colonial officials, who

exerted considerable pressure on the African

chieftains backed by a fair degree of

coercion, it was never possible to get sufficient numbers in time for the planned move
south. At the end of March 1917, the authorities had managed to come up with another
50,345 men and even by the end of June 1917 the total recruitment since the turning of
the year only reached 124,830.482

It should also be remembered that these figures only include the manpower
day
for
keep
its
limited
to
one
worth
necessary
a unit mobile with
own
reserves,usually
ammunition and two to three days' worth for food. The bringing forward of supplies
478CO 691/19, Pike Report, p. 6; CO 533/216, WatkinsReport, pp. 5-8 and 20-22; Hodges The Carrier Corps, pp. 126-128.
479CO 533/216, WatkinsReport, Appendix I to Table 2 (rotal Recruitments)and Appendix I to Table 3 (Numbers in the Field).
480CO 533/216, WatkinsReport, Appendix I to Table 4 (Deathsby Periods)and Appendix to Table 7 Crotal Desertedand Missing).
491CO 533/216, WatkinsReport, p6 and Hodges,p 54.
482CO 5331216,WatkinsReport, Appendix I to Table 2 (Total Recruitments).

155

from the rear required an elaboratesystem of its own that had to be independentof the
forward troops; the Belgians estimatedthat some 260,000 people were employed on war483
in
relatedtransportedwork their colony.
To give an example of the desperationof the situation, in late 1916 1" Division
demandedsubstantial increases in carriers in order to maintain the growing force at
Kilwa, while van Deventer's 2ndDivision neededan additional 4,000 just to increasethe
rations above half the entitlement that the troops were barely subsisting on. Despite the
seriousnessof the situation, neither demandcould be met as the main effort was directed
484
keeping
forward
Mgeta
A later staff paper
River
the
troops on the
at
supplied.
estimatedthe first and secondline requirementsas follows:

CampagnesColoniales Belges,1, p. 34.

4" CAB 44/6, pp. 6-7.

156

Formation

Battalions

Number

Mouths

Daily

to Feed

Weight

of

of

Carriers Needed

Food (lbs)
Iringa

4,865

9,370

10,568

Ruflj 1

4,534

9,066

10,568

Kilwa

16,851

33,702

16,851

4V2

10,374

20,748

8385

7540

Lindi
ForceReserve
Total

50,297

21V2

had
be
10,000
to
Thethird line requirements
that
added:
men
meant another
Force

Carriers

Iringa Force

3,244

Rufiji Force

2,500

Kilwa Force

1,800

Lindi Force

2,500

Total

10,044

THE LINES OF COMMUNICATION

If logistics appear an unglamorous subject, then its chief means of action, the
lines of communication, may be even less so. All forces require the meansof collecting
From
food,
from
theatre
to
the
their
of
war.
people,
and material
place of production
there,it is necessaryto have a systemthat can hold and sustainthem and then move them
forward to the required area of operationsat the appropriatetime. These functions must
be carried out regularly regardlessof the posture of the fighting forces, although pressure
usually increases during times of battle and its aftermath - people require feeding,
weaponsneedammunition, vehicles needfuel, and equipmentmay have to be replacedin

157

full or part more or less continuously. Add to this the need to bring in individual
battle,
for
the
while evacuating those same cases
of
sick and casualties
replacements
forward
the
for
of reinforcing
moving
plus
rearward
care as well as prisoners of war,
fully
lines
keeping
the
supplied, and a
those
communication
of
who operate
units and
huge responsibility emerges. For all of the combatants,these questions were hugely
difficult
fighting,
in
the
infrastructure
by
the
and
the
areasof movement
magnified
poor
insufficient
the
material means of carrying out the
generally
and pestilential climate, and
desired military operations. Improvisation and flexibility were the hallmarks of the
by
military and civilians alike.
suffering
period, although not without considerable
BRITISH

CASE

It may be useful to use illustrative examples from each of the major combatants to
highlight the problems that faced the opposing sides at a similar point in time, that of
September 1916. For the British, the 2nd Division had struggled to reach the thickly
General
direction
in
by
September
the
Kidodi
7
of
accordance with
vegetated area around
Smuts. This much reduced formation, which could barely muster 4,700 effectives, had
Irangi
Kondoa
its
from
to
and the
the
advance
earlier
never recovered
vicissitudes of
breakdown of supply that it entailed. However, by the time the short rains forced it to
halt opposite Abt Otto in September, it was really at the end of its tether. "" Sketch No A
forward
in
keeping
involved
distances
troops going.
the
the
shows graphically

What is

410
History
Official
Moshi;
distance
is
huge
the
to
miles
the
the
cites
striking
railhead at
while from the map it measures closer to 450 miles.

Apart from the difficulties

of

it
in
food,
fuel
this
primitive conditions,
way and
carrying
ammunition,
and supplies all
for
be
transport
types
the
made
at various points
of
should
noted that the need to change
inefficiency and invariably slowed down the flow of supplies, particularly when human
it
This
the
supply
system
as
porterage was required.
was not necessarily a criticism of
also had to contend with local road conditions; it just illustrates how precarious the
margin of existence was. It also shows how critical the reopening of the Central Railway

'" Hodges,
intheWatkins
1917
Papers.
25
February
The
Carrier
Corps,
2
General
Staff
209.
See
Appendix
of
cites
plan
p.
which
"6 GSWA, Box 26,17 September1916,Operation Report 2' Division,
Deventerto Smuts.
van

158

functioning
Dar-es-Salaam
British
to
the
as a port as quickly as
was and why
needed get
possible.
If the plan of operation seemedfamiliar, then so did the logistical situation. The
force
further
been
had
the
transport,
the
met,
of
main
east
allocation of motor
after
needs
was inadequatefor the situation. The poor and now deterioratingstateof the roadscaused
the breakdownrate to soar while local African impressedcarriers continued to desert as
quickly as they could. The divisional staff had madeevery effort to maximise its resources

into five stages,eachrelativelysuitedto

andhadbrokendownthe linesof communication
the modeof transportavailableandthe ground. The stageswere organisedas follows:

Start Point

Finish Point

Distance

Method of Transport

(miles)
Dodoma(Railhead) Kisomache

12

Carriersand Divisional
NIT

Kisomache

Temagwe

66

Heavy MT

Temagwe

Sadala(via Ruaha)

23

Light Cars

Sadala

16 Mile Post

17

Carriers

16

Light Cars

16 Mile Post

nga

This meantthat the total line of communicationfrom the railhead,which represented

the most economicmethod of transport,to the forward baseat Iringa was some 134 miles
and requiredthe crossingof numerousrivers. Furthermore,the needto exchangeloads at
five different points, while operationally essential, was wasteful and time-consuming.

Finally,it needsto be remembered

that hinga was about20 miles north-westof the start
point at Dabaga,with the main objectives another 20 miles ftuther on; for the flanking
487
doubled
by
long
the
the
columns movementof supplieswas probably
approachmarches.
Despite the hard work by the logistical staff and the supply units, there was no
possibleway of building up adequatestocks of food prior to the advance. Indeed, it was

159

y
W ES

f?
C614Aa, vICATIO-4

N*t oft

op
ilotift
sqA; -

lioN4

%-'-I;

-. --

N VOA

1,tApit
A1t

r*,

ksooor
6
O*rK L)
OR
k.

j
cl;

1000,

15 Ci01-

li

be
impossible
it
difficult
full
to
the
to
would
stood
and
situation
very
maintain
rations as
launch
his
in
full
forced
during
Van
Deventer
to
move
was
provide such
an advance.
knowledge that half rations for all troops would be the best that could be provided; he
be
fed
If
infortned
his
the
troops
this
could
not
accordingly.
command
acknowledged
and
be
is
instructive
It
followers
it
that
the
then
even
worse
off.
would
properly,
was a certainty
to note that, although van Deventer's division totalled about 5,000 fighting troops, the
carrierssupportingit reachedover 13,600. The poor weatherconditions,the hard work and
488
be
fast
desert
inadequate
diet convincedmany to
the
andthey could not replaced
enough.
BELGIAN CASE

TheBelgianssufferedfrom similarproblemsto the British, andin somewayshad

forces.
disadvantage
The
difficult
in
their
task
was that
greatest
a more
maintaining
Belgium was largely occupiedand most items had to be procuredin either Britain or
Franceprior to being shippedto Africa. Therewerethreesearoutes:one leadingfrom
La Pallice to the river port of Matadi on the River Congo, another leading from
Marseillesto MombasaandthencethroughBritish EastAfrica, anda third runningfrom
Londonto CapeTown 489The first routewasnaturallypreferableas its land portionran
.
throughBelgianterritory,but it wasfar from ideal. Arriving materialhadto be offloaded
the ocean-goingtransportat Matadi and transferredto the railway. The next stop was
Ldopoldvillewherethe rail line endedand a further transshipmentto boatstook place.
The cargothen proceededup the River Congoto Stanleyvillewhere it was separated
accordingto its type andultimatedestination.There,the river hadtwo largeunnavigable
stretchesthat wereservedby railway lines,which of courseentailedfurther loadingand
unloading.Onceon LakeTanganyika,steamerscouldbe usedto movestoresto Ujiji, the
westernterminusof the CentralRailway. Fromtherethe rail line ran to the mainbaseat
Dodoma,whereeverythingwasoffloadedandmovedto the front by MT andcarriers.

'27

CAB 44n,
and footnote.
"a GSWA, Box 29,7 December 1916,TelegramP 13,2 Div to BGGS; 10 December1916,TelegramAA & QMG to GSO 1; CAB
44/7, pp. 11-12 and footnote. Some 800 porters desertedin the first 10 days of December,leaving a shortfall of over 3,000.
489CampagnesColonjales Belges, I, p. 29. SeeMap I for overview
and Map 2 for more detailed routes.

P.10

160

The chief difficulty in the forward areaswas that they were largely uncultivated
or infertile and that they had been systematically emptied by the Germans. The British
had agreed to supply food up to the railhead at Dodoma, but the Belgians were
responsiblefor moving it forward to the front line troops. The Belgian Commander-inChief estimated that some 18,000 carriers would be needed, of which 5,000 had been
loanedto the British but were due for return.
Given the practical difficulties of identifying, moving and organising these
dispersedindividuals, the Belgians agreedto recruit a similar number in British occupied
territory. From 27 May to the end of July, this 5,865 carriers, leaving a further 13,000
outstanding. Only 10,000 could be produced from the Belgian-occupied territory and in
June authorisation was given to recruit 3,000 carriers from Katanga. These measures
were vitiated by the fact that many of the 4,000 carriers being usedto support the hunt for
Naumann were now incapable of further effort. Originally from the mountainousareas,
they found the differing climatic conditions of the plains difficult to cope with and were
susceptibleto disease. This, plus desertion,led to numerousshortfalls and also to marked
lossesof clothing and equipment.490
Representations by Huyghd to the Belgian Goverment were answered by
permission to enlist a further 10,000 carriers from the Belgian Congo. Again this was
only accomplishedwith great difficulty over the following five months. He had even
more problems in obtaining European personnel and the combat units were so undermannedby Belgians that it was impossible to provide supervisorsfor the carriers. In the
end, the British were forced to supply officers to run the most of the Belgian columns'
carriers. It was to mean a difficult campaignwith the supply of food always perilous.
GERMAN

CASE

From the German point of view, the problems of supply were quite different, but
equally as challenging. While they did not have to worry about ever lengthening lines of
490CampagnesColoniaks Belges,111, 136-137.
pp.

161

communication, indeed theirs were shrinking, their means of transport were very much
few
fly,
few
few
limited.
from
Apart
tsetse
pack animals, they
and
with
of
areas
a
more
failed
British
Although
human
limited
the
to close
to
muscle power.
were essentially
forced
Schuatruppe,
destroy
fighting
their
the
the
rapid
advance
often
with and
power of
the Germansto abandonvital stocks of food as they simply not could be carried away in
Insufficient porters hampered their efforts and there was a constant tension
betweenkeeping the various Abteilungen fully mobile and keeping the all-important lines

time.

deal
On
this
functioning.
of
great
caused a
a number of occasions
of communication
friction betweenthe civil and military authorities. The Germans also had the particular
farming
it
from
it
food,
harvesting
the
then
and
moving
problem of growing sufficient
""
lines
This
their
to
the
strain
on
system.
put additional
areas
of communication.
The Germans operated a series of depots, known as magazines, under the
command of an Etappenleiter, whose job it was to make sure that the correct supplies
went to the right location.

This task was made more difficult

by the poor

its
Railway
Central
between
the
and
theAteilungen,
communications
particularly once
telegraphsystemhad been lost. In September1916, the systemwas naturally focusedon
the main force which was operating on the line of the Mgeta River; a significant
detachmentwas further to the eastin the areaof the lower Rufiji valley and moving south
from Dar-es-Salaam. The system was based on the magazine at Kunglio on the Rufiji
which kept the forward troops supplied. The river could be used only with great
difficulty and certainly not in the rainy seasonas the currents were simply too strong, but
it did provide a link to the other major supply centre at Utete, further downstream. The
remainderof the links were basedon improved tracks; the main route ran southwardsto
Maba, it branched north and westward to Luwegu and thence to Mahenge; the main
routed continued south to Liwale. Liwale was a major centre of food production and
tracks radiated in all directions, connecting Songeain the west, Tunduru in the south, and
Massassiin the south-east,with Lindi and Kilwa in the west. The route was by no means
easy to traverse, although a number of improvements had been made and the journey
from Lindi to the Rufiji took severalweeks. Nevertheless,the possessionof interior lines
491Lettow, Reminiscences,pp. 174-178and 234-235.

162

he
Lettow's
task
of communicationmade von
significantly easierand took great pains to
ensurethat it remainedintact for as long as possible.
After sowing crops on the lower Rufiji in November 1916, the potential for
reasonablyadequatefood suppliesexisted."" This would be a vital sourceas it enableda
large element of the force to avoid drawing on the Etappenleitung for several months
while producing an extra 300 000 kg for the stocks. An example of the tensions in the
systemcomesfrom the Lindi-Bezirk in early November, in which the Etappenkommando
Liwale reportedthe movementof 450 - 500 tons of food as well as expecting to extract at
least 150 tons from native supplies during the January harvest. In contrast, the local
commander,Looff, signalled on 9 Decemberthat the removal of food from his areawas
493
causingseveresupply problems. In reply, he was told to live off the land and that the
expected150 tons must be kept in reserve.
The replacement of food was a constant concern as the Rufiji force alone
consumed about 10,000 kg of cereals per day supplemented by local foraging and
hunting. With the harvest of corn in March 1917, the next available crop would be the
mtarna in June; until then at least 900,000 kg would be needed.

But there was a

significant shortfall in stocks with only 600 000 kg held; the magazinesalong the Rufiji
had 150 000 kg, the Etappenleitung held another 150 000 kg, the magazinesin LiwaleNanganostoredanother 150 000 kg, leaving 150 000 kg in Lindi. 494
Von Lettow kept almost daily records of his supply situation. Despite his care, he
had the unpleasant surprise of discovering that the Etappenleitung held only 175 000 kg
of the expected 300 000 kg in its stores. This shortfall was due to wastage and having to
feed 20 000 rather than the planned 15 000 people. 495 In order to survive, on 26 January,
von Lettow ordered drastic cuts, with the daily ration being reduced from 750g to 600g
(rice from 500g to 400g) and meat being produced by hunting. Several days later, he
decided on large reductions in numbers, with over 8,000 Africans
492
MS
Boell,
N14/30,26.
Kapitel,
1967
1981-1983.
pp.
and
493MS Boell, N 14/30,26. Kapitel, 1988.
p.
494MS Boell, N14/30,26. Kapitel, 1990.
p.

163

followers being

0 i=

#
)
k

"VI C-47i t,

01
0

lqt
11

Jj

.01

-1
71*
.

%
".

0,

..

.3

1%

--

t-

vw,

4e

1
I

-.

.-%

. ..C4-0

,.
0,
10

.8
's

it I

41k

Ih

/63&-

dismissedwhile formations were ordered not to feed the 500 wives and boys any further.
The cuts fell on both the Feldtruppe and the Etappenleitung who were ordered to shed
Henceforth,
4,000
2,500
companies were capped at
posts respectively.
and
some
496
165
carriers each.
maximum of
Owing to a favourable harvest, conditions easedin early 1917 and rations were
be
described
than
but
5OOg
they
700g
as
more
could
never
to
rice,
raised
meal and
497
labour
that all were enduring.
adequategiven the conditions and hard physical
INFLUENCE OF SUPPLY AND TRANSPORT ON OPERATIONAL

DECISION-

MAKING

There is no doubt that the limitations of supply greatly influenced the decisionsof
from
Entente
Germans
the
both
The
suffered
the commandersof
as much as
sides.
had
Smuts
food
Although
tried to overcomethe problems of
severe
and water shortages.
his
impossible
troops
it
task
force
suffered
and
through
an
was
supply
of willpower,
heavily. After his departure,many of both sides' tactical and operational deployments
for
The
food
largely
based
adequateresources
need
and water.
were
on considerationsof
in
late
1916
Rufiji
River
lower
hold
decision
behind
Lettow's
the
and early
to
area
was
1917. Smutsfailed to realise this and ignored the area,enabling the Germansto replenish
498
far-ranging
food
Another
the
their
effect of
of
example
stocks at a critical period.
German
hold
decision
the
the
to
colony
the
of
southern
portion
supply on operationswas
Lettow's
In
into
East
Africa.
Portuguese
to
own words:
and range

"The operationsof the last few months had narrowedthe area from which
Lupembe,
The
be
for
the
troops
areas
of
productive
obtained.
could
supplies
Iringa, Kissaki and the lower Rufiji had been lost, and the newly-occupied
districts includedwide stretchesof barrenland. The productivity of the more
fertile areaswas for the most part unknown At that time I only had a general
...
493MSBoell, N14/30,26. Kapitel, p. 1991.Some 6,800 belongedto the Etappenleitung and 2,200 to the Feldintendantur.
496MS Bocll, N 14/30,26. Kapitel, pp. 1995-1996.
497MSBoell,N14/30,26. Kapitel, p. 2001; Lettow. Reminiscences,pp. 176-178.

164

idea that the eastern part of the Lindi area was very fertile and known as the
granary of the colony. 9A99

This understanding led to the despatch of Abt von Stuemer into Portuguese
territory in early 1917to harvest its food resourceswhile shortagesat Mahengeled to the
dispersal of several detachmentsto the south in order to lessen the strain on the hard
pressedlines of communication.

On the otherside,van Deventer'sadvanceon the Lukuledi hadto stopon several

occasionsowing to lack of food, while Norforce's clearanceof PortugueseEast Africa in
early 1917 was specifically intended to deny the enemy resources. Similarly, the
withdrawal of the Belgians from Mahenge in late 1917 was aimed at preventing
starvation and needlesssuffering. All had learnt the vital importance of maintaining a
functional supply and transport systemin place.

4" Lettow, Reminiscences, 160

pp,
and 178.
'" Lettow, Reminiscences, 190.
p.

165

CHAPTER 6- DISEASE AND MEDICINE

INTRODUCTION

The East African campaignwas quite unlike its contemporariesin the First World
War in many ways, but the one factor that made it notably so was sickness. Armies have
in
disease,
depredations
from
the tropics with their
the
particularly
of
always suffered
debilitating effects of sun, heat, and indigenous pestilence. The nineteenth century
in
dangers
demonstrated
had
in
imperial
Africa
the
of soldiering
campaignsof
conquest
by
However,
high
through
mortality and sick rates.
such unhealthy climates
consistently
the turn of the century, advancesin medicine and sanitation had reducedthe perils of the
Soo
levels.
continentto much more acceptable
If much progresshad beenmade in the application of medical scienceto military
in
its
low-lying
East
Africa,
deal
be
done
to
and
particularly
affairs, a great
as
remained
The
diseases
host
tropical
to
and
afflictions.
of
swampy regions, was
a wide variety
exigenciesof war and the difficulties of resupply made a complex medical situation even
be
for
both
field
To
this
the
to
the
must
campaign
sides.
worse and added
problems of
added the weakening effects of inadequate diet, prolonged exposure to the elements,
insufficient protective clothing and a shortageof effective medicines. Furthermore,many
on both sides failed to understand the importance of proper hygienic and sanitary
followers
hugely
detrimental
to
the
soldiers
and
measureswith
well-being of
effects
alike. In such circumstances,the high rates of malaria, dysentery, typhoid, pneumonia,
blackwater fever, and tuberculosis were unsurprising. Similarly, the less dangerous,but
help
bees
did
little
flies,
fleas,
leeches,
to
ticks,
ants and
equally unpleasant,effects of
501
health.
either morale or

-" Curtin, Philip. Diseaseand Empire 71e Health ofEuropean Troops in the ConquestofAftica, Cambridge:Cambridge University
Press,1998,pp. ix-xi. Henceforth, Curtin, Diseaseand Empire.
'0' CO 551/101,Union of South Africa, 1917,Volume 8, Folio 38195,31 July 1917,Court of Enquiry, Findings of the Court
pp. 474475. The entire text of the Court togetherwith the evidencesubmitted in containedin this folio - it is a bulky documentof over 200
page;Henceforth,CO 551/101, Court ofEnquiry; CO 691/10,Pike Report, pp 87-88;

166

While the transmissionof tropical diseasesthrough such vectors as the anopheles

mosquito and the tsetsefly could not be prevented,there were a number of prophylactic
The
level
implemented
be
the
to
suffering.
that
scale
and
of
reduce
measures
could
regular administration of quinine, the use of mosquito nets, the correct siting of camps,
502
in
However,
the
had
been
the
past.
effective
proven
spraying and other measures
implementation of preventative actions under difficult wartime conditions required firm
discipline and material support.

Discipline was required to ensure that medical

followed
by
hungry
and
exhausted,
were
precautions and other preventative measures
debilitated troops. Efficient organisation was essential in providing the necessary
long
disease.
difficult
In
fight
for
the
and
a
against
medicines and material
successful
drawn-out campaign, fought in virgin bush many miles away from the nearest supply
bases, both of these requirements were frequently either ignored or simply impossible to
503
carry out.

The vicissitudes of the British forces are by far the best documentedin both the
failings,
it
from
literature.
While
they
a number of
clearly suffered
popular and scholarly
is also apparentthat the other protagonistshad an equally tough time in coping with the
have
Germans
but
Portuguese
is
to
The
fragmentary,
the
seem
and
climate.
evidence still
had even higher levels of sicknessthan the British. As well, there appearto have been
marked differences between the types and rates of morbidity of Africans and Europeans
as well as soldiers and followers. Certainly all participants,military and civilian, suffered
immenselyand many left the conflict with their health permanentlyimpaired.
THE BRITISH EXPERIENCE

The British medical experience in East Africa was not a happy one, as it was
markedby significant failures in both administration and command. Overall, it proved to
be the most unhealthy theatre of the war and sickness,particularly of malaria, reached
50' Harrison, Mark, "Medicine and the Culture of Command: the Case of Malaria Control in the British Army during the two World
Wars", Medical History, X1, (1996), pp. 441-442.
'0' GG/9/96/18,21 January 1916, Letter Bonar Law to Buxton. This letter
enclosed an article in the British Medical Journal by Sir P
Manson on the measures required to ensure health in East Africa, especially precautions against malaria. A copy was almost certainly
passed to Botha and Smuts; Secretary ofDefence, Group 2, Box 839. Copies ofthe articles were also sent to the Defence Department.

167

levels
the
high
levels.
the
While
campaignsof
of
approached
never
rates
very
mortality
the early and mid-nineteenth century, the wastageof manpower was enormousand was,
in many ways, excessive. The main reasons were the high command's lack of
understandingof the need to enforce prophylactic and sanitary measurestogether with a
However,
the medical
the
command.
of
chain
own
sidelining of
medical service's
determined
the
a
more
energetic
and
as
responsibility,
servicescannot escapea share of
brought
better
have
In
the end,
to
the
well
results.
could
approach
problems encountered
it was this substantialfailure that led to widespreadsuffering and death.
BRITISH MEDICAL ORGANISATION

The medical services in the East African Expeditionary Force followed standard
British Army organisationand structures.At its headwas the Director of Medical Services
(DMS), usually a surgeon general, supportedby one or more Assistant DMS (ADMS).
Next in the hierarchy was the Deputy DMS (DDMS), a title given to the doctors running
hospitals and the line of communication medical staff. At division level, the principal
504
General,
Coming
Adjutant
these
ADMS.
the
the
title
under
medical officer used
of
officers were responsiblefor the efficient running of medical units, treatmentof the sick
and wounded and advice to their respective operational commanderson all aspectsof
medicalmatters.
The systemwas basedon a hierarchy of units designedto treat casualties,sick or
wounded, at the appropriate level. This began with the regimental medical officer, a
505
qualified doctor, and a number of stretcher bearers who were trained in first aid.
Woundedsoldierswere usually given first aid by their comrades,before being carriedback
by
to the regimentalaid post (RAP) by the stretcherbearers. There,they would be assessed
the medical officer and given limited treatment before being sent back to the field
ambulance.
5" Macpherson,Maj Gen Sir W 0, History of the Great War Basedon Official Documents Medical ServicesGeneral History
Volume1. Medical Servicesin the United Kingdom; in British Garrisons Overseas;and during Operationsagainst Isingtau, in
Togolan4 the Cameroons.and South-WestAftica, London: HMSO, 1921, p. 3. Henceforth Macpherson,Official Histo?y Medical
ServicesVolume1.
"5 Whitehead,Doctors in the Great War, London: Leo Cooper, 1999,p. 181.

168

The field ambulancewas meant to be a mobile medical formation and there were
brigade.
divided
into
It
be
division,
two
to
three
was
supposed
usually meaningone per
per
groups; the bearer division and the tent division. The former provided stretcher-bearers
who collected the wounded while the latter had the advanceddressing station and ten
ambulancesfor their treatment. Operationally,the field ambulancebroke down into three
sections,each with a third of the bearer and tent divisions' resources,that could operate
behind the forward units. A section could treat up to 50 wounded or sick, giving the field
506
150
ambulancea total of
places.
The field ambulancehad three major roles; the first was to relieve the RAPs of their
sick and wounded; the second was to assist the regimental personnel in clearing the
wounded from the battlefield, usually during pausesor at night; and the third was to
for
be
to
that
they
treatment
to
able
provide sufficient
were
evacuated more
casualtiesso
507
from
battlefield.
The next stagein the medical chain was
the
extensivetreatmentaway
the clearinghospital which was meantto bridge the gap betweenthe field divisions and the
lines of communication. The clearing hospital was meant to have 200 modified stretcher
beds,but beganthe war establishedfor neither ambulancesnor nurses.508 This, and their
lack of a significant surgical capability were major limitations in their effectiveness,
although as elsewherethese hospitals evolved into the better-known casualty clearing
stations (CCS). Unlike the Western Front, where the static nature of fighting and
widespreadpresenceof rail lines helped to overcome their limited mobility, the CCS in
EastAfrica had a much more difficult task owing to insufficient transport.
Two other types of unit supportedthe forward troops; the stationary and general
hospitals.The latter was intended to be a light, mobile hospital capable of treating 200
patients and as a link between the clearing hospitals and the base. However, improved
methodsof evacuation,notably the motor ambulanceconvoy, madethis role redundantand
over time the stationaryhospital had its capacity increasedto 500 casesand assumedthe
samefunctions as the general hospital. Finally, on the lines of communication,was the
5'6Macpherson,Official History Medical Services Volume1.
pp. 8-9.
Whitehead,Doctors in the Great War, p. 189.
Macpherson,Official History Medical Services Volume1, pp. 10-11.

169

likely
deal
types
designed
hospital.
to
This
of sicknessand
all
with
unit
was
general
divisions
between
in
capable of
and
surgical
medical
wounds the army and was split
509
500
treatingup to
casualties.
One unit that came to particular prominence during the war was the motor
implied,
to
this
As
(MAC).
move
the
used
mobile
unit
was
name
convoy
ambulance
While
MACs
in
the
between
were a great
treatment
vehicles.
units specialised
patients
in
they
were
short supply and
advanceon animal-drawnambulancesand stretcher-bearers,
limited
by the road and weatherconditions
were
EFFECTS OF DISEASE AND MISMANAGEMENT

There is little doubt about the debilitating and often fatal effects of tropical
illnessesas they figure prominently throughout accountsof the campaign. GeneralSmuts
himself was well awareof the unhealthiness,of the country and its effect on the escalating
River
Mgeta:
he
the
of
south
sick rate, particularly as contemplatedplunging
"... According to Medical report about 70% of VAN DEVENTERS Division are
Debility.
Malaria,
Dysentery
for
to
of
and
effect
combined
unfit
work owing
Position in two other Divisions not much better... Above numbers totalling 8000
be
further
5000
large
4000
the
too
to
reduced
of
enemy
and
will
or
are none
catch
by sicknessin deadly RUFIJI area..."510
Operationsin someof the most unhealthy regions of the world could never be free
from high ratesof sickness,given the stateof medicine at the time, but Smuts's telegrams
to the War Office always rested on the implicit assumptionthat the climate was solely to
blame. Unsurprisingly, they never touched on the manifold weaknessesof his own
does
It
the
that
not seem to be a
administrative arrangements
problems.
exacerbated
questionof awarenessof the number of casualties,which were well known, but rather the
inability to prevent them. As the previous chapter has pointed out, the lack of thorough
" Whitehead,Doctors in the Great War, pp.208-209.
310WO 95/5291,War Diary GHQ, 22 September1916,Appendix 32a, Telegram OA 412, Smutsto CIGS, 22 September.

170

and well thought out plans for supporting the advancing troops with sufficient food,
clothing, protective equipment contributed materially to the poor state of health by
September1916.

In the desire to wrap up the campaign quickly, Smuts and his staff ignored the
obvious consequencesof driving an already exhausted and fever-ridden force well
beyond the existing, inadequate supply lines into a primeval wilderness. The Rufiji
Valley was well known to be both dangerousto health and difficult to move through, yet
the advancesouth was continued in Septemberand October. The onset of the short rainy
seasonin the latter month forced a halt to operations, but with the troops in low-lying,
swampy and highly malarious areas. The result was to increase the level of sick
casualtiesuntil many units ceasedto exist as fighting entities and the remaining troops
were incapableof any serious effort. Concernedabout the disappearanceof much of his
force, particularly the South Africans, in October Smuts ordered a medical board to
assessthe worst-affected sick cases. The results were dire, as over 12,000 soldiers were
consideredunfit and invalided home on medical grounds. The problem was amplified by
the inability of the servicesto evacuatemany of the casesin a timely or efficient matter;
the collapse of the transport and the effects of the rains on the primitive road network
causedchaos.
His principal staff officer sent an urgent telegram to the administrative staff in
early November about the need for efficient evacuationand treatment,but there was little
that could be done in the circumstances.511 It took considerableeffort and time to move
the sick back to the railway, which still was not operating at anything like its capacity. In
the meantime, the remaining soldiers, many of whom should have been hospitalised
themselves, had to shoulder the absent troops' workload, making for a particularly
unpleasantand vicious circle.
It is worth consideringthe factors that allowed this to happen. While sicknesshad
beena significant concern during the first year and a half
of the campaign, it had never
511WO 95/5300, War Diary DMS, 3 November 1916,Appendix 111,Telegram
QC 14, BGGS to Adminstaff.

171

fighting
the
and a reasonable
the
to
of
relatively static nature
grown out of control owing
in
March
1916
launching
However,
the
warfare
of
mobile
with
systemof administration.
into
heavily
Africa
East
British
the
vegetated
the
the
and
advanceout of
and steppesof
force
habitat
German
the
lowlands
the
of the
the
entered
colony,
of
and well-watered
inability
fly.
Second,
the
coupled
with
an
tsetse
of
advance
pace
malarial mosquito and
to organise the supply services effectively meant that troops were usually underfed and
had
insufficient
Third,
the
staff
medical
services
without essentialprotective equipment.
Finally,
through a combination of
deal
the
to
the
problem.
of
and resources
scale
with
ignorance,
discipline
personaland group sanitary measures
and
operationalpressure,poor
followers.
the
troops
to
and
were often neglectedat great cost
Given this potent combination of problems, the health situation worsened
512
heavily,
Africans
South
the
If
the
the
suffered
significantly as
advance progressed.
British and Indian troops were also much depletedby sicknessand overwork resulting in
5
13
followers
The
battalions.
long
the
the need to rotate
of
was also poor
state
serving
[check]
dysentery
from
than
they
tended
to
and
pneumonia
although
suffer much more
malaria. By early 1917, matters had reachedsuch a stagethat the senior medical officer
be
Europeans
Africans
that
of
number
and a minimum
was recommending
only
food
higher
in
Rufiji
the
that
standards
of
shelter,
and clothing
employed
area and
much
514
be

applied.

The War Office beganto becomealarmed as the casualtiesbeganto mount in late

1916. While it had always expecteda higher than normal sick rate in the campaign,the
levels suffered by the East African Force seemedabnormally high. Coupled to this were
persistent stories, principally emanating from South Africa, that mismanagementand
incompetencehad led to much unnecessarysuffering and illness.515 This uneasewas
-'12WO 95/5292, War Diary GHQ, 15 October 1916,Entry 1, e& Yd Divisions. The 31dDivision was abolishedand the two
mountedbrigadesamalgamated.War Diary GHQ, 16 October 1916,Entry Lake Det. & Belgian Force. The Lake Force was
abolishedand its commanderand staff returnedto South Africa. WO 95/5292, War Diary GHQ, 23 January 1917,Entry e Division.
The 2d Division was abolishedearly in 1917.
513WO 95/5300,War Diary DMS, 31 January 1917,Memorandum SurgeonGeneralHunter to [CinC ?]. The SurgeonGeneral
recommendedthat the bulk of non-African troops be given rotated while the Africans remaining be well house,shelteredand amply
fcd. Hordcm, Military OperationsEast Africa, pp. 520-522, gives further details.
514WO 95/5300, War Diary DMS, II February 1917,Letter DMS to DA & QMG.
... WO 33/858, TelegramsD 1, No. 1313,17 October 1916,Telegram D 280/1973, GeneralBotha to War Office, p. 355. "... 1
am
especiallyanxiousto explain situation frankly to South African public, in view of political conditions here and quite unfounded

172

fever-wracked
by
the
soldiers, some of
and
strengthened
return of so many emaciated
whom had beenaway for as little as six months, that it becamea political matter. Finally,
towards the conclusion of General Smuts's time in command, one of his battalion
his
laid
formal
Kirkpatrick,
Colonel
to
Lieutenant
a
complaint
commanding officers,
5
16
GHQ.
The languagewas blunt
brigade commanderwhich was quickly passedup to
and forthright:
"... The total strength of this Regiment, with reinforcements was over 1200, so
that the percentage of EFFECTIVES is approximately 9%, leaving 91% as
INEFFECTIVES.
This, I submit Sir, is an appalling state of affairs and in view of the

been
it
is
have
from
to
time,
time
made
which
medicalreportsandrepresentations
quiteclearthat a very gravechargemust lie againstsomepersonor personswho
must be responsiblefor the existing state of affairs... Among other things,
Starvation,bad and insufficient Clothing and Equipment,dearth of Hospital
debacle,
for
Drugs,comfortsandaccommodation,
the
present
aregivenasreasons
andthesecannotbut be acceptedby everyUnit Commanderandotherpersonsin
is
This
affairs
state of
not due to any sudden
similar responsiblepositions.
difficulty arising,but has existedand continued,at any rate sincethe entry into
have
Kondoa-Irangi,some5 1/2months
ago,andthoughrepeatedrepresentations
beenmadeto you therehasbeenno improvementin the situation...
"
in
from
letter
A-Wart
threat
to
take
the
the
the
the
charges,
explicit
seriousness
of
P
subsequentaction should the matter be ignored left the high commandlittle option but to
convenea formal court of inquiry.
"... I venture to say that if the question of an enquiry is not adopted that things
will be brought to light in a very disagreeablemanner after the campaign is over,
rumoursconcerningBrits which have gainedcredencehere last July... "; No. 1388,17 November 1916,TelegramOA 84, Smutsto
CIGS, p. 377. "... GeneralBotha telegraphedme that complaints which are appearingin South African newspapersare having
disturbing effect. The complaints are to the effect that owing mismanagementtroops here have beenstarved and insufficiently
clothed...".
516CO 551/101,Court ofEnquiry, 26 October 1916,Letter Lt Col Kirkpatrick, Commanding9' Regiment South African Infantry to
GeneralOfficer Commanding3' South African Infantry Brigade, pp. 248-250.

173

in
intelligence,
the
ranks who
and
there
standing
substance,
of
are many men
as
I
Service.
from
freed
the
the
likely
of
restrictions
to
remain quiet when
are not
have only fringed on the whole question, but in support of what I say, would call
Transport
Quartermasters,
Officers,
Supply
Officers,
O's.
C.
Units,
Medical
all
Officers, etc... "
Faced with an umnistakable challenge, the court was establishedby order of 13
November 1916, some two weeks after the allegations were lodged, and ran until May
1917.517 The complaints which alleged that one unit suffered from starvation, inadequate
insufficient
from
clothing and equipment
medical arrangementsand exposure arising
lengthy
A
Central
Railway.
nd
based
largely
to
the
2
Division's
and
the
advance
were
on
detailed enquiry by officers of General Headquartersfollowed, in which a number of
5
18
length.
The
interviewed
taken
evidence was
at
witnesses were
and statements
devastatingas the other commanding officers in the brigade substantiatedKirkpatrick's
claims. In the I lth SAL
"... I left the remnant of my regiment there [KIKUMIJ. I had to point out that I
day
There
16
the
losing
were
average.
a
was
men
was
men sick at such a pace70 men sick in camp and 170 in the field ambulance... The condition of the men
becamevery bad at KIDODL They had been gradually going back for a long
time, due, as I had already pointed out, to their having been drawing on the
discussed
I
knew
Brigadier
The
these
the
time
conditions.
all
about
staminaall
...
it with him more than once and he expressedhis concernto me... "519
Things were little better in the 12'SAL
"I attribute the presentexhaustionof the men of my regiment to constantexposure
to rain and tropical heat without proper cover. Secondly, to shortnessof rations
doses
drugs
In
the
and
constant
absence
of
proper
and shortnessof
opinion
my
...
517CO 551/101, Court
of Enquiry, "Instructions of Authority Assembling the Courr, 13 November 1916, p. 238.
51' CO 5511101, Court
ofEnquiry, 17 May 1917, pp. .238-485.
519CO 551/101, Court
ofEnquiry, "Statement ofCoIN. H. M. Burn", Commanding I I'h South African Infantry, [n. d. ], pp. 329-333.

174

been
has
the
by
of
greatest
the
one
authorities,
though
medical
ordered
of quinine,
the
Commander
told
Brigade
The
all
about
was
the
campaign...
scandals of
issued,
the
inadequacy
as
of
rations
the
and unsuitability
condition of the men and
frequently and used to tell me he had done best to improve matters for them ... On
nd
in
2
10h
S.
A.
I.
nd
in
Command
told
the
2
Mpapua
my
the
of
the march to
in
fallen
had
his
that
50%
Command that
out and
my regiment was
regiment
of
,,520

much the samestate..

Even the medical evidence was damning. The Deputy Assistant Director of
Medical Servicesfor the 2ndDivision stated:
"To sum up, the troops never had full Army rations. They performed miracles on
has
been
Division
as
flour
the
action
of
put
out
water
and
and
mealie pap
...
incredibly
have
been
Drugs
by
done
if
it
had
been
the
short
enemy ...
effectively as
from MASHOTI
practically

(Feb. l7t)

non-existent

comforts, cornflour,

even from

arrowroot, stimulants, etc.,

the commencement

of

the campaign.

Prophylactic doses of quinine were ordered but not available and so on with every
521
,,
individual item...

In light of this, and much more, the Court found in favour of Colonel Kirkpatrick
blame.
It
it
to
find
stated:
to
specific
organisation
or
although was unable
any one person
"... The only conclusion possible is that all the conditions complained of were
brought about by want of facilities for transportation. The shortnessof food and
been
had
it
things
to
these
have
carry
possible
occurred
clothing would not
forward The responsibility for the want of facilities for transportationcannot be
...
fixed on any individual or set of individuals. The exigencies of the campaign

520CO 551/101,Court ofEnquiry, "Statementof LtCol Bottornley", CommandingOfficer lVh SouthAfrican Infantry, [n.d.], pp. 335339.

521
2ndDivision,[n.d.],
CO5511101,
CourtofEnquiry,"Statement
of CaptW CC Pakes",DeputyAssistantDirectorMedicalServices,
pp.331-335.

175

had
been
line
longer
than
contemplated
communications
of
necessitateda much
95522
line
was so extended...
or allowed for at the time the
While 2nd Division had clearly failed to respond adequately to very trying
had
GHQ
the
not taken any remedial action remained
circumstances,
question of why
late1916. For example,
in
health
had
been
The
to
midsince
crisis
apparent
unasked.
the chief medical officer of the 2nd Division had submitted a report stating that the
by
being
debilitated
in
fighting
the
troops
and
were
need of rest
majority of
were
by
lack
food,
fighting
drift-making,
and
and
of
continuous marching, road-making,
523
in
from
by
dysentery.
This
the
reports
medical officers
malaria and
was echoed
524
himself.
Services
For example, on 15
Medical
forward units as well as the Director of
September, 12th SAI reported that 75 per cent of its men were unfit for duty "owing
biscuits
3V2
day
dysentery
fever
by
sufficient
not
per
shortage rations weakened
and
...
525
its
5
1" SAH

ration" while

could only muster per cent of

effectives.

Theproblemswereamplyreflectedin the hospitalfigures;in September1916,lt

Division had a total strength (including porters) of 10,598, of whom 1,362 sick and 193
wounded were admitted to hospital, and of these some 1,224 had to be evacuated
rearwards. Medical officers found that some 17 per cent of the troops had enlarged
spleensand anaerniawhile 30 per cent were unfit for hard work owing to underfeeding
526
heavy
There is no evidence of either the commander-in-chief or the
and
exertions.
general staff initiating a sustained effort to sort out the serious problems in supply,
transportor health, apart from ordering more evacuations.

The choice of the principal administrativestaff officers from GHQ as court

membersno doubt influencedthe outcome,but even the most generousinterpretation
522CO 5511101,Court
ofEnquiry, "General Conclusion", pp. 476477.
'23Macphersonand Mitchell, Official History Medical Services, Volume4, 450.
p.
'24WO 95/5300, War Diary DMS, 9 November 1916,Letter DMS to DA & QMG
attaching letters 28 November 1916, Senior
Medical Officer, 2' South African Field Ambulance, attached2" Division; 28 November 1916,Assistant Director Medical Services,
2"dDivision; WO 95/5292, War Diary GHQ, 10 October 1916,Entry P, 2"d&
Divisions; and 22 October 1916,Entry P, 2"d & 31
Jrd
Divisions; and 23 December1916,Entry 2d Division. To be fair to van Deventer,
he had beenreporting the seriousnessof his supply
and manpowersituation throughout this period; GSWA, Box 26, It September1916,Telegram K 475,2 Div to Smuts.
5'sGSWA, Box 26,15 September1916,TelegramSC 548, Thirbrig to 2 Div; Box 29,16 December1916,Telegram T 63, OC
CompositeRegiment [I ObSAI] to 2 Div.
$26Macphersonand Mitchell, Official Histo? Medical Services,Volume4, 457.
p.
y

176

failings
the
hide
administrative and medical services.
the
within
substantial
could not
The matter was referred back to London where it quickly became a political hot potato
betweenthe War and Colonial Offices. General Smuts was shown a copy of the report
his
doubt
No
it
for
his
of
conscious
advice on whether to publish or not.
and asked
its
he
home,
advised suppressionuntil after the campaign was over.
political position at
He was also unhappywith the findings of the court:
"After reading the report I have come to the conclusion that, instead of
have
its
the
Johannesburg],
[in
feeling
might
publication
the
and agitation
allaying
is
the
the
I
to
campaign over
report until
opposite effect and am averse publishing
by
Division
largely
it
deals
the
commanded
administration of a
with
very
as
General Van de Venter and its publication might seriously weaken his position.
Further, in my opinion, the report is one-sided and does not sufficiently consider
the campaign as a whole and the unforeseen physical difficulties encountered nor
527
,,
by forced
for
is

there any allowance given

the results achieved

marches...

The official War Office line to the South African Government was a whitewash
that justified Smuts' conduct of the campaignwhile admitting that 2ndDivision suffered
528
It was a
difficult
from a lack of experience in novel and unexpectedly
conditions.
lame reply to a very serious matter that deserved closer attention. Doubtless, it was
hopedthat the matter would now die quietly.
If Whitehall was preparedto brush off a colonial government with a vague and
Despite
Court.
findings
the
it
the
all
of
unsatisfactoryanswer, was also concernedwith
of Smuts' reassurancesabout the unavoidable nature of the administrative problems
enduredby his forces, it was evident that the medical situation was very unsatisfactory
in
War
July
1917
independent
investigation
Accordingly,
the
that
and
was necessary.
Office instructed SurgeonGeneralWW Pike to go out to East Africa and to report on the

517C0551/101, Union of South Africa, 1917,Volume 8, Folio 38195,31 July 1917,Letter JC Smutsto Sir GeorgeFiddes, 9 August
1917,p. 234.
523CO 5511101,Union of South Africa, 1917,Volume 8, Folio 44918,10 September1917,Letter 0165/8468 (MO 2) Secretary,War
Office to Under-Secretaryof State,Colonial Office, 9 September1917,pp. 570-571.

177

529
large
four
His
took
there.
a
number of units being
medical situation
months with
study
visited and numerous personnel interviewed. The team was able to gain a first-hand
impression of conditions and in particular the difficulties under which the force operated.
The report covered all aspects of medicine and health, coming to a large number of
findings. One of the earliest commentswas the damagedone to morale and efficiency by
Smuts's ill-advised pronouncements:
"... A public statementmade in England early in 1917 that the war was over in
East Africa did incalculable harm as it causeda general feeling of "Oh! The war
is over here, why trouble to increaseunits, get things out from home for camps
and build bandas,etc., etc." The war was not over or anything like over; and all
precautionsagainstdiseasewere required with greaterurgency than before; but, in
spite of the fact that every thinking man out here knew the statementwas not
initiative,
for the thought could not but
it
inevitably
led
to
correct,
a curtailing of
intrude itself when a suggestionwas made or a requisition sent forward, 'Is this
necessaryas the war is said to be practically at an end?919530
Some areas,such as the work of the field ambulancesand hospitals together with
methods of casualty evacuation were found to be satisfactory. However, the scientific
and sanitary aspects of the medical services came in for most scathing and damning
criticism although the failure of the command to take note of medical advice was also
531

Of particular note was the failure to supply the microscopes essentialto the
correct diagnosisof tropical diseasesdespiterepeatedrequestsby medical officers.

noted.

There were also substantial criticisms of the discipline of the force, particularly
concerning malaria.

It had proven to be the biggest medical problem throughout the

campaigncausing much suffering and disability. Yet, despite its seriousnessand wellknown effects, GHQ and the medical services had failed to tackle its causesrobustly

529CO 691/19,Pike Report,

pp. 1-89.
5'0CO 691/19,Pike Report, 6.
p.
531CO 691/19,Pike Report,
pp. 87-88.

178

initial
the
Much
treatment
to
attack as well as
of
enough.
of the problem related poor
inadequatemeasurestaken to prevent infection. In the words of the report:
"Neither quinine prophylaxis nor the issue of Proper mosquito nets, nor what may
be termed "malaria discipline" was enforced as should have been the case. No
in
force
and apparently we entered
put
proper anti-malarial organisation was
lightly on a campaign in one of the most malarious regions of Africa without
532
,,
idea

much

of the seriousnessof the undertaking.

It went on to point out that the use of the former German capital, Dar-es-Salaarn,
known
led
It
had
1916
base
to
the
September
the
was
many of
problems.
as
primary
after
to be extremely malarious yet no seriousanti-malarial measureswere put into place until
the Board arrived. It was generally the first landing place of all reinforcementsand the
transit camps in the vicinity of the town were highly exposedto malarial mosquitoes. It
was unsurprising when the bulk of newly arrived troops contracted malaria there and
beganthe vicious circle of repeatedattacks that led to a marked decline in health. Once
533
it
break
the cycle.
started, was extremely difficult to
The post-war Official History was less restrainedin its language:
"Dar-es-Salaamwas probably the most malarious locality within the area of
Antiits
incidence
danger
throughout
the
the
whole
year
existed
operationsand
of
...
malaria work was not taken up seriously till the later phases of the military
largely
been
had
it
the
troops
already
operations, and when
was commenced
in
in
fact,
large
Malaria,
the early stagesof
to
riot
practically
ran
affected a
extent.
the campaign,before adequatearrangementshad been made for the protection of
new arrivals from the moment they disembarked. Had such arrangementsbeen
have
in
been
incidence
this
the
undoubtedly
possible
campaignwould
of malaria
materially less."'
112CO 691/19,Pike Report, p. 26.
533CO 691/19, Pike Report, p. 27.
534Macpherson,Major GeneralSir W G, Herringham, Major GeneralSir W P, Elliott, Colonel T R, and Balfour, Lieutenant Colonel

179

The Pike Report went on for considerable length and produced considerable
be
it
be
for
its
findings.
Nowhere
to
a ringing endorsementof the
said
could
evidence
it
Finally,
the
the
concluded:
conduct of
campaign.
medical aspectsof
"When, however, all is said and done, and every allowance made for these and
other drawbacks,we cannot but feel that there is much to regret in the medical
history of this campaign and that a great many difficulties were due to a lack of
035
forethought, of driving power, and of expert knowledge and assistance.
From this and other evidence it was clear that the link between the general and
during
For
inadequate.
the advance south of the
example,
administrative staffs was
Central Railway, there was a lack of detailed planning and poor liaison betweenthe key
departments:

"The above outline of the operations shows that they were complicated There is
...
no evidence, however, that any well-considered scheme was evolved or that
medical units were instructed regarding the part they would play during the
operations.

The medical arrangements appear to have been haphazard, and

sections of field ambulances were apparently attached to any formation that

happened to be at hand.11536

The Official History also highlighted the major shortcomings in the overall
organisationof the force:
"No doubt the abnormal sick-rate was in part due to the failure on the part of the
responsible military staff to recognize the importance of preventative measures
against disease. The D.M. S. of the force had his headquarterstoo far away. He
was not always in direct communication with General Headquartersnor with the
A, edsHistory ofthe Great War Basedon Official Documents Medical ServicesDiseasesof War, Volume 1, London: HMSO, (n.d.],
pp. 254-255. Henceforth,Macphersonet al, Official History Medical ServicesDiseasesof War Volume1.
5'5CO 691/19,Pike Report, p. 87.

180

Army Commander, and most of his recommendations had to be made by

inadequate,
Further,
and
generally
were
correspondence.
rations and equipment
there can be no doubt that men who are under-fed, exposedto hard fighting and
fall
easy victims to
unsuitably protected against weather conditions
disease There were not sufficient medical units, and when the medical staff
...
became depleted the reinforcements sent out consisted of young and
inexperienced medical officers with little knowledge of tropical medicine or
by
the
troops
their
to
charge
under
sanitation, and only able communicate with
meansof inefficient

537
,
interpreters.

Matters began to change in early 1917, when the new commander-in-chief took
steps to reduce the still-rising sick rate. A complete and comprehensive reorganisation of
the medical services was undertaken with emphasis being placed on making the field
ambulances more mobile and having a coordinated system of evacuation and treatment.
Extra staff was recruited and more effort was directed towards the medical support for the
538
lacked
Nevertheless,
the
proper
of
sanitation
much-neglected carriers.
critical area
organisation and was inadequately supervised until after the visit of Surgeon General
Pike in August 1917.539

The general advancethat began in mid- 1917 benefited greatly from the improved
system of medical support although at one stagethe rapid advanceof the fighting troops
outran the medical system and delays in the evacuation of the sick and wounded
occurred. Despite the bitterly won lessonsof the previous year, this factor does not seem
540

to have figured in the Staff s calculations and nor was it ever reported to the DMS
.
Even at the end of October 1917, during the pauseto regroup prior to the final clearance
of GEA, the medical staff failed to make the necessarycasualty evacuation plan despite
being briefed on the overall intentions. The fact that the DMS still remainedin

536Macphersonand Mitchell,
537Macphersonand Mitchell,
532Macphersonand Mitchell,
339Macphersonand Mitchell,
(rank range.)
'40Macphersonand Mitchell,

Official
Official
Official
Official

History Medical Services, Volume4, p. 453.

History Medical Services, Volume4, p. 467.
History Medical Services, Volume4, pp. 475-477.
History Medical Services,Volume4, p. 499. The post was upgradedfrom DADMS to ADMS

Official History Medical Services, Volume4, p. 485.

181

Dar-es-Salaam while the commander-in-chief was far away to the south can only have
541
contributed to the inefficiencies of the system.
One area of the administrative system that seems to have been consistently
ignored throughout 1916 and 1917 was the motor transport. Given the inefficiencies and
logistical
human
the
throughput
the
of
system
and
weaknessesof
effectiveness
portage,
forward
However,
far
dependent
transport
as
possible.
as
was
moving supplies
on motor
it required both machines and drivers to operate it and the machines seem to have
largely
health
drivers
better
feeding,
The
the
the
were
of
support.
welfare and
received
left to chanceand they suffered as badly from illness as those in the front line. Again, the
reasonsare not difficult to determine:
"... These men had been overworked and had driven their cars from early morning
till late at night over roads that were never good and often dangerous. They were
exposed to great extremes of heat and cold, while their messing and sleeping
defective.
haphazard
their
the
camps
arrangements were
sanitation of
and

No

having
tent
had
been
them
or
to
of
camps
system
rest
attempt
with
a
made provide
'banda' accommodation but it is evident, too, that very little attention was paid
...
to the welfare and comfort of mechanicaltransport personnel...99542

By 1917, opinion both within and outside of East Africa was coming to the
conclusion that it was unsuitable for Europeanor Indian troops. With the publication of
the Pike Report and the reports of returning officers, the War Office came to sharethis
view, and by the end of that year, all the British, South African and Indian fighting troops
had withdrawn to other theatres,leaving behind mainly specialist units. The fighting in
1918 was largely conducted by African troops on the grounds that only they had
sufficient resistanceto diseaseand acclimatisationto be effective there although the need
to find reinforcements for the Middle East and shipping shortages also played an
important role.

Macphersonand Mitchell, Official History Medical Services,Volume4, p. 487.

'541

182

East Africa never fared well in medical matters, although it was far from being
in,
alone
causing concern. The Medical ResearchCouncil expressedits concerns about
the Army's inability to replicate the successof the medical system in France elsewhere,
particularly in the eastern theatres, as well as the failures in disseminating effective
543
In contrast to this unsatisfactory situation, there
measuresof preventative medicine.
was both high level support and effectivenessin implementing sanitary measuresin the
BEF in France. The same staff system prevailed as in East Africa, but the chief
difference in Europe was that the officer in command of troops took responsibility for
their health and the advice of senior medical staff appearsto have been heeded.544As Ian
Whitehead has concluded:
"The combatantbranch contributed to theseachievementsby adopting a generally
sympathetic attitude towards the work of the RAMC. The latter proved to be a
critical factor as was evident from the disastrouslack of consideration given to
545
fronts,
medicine on other
such as Gallipoli, Mesopotamia,and East Africa.
By the end of the war, there was little doubt that much unnecessary and
preventable suffering had occurred amongst troops and followers, due to poor planning
and coordination. The true numbers of casualtieswill never be known, but they were
excessiveby the standardsof the day and, most particularly, of the state of scientific and
medical knowledge. Perhapsthe most telling epitaph was the minute of the Adjutant
Generalon the Pike Report:
"Had this particular campaignbeen more in the public eye, a very grave scandal
would have resulted, owing to the want of supervision on the part of the
Commanders-in-Chiefover the departmentsof their staff concerned.ti546

M2Macphersonand Mitchell, Official History Medical Services, Volume4, 474.

p.
543Whitehead,Ian K Doctors in
the Great War, pp. 144-145.
544Whitehead,Ian R. Doctors
in the Great War, pp. 219-221.
545Whitehead,Ian R. Doctors
in the Great War, p. 254.
5" CAB 103/94,WO File 0 165/9851, Minute
by General Sir Nevil Macready on Pike ReporL [check details here]

183

ANALYSIS OF SICKNESS

While there was little contemporaryquestionthat the rate of sicknessin EastAfrica

was excessivelyhigh, apartfrom the post-warMedical Services' OJJIcialHistory, there was
almost no serious coverage of the true scale of the problem. Even the official work,
diligent as it was in highlighting the failings of the campaign and providing reams of
statistics,only put a limited effort into comparisonof the levels of morbidity and mortality
acrossthe theatres. Yet, without an objective look at the levels of casualtiessuffered,it is
difficult to cometo an accurateassessment
of the successesand failures of EAEF's medical
effort. Fortunately,the final volume of the Medical Services' Official History provides an
excellent source of statistical evidence from which a preliminary assessmentcan be
formed. The only drawback is that it relatessolely to the British forces and only a rough
estimateof Germancasualtiescan be compiled from other sources.
Table I comparesthe proportion of battle to non-battle casualtiesacrossselected
theatresand in the SouthAfrican War of 1899-1902. It is noteworthy that East Africa had
by far the highest proportion of non-battle to battle casualties,over 1.5 times that of the
next nearesttheatre, Macedonia, and nearly 24 times that of France and Flanders. It is
nearly twice the earlier campaignin South-WestAffica and the South African War. Even
compared to unhealthy theatres such as Macedonia and Mesopotamia, the ratio is
extremelyhigh.

184

TABLE I
PROPORTION OF NON-BATTLE TO BATTLE CASUALTIES IN SELECTED
BRITISH EXPEDITIONARY

PERIOD

-THEATRE

FORCES547

BATTLE

NON-BATTLE

RATIO NON-BATTLE

CASUALTIES

CASUALTIES

TO BATTLE

East Africa

1916-1918

10,717

336,540

31.40x

Macedonia

1915-1918

23,762

481,262

20.25x

South-West

1914-1915

1,588

24,746

15.58x

1899-1902

26,750

404,126

15.10X

1915-1918

51,451

503,377

9.78x

Mesopotamia

1914-1918

85,207

820,148

9.63x

Franceand

1914-1918

2,690,054

3,528,486

1.31x

Africa
South African
War
Egypt and
Palestine

Flanders

'7 Mitchell Major TJ and Smith, Miss GK History

Great War Based on Oricial DocumentsMedical Services Casualties
the
of
,
and Medical Statistics of the Great War, London: HMSO, 1931, reprinted Nashville, Tennessee:Battery Press, 1997, p. 56, Table 1.
"Proportion of Battle to Non-Battle Casualtiesin the British Expeditionary Forces during the Great War". Henceforth, Mitchell and
Smith, Official History Medical Services Casualtiesand Medical Statistics. It should be noted that all figures for the South African
War are for other ranks only. All other figures are for British and Imperial troops and exclude followers.

185

Table 2 shows the proportion of killed to wounded and total permanentlossesto

temporary losses. East Africa had the secondhighest ratio of killed to wounded and of
likely
battle
to
there
losses.
In
to
temporary
were
more
casualties
permanent
other words,
die, but that overall casualtieswere much more likely to return to serviceafter treatment.
TABLE2
PROPORTION

OF CASUALTIES

IN SELECTED

BRITISH

EXPEDITIONARY

FORCES549

'THEATRE

PERIOD

KILLED TO

PERMANENT

WOUNDED

LOSSES TO
TEMPORARY
LOSSES

East Africa

1916-1918

1: 3.70

1 : 33.72

Macedonia

1915-1918

1: 6.50

1: 46.60

South-West

1914-1915

1: 3.36

1: 20.78

1899-1902

1: 4.50

1: 19.50

1915-1918

1: 5.43

1: 26.41

Mesopotamia

1914-1918

1: 5.36

17.78

Franceand

1914-1918

1: 7.80

1: 5.60

Africa
South African
War
Egyptand
Palestine

Flanders

541Mitchell and Smith, Official History Medical Services Casualties

Statistics,
drawn
from
Medical
169,
Table 6,
p.
and
"Proportions" (France and Flanders); p. 189 Table 7 "Proportions" (Macedonia); p. 226 Table 8(c), "Proportions Whole Force
(Mesopotamia);p. 211, Table 6 "Proportions" (Egypt and Palestine);p. 256, Table 7, "Proportions" (East Africa); p. 264, Table 6
"Proportions" (South-WestAfrica); and p. 270, Table 6, "Proportions" (South African War).

186

Table 3 shows the absolute number of casualties suffered in selected theatres,

injury
to
losses
important
to
how
in
relative
East
Africa's
or
were
sickness
case
showing
those incurred in battle. In terrns of troops, the sick reached86 per cent of those suffered
in the much larger South African War; together with the followers, overall sicknesslevels
were 48 per cent higher than those of the previous conflict.
TABLE3
APPROXIMATE TOTAL CASUALTIES IN SELECTED BRITISH
FORCES549

EXPEDITIONARY

THEATRE

KILLED

DIED OF

DIED OF

MISSING

WOUND-

SICK OR

WOUNDS

DISEASE

AND PW

ED

INJURED

TOTAL

OR
INJURY
East Africa

6,558

10301

7,777

330,232

349,311

44,911

635

1,333

241,688

288,943

1,299

3,744

2,778

16,888

477,518

505,024

185

61

181

- 782

560

23,565

26,334

5,774

2,018

14,048

202

19,457

390,444

430,876

7,394

2,993

5,981

3,871

37,193

497,396

554,828

11,008

5,281

16,712

15,221

53p697

803,706

905,625

381,261

151,356

32,098

319,824

1,837,613

3,496,388

6,218,540

1,965,646

1,965,646

2,689

754

(Troops)
East Africa

376

(Followers)
Macedonia

South-West

2,797

Africa
South African
War
Egypt and
Palestine
Mesopotamia
Franceand
Flanders
United

Kingdom
TOTALS

418,361

167,172

113,173

352,458

I
I
2,00,4,9761 6,074,552 1 11,096,3381

549Mitchell and Smith, Official History Medical Services Casualties and Medical Statistics, p. 12, Table 1, "Approximate Total
Casualtiesin the British Expeditionary Forces during the Great War; the numbers of South African War casualtieshave been drawn
from p. 269, Table 3, Not all theatres are shown in the table, but their numbers have been included in the overall totals. For East
Africa troops, the totals of sick and injured relate to 1916-1918only - all other figures cover the entire war. For followers, the totals
of killed and died of wounds have been combinedunder the killed column.

187

From the foregoing statisticsit is clear that East Af Hca's casualtieswere significant
is
importance.
It
to
that
now
theatres
overwhelming
as compared other
sicknesswas of
and
useful to examine the breakdown of casualtiesby diseasetype incidence of malaria:

188

Table 4 shows the

TABLE4
INCIDENCE OF MALARIA IN SELECTED BRITISH EXPEDITIONARY
FORCES
(DEATHS AND RATIOS PER 1,000 OF RATION STRENGTH)

Campaign

Period

Admissions

Deaths

Ratio per 1,000

Admissions

East Africa

1916

50,768

263

(Jun-Dec)

Macedonia

South-West

Deaths

1,039.11

5.38

1917

72,141

499

1,422.84

9.84

1918

22,941

69

559.09

1.68

1916

32,018

287

259.48

2.33

1917

71,412

228

391.12

1.25

1918

59,087

272

458.94

2.11

1914-1915

518

15.70

0.06

1899-1902

25,156

85

45.30

0.15

1916

1,423

7.57

0.04

1917

8,480

73

45.46

0.39

1918

30,241

773

130.86

3.35

1916

16,957

102.54

1917

16,063

118

52.24

0.38

1918

21,447

166

52.10

0.40

1916

58

0.04

1917

781

0.41

0.00

1918

2,739

1.47

0.00

Africa
South African
War
Egyptand
Palestine

Mesopotamia

Franceand
Flanders

This tableshowshow truly significantthe malariaproblemwasin 1916and 1917;

in the formeryeareverysoldierwasadmittedto hospitalat leastonce,while in the latter
yearthe admissionratewas over 1.4times. This rate washalvedin 1918,but was still
"0 Mitchell and Smith, official History
Medical Services- Casualtiesand Medical Statistics, p. 80, Table 19.

189

1.2 times the highest rate suffered anywhereelse (Macedonia 1918). Overall, notoriously
malarious theatres such as Mesopotamia, Macedonia and Egypt/Palestine compare very
favorably with East Africa, especiallyin the deathrates.

190

Table 5 shows the incidence of dysentery, including both the bacillary and

amoebicvarieties.
TABLE5
INCIDENCE

OF DYSENTERY

IN SELECTED

BRITISH

EXPEDITIONARY

FORCES

(DEATHS AND RATIOS PER 1,000 OF RATION STRENGTH)"'

Campaign

Period

Admissions

Ratio per 1,000

Deaths

Admissions
East Africa

Deaths

8902

306

182.21

6.26

1917

14,045

429

277.01

8.46

1918

3,294

86

80.28

2.10

1916

8,204

169

66.49

1.37

1917

5,792

124

31.72

0.68

1918

9,438

173

73.31

1.34

1914-1915

715

13

21.67

0.39

1899-1902

38,108

1,343

68.60

2.42

1916

5,597

81

29.78

0.43

1917

4,431

139

23.27

0.75

1918

4,906

264

21.23

1.14

1916

15,270

92.34

1917

11,959

286

38.89

0.93

1918

12,290

303

29.86

0.74

1916

5,776

40

2.64

0.04

1917

6,025

46

3.18

0.02

1918

12,211

46

6.58

0.02

1916
(Jun-Dec)

Macedonia

South-West
Africa
South African
War
Egypt and
Palestine

Mesopotamia

France and
Flanders

551Mitchell and Smith, Official History Medical Services Casualties

and Medical Statistics, p. 81, Table 20.
-

191

The East African figures for dysenteryare much lower than for malaria, but were
still the highestby far. In 1917the rate was 6.2x the peak incidencein Macedoniaand 3.5x
that of the SouthAfrican War.
Table 6 has been derived from Tables 4 and 5 above. It shows the ratio of
admission rates for East Africa against those of selected theatres for both malaria and
dysentery.

192

TABLE6
COMPARISON OF EAST AFRICAN TROOP CASUALTIES TO SELECTED
..
THEATRES
(BASED ON COMPARING ADMISSION RATES PER 1,000 RATION
STRENGTM552
Campaign

Period

Casualties to Selected Theatre

(Malaria)

(Dysentery)

Admissions
Macedonia

South-West

Ratio of East African Troop

Ratio of East African Troop

Casualties to SelectedTheatre

Admissions

Deaths

Deaths

1916

4.Ox

2.3x

2.7x

4.6x

1917

3.6x

7.9x

8.7x

12.4x

1918

1.2x

0.8x

LIX

1.6x

1914-1915

64. lx

93.3x

8.3x

39.5x

1899-1902

22.2x

37.3x

2.6x

6.4x

1916

137.3x

134.5x

6. lx

14.6x

1917

31.3x

25.2x

11.9x

11.3x

1918

4.3x

0.5x

3.8x

1.8x

1916

10.lx

2.Ox

00

1917

27.2x

25.9x

7. lx

9. lx

1918

10.7x

4.2x

2.7x

2.8x

1916

25977.8x

00

69.Ox

156.5x

1917

3470.3x

00

89.4x

423.Ox

1918

380.3x

00

12.2x

105.Ox

Africa
South African
War
Egyptand
Palestine

Mesopotamia

Franceand

00

Flanders

With regardto malaria, East Africa had admissionrates that were generally much
higher than Macedonia, Mesopotamiaand Egypt/Palestine all known for their malarial
problems. The relative differences in death rates for malaria were quite considerable.
The enormous differences of ratios as compared to France and Flanders can be

193

disregarded owing to the very small number of casessuffered there. Dysentery was less
it
bad
but
in
than
was
relatively
as
nevertheless
a
of problem absolute numbers
malaria,
for
death
The
&
Flanders.
France
Macedonia,
Egypt/Palestine
rate
compared with
and
dysentery was markedly than greater in all other theatres, with the difference between
East Africa and Franceand Flandersbeing exceptional.

552The figures for East Africa in 1916

are for Jun - Dec only. The comparisonswith South-WestAfrican and South African Wars
havebeencalculatedagainstthe averageEast African casualtiesfor 1916-1918.

194

Table 7 shows the incidenceof the enteric group of fevers, including typhoid and
paratyphoid fevers.
TABLE7
INCIDENCE OF THE ENTERIC GROUP OF FEVERS IN SELECTED BRITISH
FORCES

EXPEDITIONARY

(DEATHS AND RATIOS PER 1,000 OF RATION

Campaign

Period

Admissions

STRENGTH)

Ratio per 1,000

Deaths

Deaths

Admissions
East Africa

142

33

2.91

68
.

1917

124

24

2.45

1918

69

13

1.68

1916

1,105

40

8.96

1917

529

19

2.90

1918

135

1.05

1914-1915

230

26

6.97

47
.
32
.
32
.
10
.
05
.
79
.

1899-1902

57,684

8,022

103.88

14.45

1916

3,108

70

16.54

37
.

1917

573

27

3.07

14

1918

1,071

148

2.60

1916

3,060

1917

1,535

157

18.50

1918

1,071

148

2.60

1916

2,738

30

2.07

1917

1,275

24

1918

376

22

67
.
20
-

1916
(Jun-Dec)

Macedonia

South-West
Africa
South African
War
Egyptand
Palestine

Mesopotamia

Franceand

4.99

.
36
.

51
.
36
.
02
.

Flanders
01
.
01
.

5"' Mitchell and Smith, Official History Medical Services

Statistics,
Medical
p. 66, Table 7, "Incidence of the Enteric
and
-Casualties
Group of Feversduring the Great War, with Deaths and Ratios per 1,000of Ration Strength."

195

The figures for East Africa comparevery favourably with Macedonia, Egypt and
Palestineand Mesopotamiaalthoughthey are less good than those for Franceand Flanders.
The difference with the South African War, in which typhoid was a major killer, is
enormous.

196

Table 8 showsthe incidenceof pneumonia:

TABLE8
INCIDENCE OF PNEUMONIA IN SELECTED BRITISH EXPEDITIONARY
FORCES
(DEATHS AND RATIOS PER 1,000 OF RATION STRENGTID554

Campaign

Period

Ratio per 1,000

Deaths

Admissions

Deaths

Admissions
East Africa

463

119

9.48

2.44

1917

1,653

444

32.60

8.76

1918

2,023

516

49.30

12.58

1916

73

10

1917

279

43

59
.
1.53

08
.
24
.

1918

48

1914-1915

124

12

37
.
3.76

1899-1902

2,591

466

4.57

85
.

1916

1,080

115

5.75

1917

228

55

1.22

1918

1,510

1,027

6.53

61
.
29
.
4.44

1916

4,104

2.57

1917

4,924

2.89

1918

9,484

5.88

1916

1,497

135

1.13

1917

2,157

193

1.14

1918

1,921

352

1.03

1916
(Jun-Dec)

Macedonia

South-West

36
.

Africa
South African
War
Egyptand
Palestine

United
Kingdom

Franceand

10

Flanders
.
.

10
19

"Mitchell and Smith, Official History MedicalServices- Casualtiesand Medical Statistics, p. 71, Table I I, "Incidence of
Pneumoniain the British Expeditionary Forcesduring the Great War, with Deathsand Ratios per 1,000of Ration Strength." Note that
there are no statisticsgiven for Mesopotamiaand that figures for the United Kingdom have been insertedinstead.

197

'The table shows how serious a problem that pneumonia proved in East Africa.
The death rate was particularly alarming and was quite disproportionateto other theatres.

198

TABLE9
COMPARISON

OF EAST AFRICAN

TO SELECTED

TROOP CASUALTIES

THEATRES
(BASED ON COMPARING

Campaign

Period

ADMISSION

Ratio of East African Troop

Ratio of East African Troop

Casualties to Selected Theatre

Casualties to Selected Theatre

(Enteric Fevers)

(Pneumonia)

Admissions
Macedonia

South-West

RATES PER 1,000 SRENL.'Tll)`

Deaths

Admissions

Deaths

1916

0.32x

2.1 3x

16.07x

3.05x

1917

0.84x

4.70x

2 1.3 1x

36.50x

1918

1.60x

6.40x

133.24x

1914-1915

0.42x

0.86x

2.52x

6.78x

1899-1902

0.03x

0.05x

2.07x

'-. 87x

1916

0.1 8x

1.84x

1.65x

4.00x

1917

0.80x

3.36x

26.72x

30.2 1x

1918

0.65x

0.89x

7.55x

2.81X

1916

0.16x

f)

1917

0.49x

0.92X

1918

0.65x

0.89x

1916

1.4 1x

34.00x

1917

3.66x

47.00x

28.6x

87.60x

1918

8.40x

', 2.0ox

48.86x

66.2 1x

J,

Africa
South African
War
Egypt and
Palestine

Mesopotamia

France and

Flanders

United

1916

3.69x

1917

11.28x

1918

8.38x

Kingdom

... Note figures tor East Africa in 1916 are for Jun Dec only. The comparisons NN
ith South-West African and South African Wars
have been calculated against the average East African casualties tor 1916-1918.
There is no data for pneumonia cases in
Mesopotamia and figurcs tor (lie United Kingdom have been used in its place. The reverse is true ofenteric fevers

199

Table 9 has been derived from Tables 7 and 8 above. It shows the ratio of
,,
for
both
fevers
for
East
theatres
Africa
those
enteric
selected
of
admission rates
against
and pneumonia.

200

THE FOLLOWERS'HEALTH

The foregoing has compared the fighting forces in various theatres with East

-,
Africa. However, casualtiesamong the civilian followers, most notably the carriers, was
very high and it is instructive to comparetheir losseswith those of the troops. Table 10
shows the incidence of malaria and dysenteryamongstthe followers:
TABLE 10
INCIDENCE

OF MALARIA

AND DYSENTERY

AMONGST

FOLLOWERS

IN

EAST AFRICA
(DEATHS AND RATIOS PER 1,000 OF RATION STRENGTH)

Disease
,

Period

Deaths

Admissions

Ratio per 1,000

Admissions

Malaria

Deaths

7,127

187

90.33

2.37

1917

40,527

2,291

281.52

15.91

1918

21,260

361

242.63

4.12

1916

3,795

1,008

48.10

12.78

1917

26,607

7,277

184.82

50.55

1918

3,740

681

42.68

7.77

1916

0.11

01
.

1917

40

1918

42

1916

1,957

372

28
.
48
.
24.80

03
.
08
.
4.72

1917

10,009

3,013

281.52

15.91

1918

3,116

873

35.56

9.96

1916
(Jun-Dec)

Dysentery

(Jun-Dec)

Enteric Fevers

(Jun-Dec)

Pneumonia

(Jun-Dec)

556Mitchell and Smith, Official History Medical Services Casualties

and Medical Statistics, pp. 259-260, Table 13 "Principal
Causesof Admission to Hospital in the East African Expeditionary Force, Juneto December 1916,with Deathsand Ratios per 1,000
of Ration StrenW, Table 14 "Principal Causesof Admission to Hospital in the East African Expeditionary Force, 1917,with Deaths
and Ratios per 1,000of Ration Strength" and Table 15 "Principal Causesof Admission to Hospital in the East African Expeditionary
Force, 1918,with Deathsand Ratios per 1,000of Ration Strength".

201

Table 11 is derived from Tables4,5, and 10 to producea ratio of casualtiesbetweentroops

and followers for malaria and dysentery:
TABLE 11
COMPARISON

OF MALARIA

AND DYSENTERY

AND FOLLOWERS

Period

RATES BETNATEN TROOPS

IN EAST AFRICA...

Ratio of Troops/Followers

Ratio of Troops/Followers

Malaria

Dysentery
Deaths

Admissions

Deaths

11.5x

2.3x

3.8x

0.5x

1917

5. lx

0.6x

1.5x

0.2x

1918

2.3x

0.4x

1.9x

0.3x

Admissions
1916
(Jun-Dec)

The figures show that troops were much more likely to contract malaria than
followers although the latter had a much higher chanceof succumbingto the diseaseif
for
followers
death
for
dysentery,
The
the
that
rate
contracted.
except
samesituationprevails
is evenhigher than for malaria.

117
MitchellandSmith,OfficialHistoryMedicalServices Casualties
Medical
Statistics,
80-8
1,
derived
from
Tables
19
pp.
and
and
20.

202

Table 12 is derived from Tables6,7 and 10 to producea ratio of casualtiesbetweentroops

and followers for entericfeversandpneumonia:
TABLE 12

COMPARISON OF ENTERIC FEVER AND PNEUMONIA RATES BETWEEN

TROOPS AND FOLLOWERS IN EAST AFRICA"'

Period

Ratio of Troops/Followers

Ratio of Troops/Followers
Pneumonia

Enteric Fever

1916

Admissions

Deaths

Admissions

Deaths

26.45x

68.00x

0.3 8x

0.52x

8.75x

15.67x

0.62x

0.55x

3.50x

4.00x

1.39x

1.26x

(Jun-Dec)
1917

COMPARISON WITH PREVIOUS TROPICAL CAMPAIGNS

Previous campaignsin tropical Africa had suffered heavily from diseasealthough

the advancesin medical and sanitary sciencein the 19'hcentury meant that morbidity and
mortality had been decreasingover time. While the statistics have been drawn from a
number of sourcesand are not entirely consistent, it is instructive to compare previous
campaignswi that fought in East Africa. Table 13 showsthe deathsby typhoid (enteric
fever) by British soldiers in South Africa in the period precedingthe First World War.

"$Mitchell and Smith, Official History Medical Services Casualties

Statistics,
80-81,
derived
from
Tables 19 and
pp.
andMedical
20.

203

TABLE 13
TYPHOID DEATHS AMONG BRITISH SOLDIERS IN SOUTH AFRICA,
1879-1913
(DEATHS PER 1,000 PER YEAR)

PERIOD

TYPHOID

DEATHS

DEATHS

FROM ALL
DISEASES

1879-84

9.61

18.52

1886-97

1.87

6.56

1899-1902

18.11

24.53

1905-09

0.80

2.58

1909-13

0.33

3.84

TABLE 14
TYPHOID

DEATHS AMONG BRITISH

SOLDIERS

IN EAST AFRICA

1916-1918

(DEATHS PER 1,000 PER YEAR) 560

PERIOD

TYPHOID
DEATHS

1916

0.68

1917

0.47

1918

0.32

From Table 14 above, the East African death rates for 1916-17-18 were 0.68,

0.47,and0.32per thousandrespectively.Thesearemuchlower thanthoseof the South

African War period and reasonablyclose to those causedin the inter-war period.
However,1905to 1913werenot spenton major campaignsso the EastAfrican figures
5" Curtin, Disease
and Empire, p. 206, Table 8.1, "Typhoid Deaths Among British Soldiers in South Africa, 1813-1913".
560Tables
above

204

longer
fighting.
the
It
typhoid
that
the
was
no
seems
seem reasonablegiven
conditions of
deadly threat it had beenin past years.

205

TABLE
COMPARISON

(DEATH

DISEASE

THEATRES

1886-1918

RATES PER THOUSANDf"'

INDIA

BRITAIN

PERIOD

DEATH RATES BY DISEASE IN

OF ANNUAL

SELECTED

15

ALGERIA

EGYPT

PERIOD

EAST
AFRIC
A

Malaria

1886-1894

0.74

1909-1914

0.1 X

0.75

1916

5.38
9.84
1.68

Dysentery

1886-1894

0.37

2.52

1909-1914

0,10

1.01

.... ...............
.....
X'S0-IN 1)1
1909-1914

Pneumonia

1886-1894

0.50

So

1916

6.26

0.47

1917

8.46

1918

2.10

7'

. .......... 5.11
...

3.66

8.94

1916

0.68

0.62

1.50

0.36

1917

0.47

1918

0.32

0.96

1916

2.44

1 o,. l 1

1917

8.76

1918

12.58

tO.
06

1.31

0.96

1 01

o. 11

0 ', 7

1 0,77

Table 15 show how effectively typhoid, formerly the biggest threat to military
health, had been brought under control in the East African campaign.

The greatly

reduced mortality rates may be attributed to vaccination, as in all respects water supplies
and sanitation in East Africa were generally of a low standard. What is also remarkable
is the sharp rise in the lethality of malaria, dysentery and pneumonia which were many
times higher that in previous campaigns.

Interestingly, deaths from malaria declined

sharply in 1918, possibly as a consequence of the increased use of indigenous African

soldiers, but pneumonia skyrocketed.

Apart from typhoid, the British seem to have

""Curl in, Disease

and Emjwre, p. 141, Table 5.2, "European Troops in Algeria, Egypt, India and Great Britain Changing Mortal itN
1'rom Major Diseases 1890s to 1909-1913. It should be noted that the figure for dysentery includes all gastrointestinal diseases
while
those for pneurnonia similarly encompass all respiratory afflictions. 'rhe figures for East Africa have been drawn from Tables 7,8,10
and II above.

206

Mesopotamia.
Malaria
few
lessons
African
the
or
was
campaigns
applied
of
of previous
especially disastrous.
Table 16 deals with the disease mortality suffered by the French Madagascar
campaign of 1895. That country is fairly close to East Africa and was covered in dense
bush and forest with a large number of indigenous diseases.
16

TABLE
DISEASE MORTALITY

FOR THE FRENCH MADAGASCAR

EXPEDITION,

1895
(DEATHS PER 1,000 STRENCTH)562

DISEASE

DEATH

DEATH

PERCENT

RATE

RATE

OF

(monthly)

(annualised)

DISEASE

PERIOD

EAST
AFRICA

ALL

DEATHS
Malaria

32.16

72.00

24.12

1916--1918

5.38

9.84

1.68
Dysentery

3.57

8.00

2.68

1916- 1918

6.26

8.46

2.10
Typhoid

5.36

12.00

4.02

1916-- 1918

0.68

0.47

0.32
Other

3.58

2.69

8.00

1916- 1916

2.44

8.76

12.59
Total

44.67

100.00

33.50

Notes

I.
2.

The campaign lasted 9 months the figures have been annualised for comparative purposes.
The average ration strength was 15,000. "Other" figures for East Africa are for pneumonia only.

When comparedto East Africa, it can been seenthat the annualisedmalaria death
rate in Madagascarwas several times higher, while the dysentery rates were generally

""Curtin,

Disease and Empire, p. 188, Table 7.1, "Disease Mortality

for the Madagascar I xpeclition, 1895".

207

lower there. The French suffered from typhoid deathsmore than 5.9x more than the peak
British rate, but they seemedto have ignored establishedprophylactic measures.563
THE GERMAN EXPERIENCE

It is more difficult to come up with a comprehensive description of German

medical experiences as communications between Atellungen were often poor while
paper was in short supply and the majority of documentsdid not survive the war. From
the memoir literature, and despitebold statementsto the contrary, it is clear that they had
their fair shareof problems.
They began with a relatively good supply of doctors trained in tropical medicine
as a number of specialistshad been sent out to study sleeping sicknessshortly before the
war. There was a modem and well-funded researchinstitute at Amani that was far in

in the British colonies. But the flow of medicalsupplieswas

advanceof its counterparts
quickly curtailedby the RoyalNavy's blockadeandit becamenecessaryto manufacture
local substitutesfrom the outset. While ersatzgoodscouldhelp in a numberof areas,the
medical establishmentfaced major shortfalls in the stocks of medication,bandages,
surgicaltoolsandevenreferencebooks.
Isolation was damaging,but the major weaknesswas in transport. Apart from the
two railways, the Germansbeganthe war with only three motor vehicles. By September
1916, all of these mechanical means had been lost and they were almost completely
reliant on porters. This meant that the medical system was burdenedby the samemajor
limitations as the British while lacking even the inadequatemotor transport resources
possessedby their adversaries. This imposed tremendousobstaclesin the provision of
adequatehospital and treatment facilities once the campaign moved into the wilds of the
Rufiji and Kilombero valleys. Theseunfavourable conditions led to widespread
sickness,
with the hospitals usually full. There were repeatedoutbreaks of typhoid and virtually
564
illness.
nobody escapedserious
563Curtin, Diseases
and Empire, pp. 189-190.
'64Schnee,Deutsch-ostafrika im Wellkrieg, 230.
p.

208

The year 1917 was little better as the combination of insufficient food, exposure,
I,
inadequate clothing and severe exertion led to the same results as experienced by the
British: widespreadsicknessand the collapse of many individuals.

Given their lack of

transport and medical resources,the German responsewas simply to leave such casesfor
the enemy to capture, and thus to be forced to assumeresponsibility for their care and
from
River
Rufiji
to
During
the
to crossing
the
the
the
retreat
year,
evacuation.
courseof
into Portuguese territory, the number of unfit Europeans rose constantly and large
669
German
had
left
be
behind.
Just
the
to
the
to
of
colony,
evacuation
prior
numbers
Germans, or two-thirds of the available strength, and a large number of Askaris
565
surrenderedto the British from field hospitals.
Hospitals and their staffs suffered as much from the rigours of the campaign as
the fighting troops. As the withdrawal into the heart of the colony continued, more and
1916
due
for
lack
had
be
to
transport
to
action.
or
enemy
more equipment
of
abandoned
had seen a total of 10 field hospitals, 2 base hospitals and 2 convalescenthomes in
field
3
hospitals operating on
but
been
by
had
1917
to
the
this
operation,
reduced
end of
much reduced scales of equipment. The later march into PortugueseEast Africa saw
numbers reduced to 2 hospitals, with one being lost at the end of August 1918. The
combination of a dwindling number of trained staff and severe shortfalls of critical
supplies made the treatment of the sick and wounded increasingly difficult. Only those
incapable of walking or requiring serious treatment were admitted to hospital while
566
in
Losses during the campaign also hurt the
malarial caseswere treated the ranks.
medical establishment's ability to treat its sick and wounded; the number of doctors
declined from 63 at the outbreakof war to 13 by the end of 1917. During the latter stages
of the fighting in PortugueseEast Africa, medical officers were reducedto a scant6.567
Staff-Surgeon Taute noted the difficulties engendered by the long marches
characteristicof the latter year and a half of the war. First by early 1917, owing to food
565Schnee,Deutsch-Ostaftika im Welikrieg,
pp. 257-258.
5'6Taute, "A GermanAccount of the Medical Side the War in East Africa, 1914-1918", TanganyikaNotes
of
andRecords, pp. 7-8.
Henceforth,Taute, "A GermanAccount".

209

flour
600
400
been
had
daily
African
to
the
g
of
or
g of rice.
reduced
shortages,
ration
This meagreportion, which also had to provide for the soldiers' wives and servants,had
to provide sufficient nourishment for marches that averaging 25 krn per day with only
one rest day in seven. An extreme example followed the seizure of Ngomano on the
Rovuma River on 25 November 1917. Desperatefor food, the columns marched for 28
for
difficult
days,
halting
While
22
December.
this
exertion was
enough
straight
only on
the fit, the wounded, many suffering from severegunshot fractures, had to be carried on
hammocksthroughout - including the crossingof a number of rivers.568
The marching was especially difficult for the medical personnel who had to take
their place in the column all day long with their real work only beginning once a halt was
made and camp pitched. Frequently, they had to treat patients through the night, working
only by candle or firelight and very short of bandagesand medicines. Regardlessof their
569
had
be
dawn
before
tiredness,they
the next morning.
to ready to resumethe march
The campaign in Portuguese East Africa was only slightly less difficult

with

regard to medical supplies as the Germans were able to capture large amounts of
material, notably quinine, from the Portuguese although they lost a great deal to British
action. The reduction in strength meant that there was less pressure on quinine stocks,
570
from
from
Africans.
the Europeans who always suffered more
particularly
malaria than
However,

the increasing toll

on endurance through

overwork

and underfeeding

undermined health, while malaria, dysentery and blackwater fever remained constant
Typhoid made repeated visits and numerous deaths through exhaustion
571
were noted.

companions.

From these accounts,it is apparentthat sicknesswas an overwhelming aspectof

the campaign. Coming up with an accurateset of casualty figures is more problematic as
the Germans were widely scattered and often in small detachments while usually
M7Taute, "A GermanAccount", 3.
p.
5" Taute,"A GermanAccount", 9.
p.
5" Taute, "A GermanAccount", 8.
p.
570Schnee,DeutscheOstafrika, 357.
p.
571Schnee,DeutscheOstafrika,
pp. 374-375.

210

withdrawing under pressure. The overall health situation in 1917 was very bad with
Africans and Europeans suffering heavily from malaria, increasingly of the tertian
variety. Dr Taute estimatedthat, in the worst affected areas, up to 80 per cent of the
Europeanswere sick from a variety of causes,with blackwater fever (64.2 per cent of
deaths) being the greatestnon-battle killer. Dysentery proved to be a bigger problem
572
African
being
the
with
soldiers and carriers
particularly afflicted.
Dr Ludwig Deppe, one of the more accurateand precise observersof the East
African campaign,claimed that von Lettow had overturnedthe existing rules of tropical
hygiene and behaviour. Noting the ability of the force to march for up to 12 hoursper day,
including during the hottesthours,as well as the apparentlack of problemsresulting from a
lack of mosquitonets or from drinking untreatedwater, he proclaimedthat von Lettow had
"eine neueTropenhygienebegrundef'."' While the achievementsof the Schutztruppedid
demonstratethe resilience and toughnessof the human body in very adverseconditions,
such a pronouncementmust reviewed in light of the available evidence. His claims of the
generally good health of the German troops by the armistice seemsto be based on an
unfavourable comparisonwith that of the British, who were by then suffering from the
ravages of the Spanish influenza epidemic. He did admit that tropical diseaseswere
unavoidablein such a place as East Africa althoughno figures were given to supportthese
statements.574Schneealso madea similar comment,noting that the Germanswho had lived
in the tropics for sometime did hold up much better than the South Africans 57' There is
.
probablya degreeof truth in both statements,in that the Germansshowedmore stamina,but
the British also noticed that younger men, who were more numerousin their ranks, often
had difficulty standingup to the climate. Nonetheless,an analysisof the availablestatistics
showsthat the Germanwhites sufferedfrom diseaseon a massivescale.
Interestingly,Deppe did provide casualtyfigures, for Europeansonly, although he

madeno attemptto analyseor comparethem with others. When examined,his own

numbersunderminehis assertionas they showthat diseasewas an importantfactor. The
572Taute, "A GermanAccounf',
pp. 11- 14.
373Deppe,Durch Afrika, 149.
p.
374Deppe,Durch Afrika
pp. 149-152-

211

* breakdownwas asbelow:
TABLE 17 - GERMAN EUROPEAN CASUALTIES 1914-1918576
Nature of Casualty

Numbers of

% of Total

% of Total

Casualties

Casualties

Served 3600

Killed in action

285

7.9

7.9

Wounded in action

875

24.1

24.3

51

1.4

1.4

190

5.2

5.3

Prisonersof War

1742

48.0

48.4

Missing in action

486

13.4

13.5

3629

100

100.8

Of whom died of wounds

Died of disease/injury

Total

If the numbersarebrokendown further, it can be seenthat deathsfrom disease(190)

representsome36 per cent out of the total deaths(526). It is not known how many of those
listed as prisonersof war or missing in action died as a result of either wounds or disease,
but even if thesecategoriesof lossesare assumedto have suffered total deathratesof, say
10 per cent and 25 per cent respectively, then the extra losses would amount to
(174+122=296). This would give a new mortality figure of 822; diseasewould still
represent23 per cent of this increasedfigure.
The most authoritative figures available, given by Major Boell in his semi-official
Die Operationenin Ostaftika, presentthe following casualtiesover the entire campaign:

s"3Schnee,Deutsch-Ostafrika, 230.
p.
576Deppe,Durch Afrika, 156.
p.

212

577

TABLE 18 - GERMAN CASUALTIES 1914- 1918

Nature of Casualty

Askari

% of Total

Europeans

Casualties

% of Total

Porters/

Casualties

Boys

(A)

(E)
Killed in action

354

7.4

1200

7.1

1000

Wounded in action

-874

18.4

3669

21.6

80

1.7

Died by accident

23

0.5

Died of disease

277

5.8

508

3.0

5000-6000

Prisonersof War

2718

57.1

4275

25.1

Missing in action

4510

26.5

Deserted

0.00

2847

16.7

4762

100

17009

100

6000-7000

Of whom died of wounds

Total

UsingBoell's figures,it canbe seenthattotal deathsby disease(277Europeans

and
508 Askaris)represent37.74per cent and 29.74 per cent (734 and 1708)respectively.
Theseshouldbe examinedin relationto the approximately3,600 Europeanand 14,600
be
"'
best
The
that
the
Askariswho servedin the Schutztruppe
can
the
war.
over courseof
both
in
huge
types
to
is
of
they
that
the
comparison
the
were
saidof
porters
mortalityof
fighting
than
their
Clearly,
times
comrades.
their sufferingswere many
greater
soldier.
faced
by
health
Bothsetsof figuresarebroadlycomparable
to
the
problems
severe
andpoint
all participants,Europeanand African.
By way of comparison,the representativefigures for the major participantsis shown
in the table below:

577Boell, Die Operationen,p. 427. Boell's calculation of the total Askari casualtieshas omitted the wounded, hencethe revision from
13,430to 17,009.
3" Boell, Die Operationen, 429.
p.

213

COMPARISON OF BRITISH AND GERMAN CASUALTIES

TABLE 19 - COMPARISON OF BRITISH AND GERMAN CASUALTIES579

British

German

(Total Force)

(Europeans)

Ratio

German

Ratio

German

(Askaris)

German
(Askaris)

(Europeans)

British

British
% Killed in action

0.60

9.87

16.5x

8.22

13.7x

% Wounded in action

2.02

24.28

12.Ox

25.13

12.4x

% Died of wounds

0.20

2.22

1LIX

% Died of disease/injury

1.82

8.33

4.6x

3.48

1.9x

% Prisonersof war

0.06

75.50

1258.Ox

29.28

488.Ox

30.89

154.3x

19.38

-I

% Missing in action
% Deserted

0.20
-

0.06

From Table 19, it can be seenthat GermanEuropeanswere 4.6x and Askaris were
1.9x as likely as the British to die from diseaseor accident.This may be attributed to the
Germans
drugs
to
the
as well as to their
reducedquantities of
and medicationsavailable
difficulties in evacuating the sick and wounded. It may also representthe continued
attrition sufferedin trained medical personnelthat worsenedasthe campaignwore on.
It is notable that the German casualties were much more likely to have been
limited
in
have
been
battle
the
British,
than
the
more
maimed
a result of
which may
Europeans
Askaris
by
German
loss
Interestingly,
the
and
manpowerpool.
rates suffered
were broadly similar, apart from desertion,indicating that the risk was evenly shared.
SUMMARY

Disease and ill-health were factors of overwhelming importance to both sides

'" Mitchell and Smith, Official History Medical Services Casualties
Services
StatislicOH,
Medical
Casualties
Medical
and
and
Medical Statistics,p. 254, Table 4, "Classification and Percentagesof Casualties1916-1918",p.254. The table in the text has been
calculatedwith casualtiesbeing shown as a percentageof the total strength deployedduring the period 1914-1918for the Germans

214

throughout the campaign. The levels of malaria, dysentery and pneumonia were well
for
British
in
the
those
at
while,
campaigns,
concurrent
above
suffered previous or other
least, the menaceof typhoid had beenlargely solved. The high levels of sicknesssuffered
by followers as comparedto troops is notable and were a result of their poor diet and much
more limited health care.
Claims the Germanshad founded a "new tropical hygiene" must be taken with
it
is
Finally,
high
levels
indicate
the
clear that significant
caution as
otherwise.
of sickness
failures in administration and the medical servicesoccurred during General Smuts's term
of command and these contributedto the alarmingly high levels of morbidity suffered by
the troops and followers.

and 1916-1918for the British. The Germanfigures have beenderived from Boell, Die Operationen, pp. 427 and 429.

215

CHAPTER 7- THE OPERATIONS OF JUNE - DECEMBER 1917

VAN DEVENTER'S APPOINTMENT
VAiile Smuts's involvement in Hoskins's sacking remains unclear, it is certain
580
The
his
that he played a major role in nominating van Deventer as
replacement.
decision was taken quickly by the War Cabinet and he was confirmed as commander-in581
British
in
it
May.
Initially,
the
amongst
officers who
popular
chief
not
early
was
believed it to be another political appointment that unfairly deprived Hoskins of his

command.
Nevertheless,van Deventer appearsto have had little involvement in these
him
had
Smuts
he
Although
to
a number of
with
served
and
machinations.
owed much
times from the Anglo-Boer
temperament.

War onward, they were very different in outlook and

He had much more of the professional soldier's outlook and was not

known for his politics.

Transvaal Artillery

Van Deventer had begun his military career as a gunner in the

in 1896, rising rapidly through the ranks. A battery commander by

the outbreak of war in 1899, he had served under Smuts's command during the last year
of the war before retiring to his farm after the peace settlement.

He was recalled to

service as a colonel in 1914 to participate in the South West Africa expedition, but was
given the task of suppressing Maritz's troops when the Rebellion broke out in late 1914.
Having completed that mission, he was promoted brigadier in command of the Upington
column in the advance against German South-West Africa, finishing as a divisional
582
commander.

After a brief period in South Africa, he was called upon to lead a mounted brigade
in EastAfrica under Smuts. He was rapidly promoted to commandthe newly-formed 2nd
5sOMSRobertson,1133/4611,23April 1917.Letter Robertsonto Smuts; 1/33/47/2,23 April 1917,Telegram Smutsto Botha; MS
Smuts,Box 100,9 April 1917,Letter Robertsonto Smuts.
-"' CAB 23/2,1 May 1917,War Cabinet Meeting 128.
Biography,
Volume V, Pretoria: Human Sciences Research Council,
-"2 Beyers, CJ and Basson, JL (eds), Dictionary
ofSouth Aftican
1987, pp. 809-8 10; South African National
(A)PI/4139
1/1, Record ofService
Defence Force, Personnel Archive
and Reserves, AG(l)
General Sir Jacob Louis van Deventer in the Union Defence Force.
ofLieutenant

216

Division in April 1916 which he then commandeduntil January 1917, when he and his
troops were withdrawn to South Africa on medical grounds. After journeying back with
Smuts, he had barely a month at home, before being asked to assumecommand of the
East Africa Expeditionary Force as a lieutenant general.
Van Deventer had learned a great deal about the need for efficient organisation
and administration during his time as a divisional commanderin 1916. Although he was
most at home leading mounted columns, he had learned that the difficult conditions in
East Africa demandeddifferent solutions. He was much more of a fighter than Smuts
and did not shrink from battle. His methods were direct and while he could not be
described as an administrator, he did use his staff more effectively. One British officer
describedhim:

"Van Deventer is calm and collected, divulging his plans to none, not even his
staff. He is cunning as an old fox and does not make up his mind till the last
moment. Then he acts like lightning; up to that moment he appears dense and
slow. To him a decision is final; there is no swerving, no delay, no alteration of
plan. 11583

He believed in using mounted and irregular troops whenever possible although he

recognisedthe limitations imposed by the tsetsefly. He had realised the futility of trying
to out-manoeuvrethe Germansin such difficult terrain and that they had to be defeated
militarily. Also important was his recognition that their chief vulnerabilities were lack of
replacement manpower together with low stocks of food and munitions. Disease
remained the deadliest foe, but better planning and support for the troops could help
reduceits ill-effects.

These factors led van Deventer to adopt his predecessor'saggressiveapproach

together with greater tactical ingenuity. While the power of the machine gun and the
strength of defensive works meant that the casualtieswould generally be higher for the
... Meinertzhagen,Army Diary,
p. 180.

217

be
defender,
he
that
than
the
could
an effective option.
attrition
attacker
also realised
However restricted his manpower and equipment reserves were, they were still vastly
NCOs.
in
European
Lettow,
and
to
those
officers
particularly
superior
of von
Accordingly, this led to a new strategyof hard-hitting wherever possible, with the aim of
inflicting the maximum casualtieswhile lowering morale and the will to continue. It was
in
lines
German
the
particular
communication,
and
of
combined with a campaignagainst
584
food
eliminating their
growing areas.
On arriving in East Africa on 29 May, the new commander-in-chief was given
585
instructions.
The CIGS came immediately to the point:
very explicit

fact
it
is
"In view of requirements
that
the
theatres
essentialto release
of
and
other
by
of your
at earliestpossiblemomentvesselsabsorbed supplyand maintenance
force,His Majesty'sGovernmentattachgreatimportanceto early terminationof
during
importance
impress
I
remainder of
you
campaign.
upon
must also
campaignof limiting demandsfor tonnageto minimum.

086

The attitude of the British Governmenthad changedconsiderably since it had so

lightly sent Smutsoff to take the whole of GermanEast Africa. There was now the
realisationthat von Lettow had to be defeatedmilitarily and that simple occupationof
groundwasnot sufficientfor victory. Furthermore,with the growingpressuresof the UBoat campaignandthe war in Europe,it recognisedthat the campaignhad to be wound
up asquickly aspossiblein orderto releaseboth shippingandmanpower.
Van Deventer was authorised, if necessary,to conduct operations in Portuguese
East Africa, although the Portuguesethemselvescould be expectedto give little useful
assistance. Van Deventer was instructed to try to prevent the enemy from leaving
German East Africa for the south as it would prolong the campaign. There was no

WO 33/953, TelegramsD 2, No. 1901,11 June 1917,Telegram G 843, van Deventerto CIGS, p. 97.
WO 33/953, TelegramsD 2, No. 1868,29 May 1917,Telegram 0 666, Hoskins to CIGS, p. 85.
5'6WO 33/953, TelegramsD 2, No. 1852,22 May 1917,Telegram 340907, CIGS to
van Deventer,p. 81.

218

implied
it
but
that
they
the
was clear
explicit mention of any other campaign objectives,
destruction of the enemy force and its ability to fight. 587
Although Hoskins had done much to reorganise the British forces, the exertions
heavy
had
toll.
taken
the
a
very
and overwork of
preceding year

Furthermore, the

Germans retained interior lines in a territory still some 300 miles square and difficult to
588 For
Mahengeto Liwale
sevendays' march for the Germans

penetrate.

example,

was a

Kilwa
Lindi,
from
Mahenge
for
force
have
to
British
the
still
to
either
or
a
while
moved
lack
The
distance
from
have
Liwale,
the
taken
of
same
number
of
weeks.
some
would
589
disadvantages.
maps and difficulty of the terrain amplified these
BRITISH STRATEGY IN MID-1917

By mid- 1917Britain was running out of money to fund the war and the
four
first
In
increasingly
U-Boat
the
months of
successful.
unrestricted
campaignwas
1917, British lossesto submarineshad risen from 109,954tonnes (35 ships) in Januaryto
516,394 tonnes(155 ships) in April. 590At this rate, Britain would soon be starvedinto
submission,as replacementships could not be built fast enoughto offset the losses. The
War Cabinetwas underno illusions asto the severity of the crisis and it closely scrutinised
the amountof shippingbeing usedto sustainthe overseastheatreswith the aim of making
591
the maximum reductions. By early July, the figures showedthat, overall, the shipping
employedon the overseastheatres,-some328 ships, was the equivalentof lessthan two
months' lossesto submarinesat peak levels and about nine months at the earlier rates. East
Africa representedonly 10.7per cent of the shipping employed,but it took up 34.3 per cent
592
horse
hospital
35
In
troop
21.9
of
and
any
event,
ships
ships.
shipsand
per cent of
5'7W033/953, TelegramsD2, No. 1852,22 May 1917,Telegram 340907, CIGS to van Deventer,p. 81.
'" Van Deventer,Lt Gen J 1, Despatchfrom
the General Commanding-in-Chief British Forces In East Africa, Relating to Operations
From 31y'May to I' December,1917, dated21 January 1918,publ ished in London Gazelle, "Supplement7,No. 30611,5 April 1918,
HenceforthVan Deventer,DespatchL
F,
9172Van Deventer,Despatch1, pp. 172-173.
Marder, Arthur, From Dreadhought to ScapaFlow: 1917 Year of Crisis, Volume 4, London: Oxford University Press,1966,p.
102.
"' CAB 23/3,29 June 1917,War Cabinet Meeting 172. Smuts,
now a memberof the War Cabinet, had beenaskedto compile a
report and stated"In view of the submarinecampaignandthe urgentnecessityto economiseshipping,die CabinetCommitteeon War
Policy hasaskedme to investigatethe possibility of reducingthe quantity of shipping allottedto overseasexpeditionsoutsideFrance.
m CAB 23/3,29 June 1917,War Cabinet Meeting 172,29 June 1917,Appendix I, "Shipping Allocated to OverseasExpeditions
Outside France".The figures for troop and horseships are 24 of 70; for hospital ships 7 of 32.

219

for
in
the year was
deal
the
shortfall shipping
representeda great
of capacity,particularly as
593
War
Cabinet's
did
be
8,000,000
This
the
to
eye and at a
tons.
estimated
not escape
subsequentmeeting it considered:
"The urgent necessityof clearing up the position in East Africa once and for all, and
the fact that [sic] the conductof operationsby us in that quarterwas monopolising
the carrying power of thirty-five ships,were [sic] pointed out. It was agreedthat, in
be
brought
in
to a close at the earliest
that
the
this
might
order
quarter
campaign
found
General
it
to
to
that
send
was
possible
possiblemoment,every reinforcement
594
do
him
Van Deventershouldbe sentto enable
to this.
This was in partial responseto a plea by van Deventerfor more artillery and limited
595
despite
But
the rhetoric, thesewere only
reinforcementswere subsequentlyauthorised.
tiny forces and could not transform the situation in East Africa. They reflected conflicting

imperialists
hand
the
there
the
who
were
political pressuresregardingwar aims; on
one
lookedto displaceGermanyasa colonialpowerandon the otherthose,who despitetheir
596
key
Force
to victory. For all, concerns
misgivingsaboutstrategy,sawtheWestern
asthe
aboutthe stateof Ireland,conscriptionand war-wearinessloomedlarge. Any political
hopesfor a quick andeasyvictory in EastAfrica hadbeendashedby the realisationthat it
would remain a lengthy and hard-foughtcampaign.
THE OPPOSING FORCES IN JUNE 1917

The deploymentsof the British had changedconsiderablysince van Deventer had

left in January. Only the Nigerian Brigade, based on Kibambawe, was situated on the
Rufiji front, while further to the eastat Iringa there was a small column of two battalions.
The Belgians had begun to arrive in the Kilossa area and had pushed several battalions
forward to Kidodi although it would be several weeks before the planned brigade was
"3 French,Lloyd George Coalition, 77.
p.
5" CAD 23/3,23 July 1917,War Cabinet Meeting 193.
595WO 33/953, TelegramsD 2, No. 2002,22 July 1917,TelegramG 447,
van Deventerto CIGS, p. 129; CAB 23/3, War Cabinet
Meeting 194,24 July 1917. Although severalartillery batterieswere also agreed,only a mounted regiment was sent to East Africa
"96French,Lloyd George Coalition,
pp. 62-64; Turner, British Politics, p. 202.

220

Songea
Ubena
between
Northey's
force,
and
with a
spread
complete.
now reinforced,was
detachment operating in PortugueseEast Africa against Abt von Stuemer. Brigadier
General O'Grady commandedthe Lindi column while Brigadier General Hannyngton
Far
Kilwa.
the
from
to
the
brigade-sized
and
north,
two
west
controlled
columnsoperating
equivalent of a brigade of Belgians was pursuing Naumann and there was a small central
KAR
form
in
two
Morogoro.
Local
the
raised
of
newly
reserve at
reinforcements
battalions were due to be fielded by the end of Junewhile other battalions were in various
before
be
it
they
but
formation
in
the
considerable
period
a
would
stages of
areas,
rear
597
from
due
help
Further
be
for
also
was
operations.
could
consideredremotely ready
South Africa with two new reformed battalions scheduledto arrive at the end of the same
598
force
4,500
landed
The
had
Portuguese
at the port of
of
month.
a new expeditionary
599
field.
Palma, but it would be severalmonthsbeforethe troops were ready to take the
The Germans too had redeployed markedly in the preceding months. They
based
Lettow,
0suruppen
in
at
the
von
two
under
remained
major groupings, with
Liwale, and the Wesuruppencommandedby Tafel at Mahenge. Exact numbers are not
available, but an estimate of 6200 and 2500 based on known losses would seem
four
had
1500.600
The
0suruppen
4100
leaving
main
and
reasonable,
combat strengthsof
detachments;in the far south, At Looff had six companiesand two guns between Lindi
held
the ground
Portuguese
border;
At
G6ring
the
and
with sevencompaniesand a gun
Rufiji;
had
Mpotora
Otto
the
Hanforce;
Abt
on
companies
a
gun
seven
and
near
against
in
its
direct
four
Kommando
the
had
chiefly
control,
while
guns under
six companies and
Mpotora-Liwale area.601

The Wesuruppenunder Tafel also had four substantialAteflungen radiating some

70 miles west, north and east of Mahenge down to Likuju, about 50 miles from Songea.

TheywereAbt vonBrandiswith threecompaniesandtwo gunsnearKidodi; Abt Aumann

with threecompaniesand a revolvergun alongthe RuhudjeRiver; Abt Linckewith five
"7 Van Deventer,Despatch1, 172.
p.
"8 WO 95/5329,War Diary Norforce, Entry 8 June 1917.Telegram G 775, van Deventerto Northey, 7 June.
5" Cann, "Mozambique
and German EastAfrica7, pp. 131-133.
600MS Boell, N14/30,26. Kapitel,
pp. 2032 and 2044; On I April 1917,they had numbered6,534 and 2,854 respectively,with
effective battle strengthsof 4,419 and 1,712. The revised figures are an estimate.
601MS Boell, N14/30,26. Kapitel,
pp. 2031-2032.

221

finally
body
Norforce;
facing
the long absent
two
the
and
companies and
of
guns
main
Abt Wintgens of six companies and gun now well to the north and west of the other
German forces.

PLANS FOR THE GENERAL

OFFENSIVE

OF 1917

The generaloffensive for 1917 could not begin before the end of the rains and the
subsequentdrying out of the countryside in late June at the earliest. Van Deventer had
very little time to settle into his new command, but, fortunately, Hoskins had set the
necessaryplanning, reorganisationand regrouping in train. In conducting its appreciation
of the situation, GHQ had identified five possible lines of advance for 1917. The first
was the Dodoma-Iringa-Mahengeroute in the west. While it was the healthiest, it was
'
also the longest and the country between the highlands of Iringa and Mahenge was
difficult and low-lying. Furthermore, only a small portion of it could be made passable
for MT and it was highly vulnerable to flooding in the wet season. The second option
was slightly further eastalong the line Kilossa-Kidatu-Mahenge. It was shorter and more
practical for MT than the first approach,but it too suffered from its vulnerability to the
rains.

The third approachfollowed Smuts' route acrossthe middle Rufiji going via
Mikesse-Kibarnbawe-Mahenge.Plunging as it did through the Mgeta and Rufiji valleys,
it was by far the most physically difficult to support. The nightmarish conditions of the
wet seasonwere still all too evident while the country south of Kibambawe was a wild
tangle of bush and almost totally impracticable for MT. The fourth option was an
advance on the line Kilwa-Liwale. It had the unhappy reputation of having the
unhealthiestclimate of all, although it did offer the advantagesof being close to a large
proportion of the enemy's forces and to a first-class harbour. There was also the
probability that sicknessrates would fall with the onset of dry weather. The final choice
was the approachLindi-Massassi. Its chief advantagewas that it offered a direct line of
advancethat could potentially cut off any retreattowards the Portugueseborder, although
the country was very difficult.

Health, too, was a problem, and Lindi harbour had

222

H
iz
4k

lz
,

%.

-Z

E-I

r. i

1;1

at
t-6

4.

It

dab

E5
e

l
'O

qr

l<

%
ia

It
, it

%%Z '.
we

r,

...

tu

I
40

(z

z4

*.

it

M"-I

'R
IC

Ic

It

. 3L

ab.

g
T-

00

Nb

considerable
602
shipping.

limitations,

by
be
a proportion
used
and could only

of the available

The attractivenessof Kimbambawe as the base for a renewed offensive, which

had never been high, now waned significantly.

Apart from the unhealthiness of the

climate, the difficulty of the countryside, that had so recently been heavily flooded, made
the supply of the existing small garrison very difficult.

The need simultaneously to build-

from
it
least
desirable
build
the
to
option
up supplies and
roads and river-crossings made
van Deventer's point of view and fortunately Lettow's redeployments made it possible to
drop this line of approach. A contributing factor was the insufficiency of MT; there were
simply not enough vehicles to support concurrent advances across the Rufiji as well as
from the coast.603

General Northey added his views, believing that the Germanswould only move
into PortugueseEast Africa if pushedthere, and, until the situation east of Mahengewas
break
He
force
for
to
there
the
westward.
cleared up,
another raiding
was
potential
proposed similar converging simultaneousthrusts against the enemy, but, conscious of
the time neededto get Murray's Column back and suitably rested, he expectedthat his
604
importance
Deventer
Van
the
of
the
agreed,realising
advancewould start after
others.
having Northey attack Mpepo and Mtarika as it would draw off enemy forces on the
Ruipa line prior to the arrival of the Belgians. Once they were ready, Northey could then
shorten his line by handing over the Ruhuje sector to them. By that stage, Murray's
605
in
be
column would
restedand ready to participate the generaloffensive.
This would leave the SongeaColumn at about 1,000 rifles with 16 machine guns
and two mountain guns, and the Fort Johnston Column with a similar amount. While

thesecolumnswould be readyto startat the end of June,Murray's Columnwould need

extratime to reachSongeaandwould not be readyto advancebefore7 July. This would
provide another 1,000 rifles, giving him over 3,000 men advancing on two axes. He also
6'2 Van Deventer, Despatch I,
p. 173.
603CAB 44/9, p. 67.

6" WO 95/5293, War Diary GHQ, 26 May 1917,Appendix A195, TelegramNF 2957, Norforce to Hoskins, 25 May.

223

in
battalions'
be
July, but neither
that
two
arriving
noted
worth of reinforcementswould
606
be
for
field
before
August.
Likuju would also form an important
the
would
ready
staging base,as both a wireless station and small airfield were establishedthere.
At the end of May, Norforce had been stretchedout over enormous distancesand
concentration was impossible in the short term. The pursuit of Naumann together with
the countering of von Stuemer's threat to Nyasaland had drawn off all Northey's
available reserves. However, as the campaign season approached, he began to receive
more resources to achieve his difficult task. On 29 May, some 50 miles south of the
Central Railway, Murray's Column was released from the pursuit of At

Naumann and

began its long march home. It was not expected to reach Neu Langenburg until 15 June
and a considerable distance would then need to be covered until it could reach the
forward areas.607

On 10 June 1917,van Deventersignalledhis plan back to London. It reflected

the changedstrategic situation as well as the imperatives given in his own instructions
during the previous month. The German re-deployment of its main body south from the
Rufiji River to an areaopposite Kilwa showedthat Colonel von Lettow was aware of the
possibilities of being cut-off and wished to maintain an escaperoute to PortugueseEast
Africa. It also reduced the importance of an advance from the north, and the need to
supply a major force acrossthe Rufiji Valley. The plan also reflected the likely increased
importance of Lindi and its hinterland as a meansof cutting off the enemy's withdrawal
to the south.The elimination of Naumann's raiding force was a priority as the pursuit was
tying up Belgian and British forces that could be better used againstthe main body of the
Germans.

Van Deventer'splan was as follows. First, to catch up with and destroyAbt

Naumannas quickly as possiblein order to rest and redeploy the forces involved.
Second,to launch a limited attack aroundLindi in order to securethe high ground
"-'CAB 44/9, pp. 43-44.
" WO 95/5329,War Diary Norforce, Entry 9 June 1917.
6'" WO 95/5329,War Diary Norforce, Entry 29 May 1917.

224

Ruipa
Third,
better
to
the
attack
the
town
to
supply.
water
and secure a
surrounding
River position from the direction of Iringa in order to deny that rich food-growing areato
the Germans. Fourth, to generate an offensive from Kilwa as soon as sufficient
General
Northey
During
be
brought
these
was to
preparations,
up.
reinforcements could
back,
Murray's
Colonel
having
his
force
Songea,
column
and then to
got
at
concentrate
608
he
advanceeastwardsas soon as was ready.
The commander-in-chiefunderstoodvon Lettow's likely responseto his plan:
"The establishment of considerable food depots in the MASSASSI (GGF. F. 8.c.)
force
in
PORTUGUESE
together
the
enemy
considerable
of
a
area,
with
presence
EAST AFRICA, appears to point to his eventual retirement by this line. One of
force
breaking
his
be
through
to
to
therefore
main
prevent
my chief aims must
into

PORTUGUESE

EAST

AFRICA,

as that

might

necessitate a new

609
,,
campaign.

As he beganto finalise his plans, Generalvan Deventer neededto liaise directly

his
disappointment
his
Hiding
Belgian
Hugyh6.
Colonel
over the
with
counterpart,
slownessof the Belgian build-up, he agreedto meet Huyghd at the Belgian Grand
Quartier Gineral at Dodoma on 18 June. Prior to the conference,Huygh6 was requested
to senda column of 1,200rifles as quickly as possibleto Iringa in order to operateagainst
the Germanson the Ruhuje-Ulangariver line betweenMpepo and Malinje. This column
from
be
forward
Tafel's
the Ulanga Valley
troops
the
would
chargedwith
clearanceof
betweenthosetwo settlements. However, in the circumstancessuch speedyaction was
not possibleas the reorganisationof the Force Publique was taking more time than
anticipated. Shortagesof transport and the needto shuttle troops along the railway meant
610
be
July.
date
that the earliest
at the end of
of readinesswould

6"W095/5293, War Diary GHQ, 10 June 1917,Appendix A74, TelegrainG 843, van Deventerto CIGS, 10 June.
"9 WO 95/5293,War Diary GHQ, 10 June 1917,Appendix A74, TelegramG 843, van Deventer to CIGS, 10 June.
6'0CampagnesColonialesBeiges,III,
pp. 127-128.

225

This was unfortunate as van Deventer was very anxious to start operations as well

I
bolster
In
force
breaking
the very
German
to
to
order
west.
prevent another
raiding
as
thinly stretched net around the Schutztruppe, the British commander-in-chief
Huygh6 to provide at least a single battalion for the Iringa front.

pressed

Despite having strict

instructions not to allow his forces to intermingle with those of his allies, Huygh6 agreed
611
do
to
so.

The conferenceproved to be a successand both men acceptedto the new strategic

Tafel
Mahenge
The
Belgians
taking
the
the
on
and
securing
of
were
given
role
plan.
British
MT
be
dependent
they
and carriers to move their
on
although
area,
would
between
itself
Colonel
Tytler's
The
Brigade
Sud
to
column at
supplies.
establish
was
Iringa and Norforce at Lupembe ready for the advanceeastwards. In the vital matter of
food
its
fleet
MT
it
British
the
that
and that
via
supply
supplies, was agreed
would
Belgian pioneers would assist in the improvement of routes to the forward forces. The
Brigade Nord had already commenced operations on 15 May with three battalions in
612
rhiS
pursuit of At Naumann while the remainder continued their concentration.
formation would now take over sole responsibility for the pursuit and destruction of that
force with all British columns being withdrawn into reserve. The Belgian supply base
was fixed at Dodoma and the provision of pioneers and other troops on road building
613
taskswas also agreed.
After this meeting, van Deventertelegraphedhis instructions to both Northey and
Tytler on 20 June. He informed them that the Belgians would take over the conduct of
the operationsagainstNaumann from 25 June and that they would also provide a column
of 500 rifles to move from Dodoma via Iringa to Mtua to operateon the Malinje-Mpepo
line. A further column of 1,200rifles would be made available in the secondhalf of July
and it would probably advance on Mahenge from a base at Kilossa and Kidoti. Once
Naumannhad been dealt with, a further force of 700 rifles would then take over the area

611CampagnesColoniales Beiges,III, pp. 128

andAnnexeNo 27, Instructions au CommandantHubert, 19 June 1917,pp.329-330.
612CamPagnesColonlales Beiges, 111,pp. 297-298. Instruction
au commandantde la Brigade Nont ler Bureaz4No 96, dated 15 mai
17.
613CampagnesColoniales Beiges,III,
pp. 129-130.

226

Effectively,
the
this
turn
be
by
Tytler
would
elsewhere.
used
who would
controlled
614
whole of the Mahengeoperationsover to the Belgians.
Following these discussions, van Deventer issued his final timetable for the
day
begin
in
PEA
the
27
June.
A
same
would
on
preliminary operation
offensive on
The
Mwembe.
its
begin
Norforce
main phasewould
on
advance
column would
when a
by
Ruipa
Column
the
Iringa
2
July,
position supported
attacking
commenceon
with the
250 rifles from Norforce moving from Lupembe against Mpepo. The principal attack
Mtua,
follow
4
Linforce
July
with an assaultplanned there on
on
moved
on
when
would
the following morning. Also on 5 July, Hanforce would hit the Nguara River position.
In the south, Norforce was to send its main column from Likuju towards Liwale on 5
July. Finally, on the middle Rufiji, a battalion-sized force from the Nigerian Brigade
61
5
Furthermore,
10
July.
9
Msswega
the
leave
for
Kibambawe
or
on
on
an
attack
would
Belgians would take over responsibility for the hunt of Abt Naumann as well as providing
8
July,
Kidodi
from
Kilossa
towards
500
on
to
move
an advance column of
rifles
616
into
position.
although the remainderwould take sometime to get
GERMAN STRATEGY FOR 1917

Imperial German strategy for 1917 continued along the same lines as the earlier
stages of the war.

Solf and his colleagues continued to argue for a greatly expanded

African empire largely at Belgian and Portuguese expense, they still lacked any means of
directly influencing events there. This led to policy being decided in a partial vacuum,
isolated from the events on the ground and totally dependent on military
Europe.617

success in

Colonial German strategy for 1917 was much simpler: it was to continue
forces.,
Von
Lettow
for
face
in
long
the
surviving
of superior enemy
as
as possible
614WO 95/5293, War Diary GHQ, 20 June 1917,Appendix Al 16, TelegramG 976, BGGS to Northey and Tyder, 20 June.
`5 WO 95/5293,War Diary GHQ, 27 June 1917,Appendix A140, Telegram G 91, Genstaff to Hanforce, GOC Lindi, OC Iringa and
GeneralEdwards,27 June.
`6 WO 95/5293,War Diary GHQ, 28 June 1917,Appendix A142, Telegram G 107, van Deventer to CIGS, 28 June; WO 95/5329,
War Diary Norforce, July 1917,Attachment by War Office, M02(B), 29 June 1917,"East Africa - ProspectiveAdvance".
6" Fischer,Germany's War Aims,
pp. 317-319 (1917) and 586-590 (1918); Fischer, Griffnach der Weltmacht,pp. 258-260.

227

bulk
his
he
himself
the
that
to
the
particularly
as
of
colony,
realised
need not restrict
Portuguese
infrastructure
lost.
had
been
He
that
the
were
also
understood
supply
already
the weak link in the situation as their unpopularity amongst their colonial subjects and
military incompetencemade for easy pickings. The sheer size of the potential area of
operationsmade his main problem one of supply; he neededto protect the key mtama and
rice growing areasfrom the British whilst they ripened. He resolved to continue his war
inflict
defeat
delaying
local
to
tactics
on any British column
of
opportunities
while using
that strayedinto him.
He had already reducedhis forces substantially during the wet seasonthrough the

elimination of a large part of his lines of communicationorganisation. He was very

concernedabouthis food supplyandtook greatpersonalinterestin the growingof crops
for the comingyear. As ever,he realisedthat he wasoutnumbered,
but resolvedto give
anyweakor isolateddetachments
a hardknockasthe opportunitypresented.
THE INITIAL

ADVANCE BEGINS

The question of effective numbers remained a major concern, as sickness

constantly depleted the units and those who were well enough to be forward were still
weak. Two South African battalions were due to arrive, but not before the end of July.
Furthermore, although they bore the titles of units experienced in the 1916 campaign,
they were essentiallynew with only partial training at this stage. Finally, it is interesting
to note that theseunits could not be sent to GeneralHoskins, but were so easily available
to Generalvan Deventer.618
Before the general offensive could begin, it was necessaryto make a number of

adjustmentsto the variouscolumnsand to launchseveralpreliminary operations. The

Kilwa force pressedfor reinforcementsbeforebeing launchedinto the attackonto the
Nguara River position; although Hannyngtonhad two columns of 1200 and 500
respectively,a shortageof officers and a large numberof inadequatelytrained troops
618CAB 44/9, P. 72.

228

to
this
Van
Deventer
a
send
and
promised
their
agreed
quickly
effectiveness.
reduced
619
South African and up to two KAR battalions prior to beginning the advance.
Most of June was taken up with preparationsfor the advance and the return of
left
Johnston
Norforce
Fort
Headquarters
from
on
to
the
the exhausting chase
west.
units
24 June for Songea, arriving there on 2 July. The remainder of his troops made their way
to the jumping-off position around Kitanda and LikuJu in the north while a separate
;striking
620
Africa.
force under Colonel Shorthose prepared itself to clear Portuguese East
Northey too, had insufficient trained manpower, and a number of units had to be
duties.
lines
to
communication
of
relegated
,advance.621

By 28 June, all was in readiness for the

THE PURSUIT OF NAUMANN

Naumannhadout-marchedandeludedhis pursuers,reachingthe southernshores

but
down,
him
hard
had
to
in
late
Lake
Victoria
June.
The
Belgian
tried
slow
columns
of
difficulties in obtainingsuppliesandpoorintelligencemadethis difficult. Finally, on 29
June,a combinedAnglo-Belgiancolumncaughtup with the Germansat Ikoma,a few
had
fought
to
battle
inland
the
from
lake.
An
attackers
and
the
was
miles
unsuccessful
Naumann
turning
to
This
and
escape,
a
chance
gave
withdraw with numerouscasualties.
After
Central
Railway.
Irangi
for
his
headed
Kondoa
the
Abteilung
and
south,
now
dealing with their losses,the pursuersthen spentmuch of July following the Germans'
trail. "'

THE MAIN ADVANCE BEGINS

GeneralO'Grady faced a difficult situation in early June 1917. Operationsin the
full
fight.
in
had
of
the
good condition and
previous month
enemy was
shown that
"' WO 95/5293, War Diary GHQ, 4 June 1917,Appendix A23, Telegram 0 884, Colonel Grant, Kilwa to BGGS, 4 June; War Diary
GHQ, 7 June 1917,Appendix A 47, TelegramG 776, Genstaff to Colonel Grant Kilwa, 7 June.
620WO 9515329,War Diary Norforcc, Entries 24 and 29 June 1917. Stevens'Column now consistedof V/1", 2d/l" KAR, and 1'
SAR. 2d SAR was employed on lines of communication security back to Lake Nyassa621CAB 44/9, p. 47.
622CampagnesColoniales Belges,111,pp. 56-57 and 105-107.

229

!r
VC

41

->-

d tv

_4f'

IF

'Ir
/, -/ 11211,1 -1-

q?

4-

- -...

its,

ex

tz

211

;
10
Ci a

0
-4

Cp
%"

; .

4,,IZ,,

Ira,
Nya
rF

Q4

Id

*,
.-Z

I0,

-.--- -Zz

-- --- -

/ep
C
i

Equally, the brittleness of the new KAR battalions was also evident and reinforcements
would be necessaryfor any successfuladvance. This meant securing the mouth of the
Lukuledi River, which emptied into Lindi Creek about one mile south of the River
Mohambika about five and a half miles south-westof Lindi town. The small village and
port of Mingoyo, which lay along the Mohambika, was vital to any advance as it was
here that lighters could offload supplies from ships in the anchoragedirectly to the trolley
line. This was a peacetimeline that had been built to service the local plantation and its
capture would enable the British to bypass nearly 25 km of swamps that faced any
623
The other advantageof the confluence of the two rivers on the
overland advance.
'
Indian Ocean was that the water was deep enough for monitors to approachto within a
624
fire
mile and a half of Mingoyo and provide heavy
support.
O'Grady had a battle strength of about 3000 rifles with which to dislodge an
625
600
In reality, the situation was worse as
estimated six companieswith around
rifles.
626
16
there were seven companiestotalling over 900 rifles,
machine guns and two guns.
All in all, it was a difficult position to crack, particularly as the approacheswere so
vulnerable to observationand fire.
At

Wahle was deployed as follows: Abt Looff with two companies was nearly

five miles south of the mouth of the Lukuledi near Namunda while the much strongerA bt
Kraut musteredfive companies,of which three were at the Ngurumahambaposition with
Wahle and the remaining two companies at Majani.627Conscious of the dangers of a
frontal assault,O'Grady decidedto make a bold pincer movement by land and sea. In the
north, a column would move out from the northern edges of the Lindi defences and
march west for the Lutende Hills. Once there, it would turn south and head for Naitiwi,
thereby cutting off the Ngurumahambaposition and threateningthe trolley line along the
Mohambika River. At the same time, an amphibious move would be launched up the
Lindi Creek, landing two battalions which would then march for Tandamuti Hill before
623MS Boell, N14/32,28. Kapitel, 2247; Boell, Die Operationen, 333.
p.
p.
624CAB 45n3, Sketch 100,"Operations Lindi June 1917".
at
625CAB 44/10, p. 9, footnote 1.
"6 Boell, Die Operationen, 333. Wahle had total
p.
strengthof 900 rifles, 17 machine guns, and 2 guns.
a
627MS Boell, N14/30,26. Kapitel, 2055
p.
and 28. Kapitel, p. 2247.

230

turning north towards the Mohambika River. Once there, they would cut the trolley line
link
and
up with the northern column.
Working closely with the naval commander-in-chief, O'Grady drew up a detailed
plan of action. The Germans were alert to the threat posed from the sea and watched the
approaches carefully. Therefore, a night move was decided upon; a blacked out force of
boats and lighters would move silently past the German outposts at high tide before the
moon had risen, and would land between the Rivers Mohambika and Lukuledi at the
Kenjengehe pier marking the southern terminus of the trolley line. Once landed, a twobattalion force would move initially south-west towards Mkwaya and then make the
decisive turn north towards Mwreka. 628 However, this did not account for the presence of
A bt Looff south of the River Lukuledi in the area of Namunda, rather than fw ther west as
believed. 629

The operation began on 10 June and the boats moved undetected past the forward
German posts. However, observation posts in the hills had detected boat movement and
reported it to Wahle. He immediately ordered his troops closer to the river, while telling
At

Looff to march north and rejoin the main body.

Similarly, he recalled the other

forward companies, leaving only a single one facing Lindi

at the Ngurumahamba

position. His intention was to hit the British from two sides; his main body would move
from the north, while Looff converged from the south with the aim of driving O'Grady
back into the water. 630

British planning had been good and the leading troops landed safely at
Kenjengehe. The lead battalion was ashore by early evening and disembarkation
continued through the night. With the rising of the moon, visibility was good and early
on 11 June the Konigsberg gun opened fire on HMS Thistle leading to a lively
engagement. Wahle was now sure that this was no feint and ordered Kraut to attack the
landing site with two companies. However, by this time O'Grady's troops had pushed
62"CAB 44110,pp. 8-10.
629Boelt, Die Operationen, 333.
p.
6 Boell, Die Operationen, 336; MS Boell, N14/32,28. Kapitel,
p.
pp. 2250-2251.

231

PE

7D

;4 I'A

Al

1w

f IL
"Lo

Iq A.TAw

0
J4

klA

-4;
Of 40

I r'

A Aua -1,4

Dia-

Aolh. -4&,
4wq

Hill,
to
the
Tandamuti
had
five
inland
south-west of
taken
well
and
miles
nearly
Mingoyo-Mrweka. The secondbattalion had turned north-west near Ziwani Ridge and
headed for Mrweka. Both sides collided there after daybreak and a heavy engagement
followed with the British being held in the Mohambika valley. Both sides reinforced,
631
into
back
drive
the
he
Wahle
believed
the
sea.
attackers
could
still
although
Looff was the key to the plans, but despitefrequent signals he and his troops were
in
Wahle
day
fighting
the
be
The
received
to
afternoon
and
all
on
seen.
raged
nowhere
the highly unwelcome news that a second,undetectedBritish column had marchedout of
Lindi moving via the Lutende hills and was now in possessionof the key track junction at
This immediately rendered the Ngurumahambaposition untenable as well as
threatening his forward forces concentratedat Mwreka. He immediately ordered the

Naitiwi.

five
Narunyu,
back
from
Ruaha
baggage
to
train
miles up
the
evacuation of
vulnerable
the Mohambika River.

In the meantime,the battalionat TandamutiHill was orderedto the main fight at

Mrweka. Its piecemeal arrival late in the day ended chances of German victory.

632

Looff s appearancearound 1600 did little to changematters and the whole position was
633
had
At
for
Looff
by
darkness.
This
actually run
the
entire story,
evacuated
was not
634
Elsewhereon
broke
into the rear of that battalion, severalcompaniesof which
and ran.
the northern flank, another KAR battalion securedNaitiwi while an Indian unit marched
into Mrweka via Majani. The German forward position was lost and a withdrawal to
Narunyu becamenecessary.
The result of the action of 10-11 June was to advancethe British line well inland.
The Germansnow occupiedthe line Tandamuti Hill - Mohambika Valley - southwestof
Ruaha,while the British faced them from positions along Mkwaya - Mrweka - Majani Naitiwi 635 However, neither side possessedsufficient strength to defeat the other and
.
631MS Boell, N14/32,28. Kapitel, pp. 2252-2255.
632MS Boell, N14/32,28. Kapitel, pp. 2258-2260; WO 95/5330, History of I'/2"dKM PP. 14-15.
633MS Boell, N14/32,28. Kapitel, p. 2262.
634Boell, Die Operationen,pp. 337-338. The Germansdeployed 741 soldiers of whom five were killed, 20 were wounded and two
were missing. 3/2' KAR lost two killed, nine wounded and 10 missing.
633Boell, Die Operationen,p. 338.

232

I Itupo

IL

io

FA

&i'

W1'
, A(
Iq A.TAO I

M,

4Uf1

&. V*44

\
I

//
t,

. v

A""

-' %

231c,

for
had
been
Success
for
further
close
offensive operations.
reinforcementswere needed
both sides: the audacious and well-executed landings at Kenjengehe had caught the
battalions
by
two
the
but
the
Germans off-balance,
this was squandered
splitting of
between Mrweka and Tandamuti. Wahle, in his quest for a decisive victory, had failed to
his
flank
had
Lindi
so much
northern
weakened
and
maintain adequatesurveillance of
had
light,
been
Losses
the
being
fighting
it
fell
with
not
necessary.
that
without any
636
British
31
Germanssuffering 62 casualtiesas opposedto
side.
on the
While the advancesfrom Kilwa and Lindi were underway, Colonel Murray and
his column were reaching their destination after a long and arduousabsence. Having had
for
Wiedhafen
they
Mwaya
to
where
to
a
steamer
on
they
embark
marched
a week's rest,
landed on 25 June. Murray then moved up to Songeawhere he assumedcommandof the
force. It was joined by a Rhodesian battalion on 4 July, while the remainder of his
637
former column remainedin SongeaasNorthey's generalreserve.
On the northern flank, a smaller force was preparing to move against Abt
Aumann, estimated at having some 300 rifles and five machine guns, at Mpepo. Ibis
for
Mahenge
German
thereby
draw
intended
and
to
reinforcements
any
off
move was
Capt
had
instructed
from
Iringa.
Northey
ftu-ther
Belgian
the
north
assist
advance
Dickinson with three companiesto move from Lupembe in conjunction with Lieutenant
Colonel Carbutt's moving north from Kitanda.
On the German side, on 27 June, Abt Lincke had four companieswith a battery
638
his
decided
build-up,
Lincke
British
to
Likuju.
Unaware
the
the
attack
scaleof
of
near
former camp. His troops moved through the bush on 28 and 28 June, putting in a strong
fight
fortunate
break
failed
in.
break
Lincke
the
to
that
to
off
as two
was
assault
was
639
him
ff
As soonthe areawas clear,
0 .
strong enemy columns were nearby and nearly cut

the various units convergedon Likuju for rest and final preparations. Before the
636NIS Boell, N14/32,28. Kapitel, pp. 2270-2273.The Germansfought with 741 soldiers and lost five killed, 20 wounded, and two
missing. Td/2d KAR lost three dead,nine wounded and four missing.
637WO 95/5329, War Diary Norforce, Entry 3 July 1917; CAB 44/9, p. 48. Three companiesof RNR and one BSAP were numbering
400 rifles and eight machinesconstitutedthe reservein Songea.
638NIS Boell, N 14/32,Doell, 29. Kapitel, pp. 2621-2622.
639CAB 44/9, pp. 48-49.

233

battle,
disbanding
final
his
began,
Northey
to
several
of
order
change
made one
offensive
depleted European units and converted the remaining personnel into Lewis gunners.
They were then distributed amongst the various African battalions to provide additional
firepower and experience.640
While thesepreparationswere underway, Tafel was making alterationsto his own
deployments, forming an enhanced reserve at Mahenge in early July. This was
imminently
he
Kilwa
important
expected
at
and
needed
as an offensive was
particularly
to react quickly to changes. One of the major problems of the disperseddeployment was
that it took two to three days for reports from the forward areasto reach him and a similar
64
1A
back
Kommando.
information
to
stronger reserve
period was required to pass the
would weaken the forward fighting strength, but would allow him to react to the main
threat more quickly.
As June progressed, the weather improved and the country began the rapid
processof drying up. While this was very helpful for supply purposes,it also meant that
sources of spring and surface water were also disappearing at the same time. This
imparted a degree of urgency to General Beves, temporarily in command at Kilwa. He
ftirther
inland
Germans
through lack of water before
that
the
considered
might withdraw
decided
he
decisive
battle
be
fought.
Accordingly,
to launch a pre-emptive attack
a
could
whilst the enemy was still in easy range of his own supply bases. He therefore proposed
to attack as soon as reinforcementshad arrived, but without waiting for more transport.
He saw the advantagesin fighting close to his own base and trying to inflict heavy
casualties as greater than the disadvantages of an immediate, prolonged advance
642
thereafter. To that end, on 13 June,he askedGHQ for permission to divert a battalion
from Lindi as O'Grady had already achieved his initial objectives and his own
reinforcementswere still chasing Naumann. This was agreed by van Deventer and the
latter increasedthe reinforcementto include a battery of 8 Stokesmortars.643

" WO 95/5329, War Diary Norforce, Entry 3 July 1917.

641MS Boell, N 14/32,29. Kapitel, pp. 2621-2622.
642WO 95/5293, War Diary GHQ, 14 June 1917,Appendix A 95, Telegram0 955, Beves to BGGS, 13 June.

234

Beves's assessmentproved to be correct as the Germans had begun to evacuate

in
Hills,
Kiturika
located
June.
Patrols
the
them
forward
still
the
their
end of
positions at
African
As
South
the
logistical
force's
range.
the
operational
and
within range of
began
General
Beves
to
than
began
to
anticipated,
arrive somewhatearlier
reinforcements
had
four
had
No
I
battle.
He
three
for
his
dispositions
columns:
the
now
coming
make
battalions, eight guns and eight 8 Stokesmortars; No 2 had three battalions, six guns and
four Stokesmortars; and No 3, under Colonel Taylor, had about a battalion. His plan was
to converge on the enemy's positions near Mchakama, and bring him to battle while
644
faced
force,
He
his
to
the
to
as von
a
substantial
westward.
retreat
attempting cut off
645
hold
the area.
Lettow had left Abt von Lieberman someten companiesand two guns to
General Hannyngton decided to use both his columns in a converging movement
from the north, while the third, smaller column under Colonel Taylor would march southits
jumping-off
Leaving
from
Kilwa,
the
route.
southern escape
cutting off
west

point

linking
for
before
Column
dark
6
July,
No
I
up
seven
south-west
miles,
marched
after
on
While
its
forward
these troops attacked the next morning
elements.
with
reconnaissance
in the thickly wooded positions, No 2 Column was engaged in an arduous approach
by
No
I
flanking
The
from
the
and
a
movement
successful
attack was
north-west.
march
Column forced the Germans to retire to Mtshikama and the two columns linked up the
646
day.
next

Hannyngton then took the opportunity to rebalance his columns, drawing No 2

back to Kirongo in preparationfor another flanking move while No I Column took over
Mnindi. After four days of reconnaissanceand planning, a move by that force coupled
led
defenders
by
3
Column
from
No
to evacuate and move
the
the
with pressure
east
17
Nos
I
Late
July,
10
further
Narungombe.
the
to
on
evening
of
and 2
some miles
south
Columns linked up and readiedthemselvesfor the next stage.

643CAB 44/9, pp. 73-74.

6" CAB 44/9, pp. 74-75. No I (Orr) had four battalions; No 2 (Grant) had three and a half battalions; No 3 (Taylor) had two
battalions.
"s MS Boell, N14/32,28. Kapitel, pp. 2287-2288.
646Orr, "Rumbo to the Rovurnar, pp. I 10-111.

235

Orr was given command of Nos I and 3 columns for the planned fight at
Narungombe, although sickness reduced one of his Indian battalions to company
followed
by
Column,
No
3,
No
I
18
July
The
going
with
strength.
moves commencedon
for the water holes at Kihumburo, about six miles from Narungombe. Apart from its
it
in
latter
had
the
made
vital
such a parched
of
which
garrison,
place
a good source water
land. Indeed, failure to take it would inevitably force a return to Kibumburo, and, if its
water proved to be insufficient, then a move even further back might be necessary.
By 1730 hours that day, contact between the opposing sides was lost, and No I
Column bivouacked about two miles short of Narungombe. Captured prisoners revealed
the presenceof eight companiesand two guns around the water pool with the possibility
of reinforcements arriving from the south.

Receiving permission to employ No 2

Column in his operation, Orr plannedto use No I Column to engagethe enemy from the
front while No 2 moved against the enemy's left flank and No 3 on his right flank. A
bombardment by mountain guns and Stokes mortars would commence as soon as
daylight permitted.
Abt von Lieberman had its position across the road and facing generally north.
On its right flank was a substantialswamp while to the left was a hill coveredin virtually
impenetrable bamboo and bush. The approacheshad been carefully cleared and were
backed by three sets of trenches and rifle pits. The overall frontage was about 1800
647
yards. The attack went aheadearly the next day. Only limited progresscould be made
to the front while the two Ranking moves went ahead,with No 3 Column making good
progresson the right. After severalGerman counter-attacks,Orr put in heavy attack in an
attempt to break through the defences. A newly raised KAR battalion in No 3 broke and
for a time the column was nearly overwhelmed. However, a final push on the left by No
2 Column broke into the defencesand savedthe situation. Unable to progressfurther and
unwilling to give up their gains, the British dug in for the night preparatory for another
attempt in the moming.64'

647Ridgway, "With No 2 Columie', 1,pp. 22-23.

236

Before dawn, patrols found that the enemy had evacuatedhis remaining positions
joined
by
holes.
They
Narungombe
Orr
the
were
water
and
vital
and
promptly occupied
Force HQ and, in light of the casualtiesand need for water, a further advancewas called
back
12
No
2
Narungombe,
to
in
No
I
miles
was
sent
while
of
off.
remained possession
Rungo where water was plentiful. Liebermann now moved back to Mihambia, where he
few
been
former
had
joined
by
irate
Lettow
the
a
miles
away
when
who
only
was
an
von
had evacuatedthe most important position for many miles.
Narungombewas one of the hardestfights of the campaign and one of the first in
which the Germans had been fought out of rather than manoeuvred out of position.
Casualtieson both sides had been heavy and the fighting had been conclusive before the
degree
It
four
Lettow
the
to which the
also
revealed
arrival of von
extra companies.
and
P/3
rd
in
KAR
No
3
Column
battalions
KAR
was only
new
still needed experience;
high
it
lost
its
disintegrated
had
trained
a
proportion
officers and a
when
of
partially
and
severebush fire swept through its position; its collapse had accountedfor the difficulties
in which the column found itself that afternoon.649
A strong position was now developed at Narungombe with a line of forward
front
defences.
in
later,
A
the
two
of
week
on 28 July,
outposts establishedabout
miles
No 2 Column moved south-west towards the enemy post at Mssindy, about ten miles
distant. After a number of encounters,the British occupied that place and now had a
650
jumping-off
from
further
secondwell-watered
point
which to conduct a
advance.
Norforce began its advance on 2 July with the aim of catching up with At Lincke.
It was located the next day, some 23 miles from Likuju at the important track junction
linking Songea-Liwale and Songea-Mahenge. A KAR battalion was ordered to attack,
while the remainder of Murray's Rhodesians arrived at Songea, completing the arduous
65
1
The move on Lincke took a few days to develop as Murray endeavoured
return march .
to cut Lincke off and the deep ravines and densely vegetated valleys were difficult to
6"WO106/273,
History
KAR,
70.
P.
of3
"9 Orr, "Rumbo to the Rovume,
114-115.

pp.
00 Ridgway, "With No 2 Columrf', 1,pp. 26-27.

237

KilwaK!

vinjL

S,
14A(dFOACIE
iA&

M.

N[poto

ASA

-f

lwaK-. Siwa. u
Kiwat

N ahands

gom'"

ar

Ad

If,

k; I

A,

Na. anj I
ti

ngo

4 ny
Bw
Chini,

lz
-

/I

116

&tun
Niciu

:?

. Ir iA.

wile

II

..

c
%%

itoneno

Sudlo
'o

<
Z-;

--

dandiL
'Vrl

2!"

1-I,

CCU

MZ-SSZLSSI

C12iw

-atcL

NamyaMbp

wm b

It
v V

T
'U3.
rl-

;L

6/7
Germans
July,
the
the
Finally,
the
of
night
encirclement complete on
with
negotiate.
hastily evacuated the very strong position, leaving behind a considerable amount of
darkness
they
Escaping
and
of
good
use
made
with a number of casualties,
equipment.
Mahenge
himself
to
Lincke
bush
and
sick
evacuated
was
paths.
some newly cut
652 Further south, a battalion under Colonel Shorthose
devolved
Poppe.
onto
command
Fort
Johnston,
to
in
150
Mwembe,
of
preparatory
north-west
miles
some
on
was closing
battalion
July,
5
German
the
After
attacked
there.
on
patrol
ambushing a
attacking
653
defenders.
Mwembe the next day and drove out the
The main advancecontinued with the retreating enemy moving north rather than
Good use was made of the slender air

towards the expected destination of Liwale.

bombing
German
Likuju
based
the
and
new
the
reconnoitring
at
aircraft
resourceswith
favoured
flammable
huts,
highly
in
based
dumps,
Supply
a
grass
were
usually
positions.
target and the Likuju aircraft as well as another based in Mwembe were able to bum
654
MT,
beginning
to
the
Murray's
outpace
supply
as
situation,
was
several.
advance
food,
lack
Poppe's
Owing
had
to
the
to
of
time
arrive.
yet
ordered some
previously,
tactics were strictly delaying and he took care not to becomedecisively engaged.
On 13 July, Northey succeededin having his orders changedfrom an advanceon
Liwale to one on Mpondas, as he wanted to concentrateon destroying the enemy rather
than seizing empty territory. His requestto continue the simultaneousdrives againstAbt
Aumann in the north and Abt von Stuemer in the south were approved by van Deventer,
Once
build-up
he
Belgian
that
the
slowly.
very
was
proceeding
although
warned
Mpondas had been reached,the commander-in-chiefwould consider the situation before
issuing ftirther instructions.655

By the middle of the month, Norforce was fully engaged in all three areas,
pushing back the various enemy detachments. The welcome news of the arrival of the
"' WO 95/5329, War Diary Norforce, Entries 2 and 3 July 1917.
6" WO 95/5329, War Diary Norforce, Entries 6 and 7 July 1917.
... WO 95/5329,War Diary Norforce, Entry 8 July 1917.
6" WO 95/5329,
655 WO 95/5329,
300 BGGS

War Diary
War'Diary

Norforce,

Entries

Norforce,

Entry

10 and II July 1917.

12 July 1917. Telcgrarn

NF 4139,12

to Norforce.

238

July Norforce

to BGGS;

Entry

13 July, Telegram

first Belgian contingent at Iringa was received on 15 July together with reports of an
impending attack at Mpepo.656 This was countered a few days later with the
disappointing news that the Belgians would not now take over the Lupembe- Mpepo
KAR
Deventer
to
send
another
promised
sector as previously agreed, although van
657
in
battalion from Iringa recompense.
The SongeaColumn advancedsteadily and fought a number of minor, but timefrom
Mpondas,
three
to
while
miles
about
consuming, rearguardactions reaching within
658
from
Mahenge.
him
Madaba
to
Tafel had brought reinforcements with
on moving
Further to the north, the column now commandedby Lieutenant Colonel Fair had forced
back At Aumann between Mpepo and Mkapira on the right bank of the River Ruhudje
by 26 July.659 Progresshad been good considering the relative closeness.in strengths
between attacking and defending columns: Hawthorn's Songea Column had about 900
faced
650
Fair's
350
Aumann's
600,
Lincke's
to
around
of
while
rifles as compared
660
400
Mpepo and Shorthose'swere pushing back von Stuemer's
or so.
The time spent in thesestatic locations was not wasted, as daily patrols were sent
find
his
disrupt
locate
the
to
to
supply
and
as
enemy
well
as
out
water sourcesand roads,
denying
burning,
the enemy
Deforestation,
through
used
as
a
means
of
also
was
system.
joined
flight
by
2
Column
No
defences.
to
the
of aeroplanesat
was
a
coveredapproaches
Mssindye and later by the mounted troops of the 250' (Indian) Cavalry, who immediately
began raiding the enemy lines of communication, often with considerable success,and
661
disrupting the enemy's supply arrangements.
Elsewhere, von Lettow was encountering other difficulties as the Germans'
control of the Makonde Highlands was becoming unsustainable. Largely supported by
the British, the Makonde tribe rose in active revolt against the colonial regime. Two

656WO 95/5329, War Diary Norforce, Entry 15 July 1917.

657WO 95/5329, War Diary Norforce, Entry 24 July 1917.
619WO 95/5329, War Diary Norforce, Entry I August 1917.
659WO 95/5329, War Diary Norforce, Entries 26 July and I August 1917. Abt Aumann was about 350 strong with three companies.
"0 WO 95/5329, War Diary Norforce, Entry 31 July 1917.A capturedfield state,signedby Lincke on 24 July, gave a fair indication
of the strengthof the Germanopposition with some63 Germans,540 Askaris and 428 carriers on the strengthof his Abteflung.
661Ridgway, "With No 2 Column", 1, p. 28.

239

German
to
in
despatched
to
protect
as
well
as
rule
re-assert
attempt
an
were
companies
fighting
Von
Mrweka,
hearing
in
On
food-gathering
the
the
at
of
area.
parties
vital
Lettow also sent six companies to reinforce Wahle, with another three drawn from Abt
Otto being moved to the Kilwa

front.

Marching

with

the reinforcements,

Kommandeur set up his headquarters at Nahungo on 24 June.

the

His intention was to

detached
Linforce
from
that
breakout
Lindi
to
portion
of
any
attack
well
as
as
prevent a
had
Lutende,
he
days
later,
Five
to
the
area
not
of
which
on
moved
came within range.
662
At
his
began
He
been occupied following the British advance,and
sent
preparations.
K6hl
Lutende
Hills
Abt
Chappuis
the
the
to
two
with
of
while
area
with
companies
von

663
A subsequentpersonal reconnaissanceby
Naitiwi.
three companieswas held around
von Lettow on 8 July of the Naitiwi position concluded that the chancesof a surprise
664
it
attack were unfavourable and without no chanceof successwas possible.

Back at Lindi, O'Grady wantedto develophis earlieradvance,but neededmore

troops. He had been promised anotherKAR battalion, but as it would not arrive before 5
July, large-scale operations were out of the immediate question. However, reports of
June.
investigate
29
his
him
in
led
to
Lutende
to
the
on
own
patrols
order
patrols
area
Not realising the presenceof enemy reinforcements, a half battalion had successfully
immediate
defenders
The
30
June.
the
Lutende
was
and,
as
response
attacked
on
blow
his
fresh
The
back,
three
Lettow
companies.
with
streamed
counter-attacked
von
fell heavily on the Indians and they were driven out of their newly acquired position.
Casualtieswere heavy, including the loss of their commanding officer and two machine
665
hands
once again.
guns,and the Lutende position was in German
Following this setback,O'Grady, always an aggressivecommander,proposedthat
Linforce advancealong the road running from Lindi to Massassi,in order to cut off von
Lettow's escaperoute. Believing that the Germanswould not hold Liwale but would fall
back to Massassi and thence into Portuguese East Africa, he pushed for an initial
662Boell, Die Operationen,p. 338.
663MS Boell, N14/32,28. Kapitel, pp. 2287-2288.
6" MS Boell, N14/32,28. Kapitel, pp. 2292-2293.
665Boell, Die Operationen,p. 339. The British lost 73 dead, 19 wounded and 18 taken prisoner,while the Germanslost nine killed
and 26 wounded.

240

666
Massassi.
to
far
Mahiwa,
onward
move
the
subsequent
with
a
as
road as
advancealong
Given the temporary exhaustion of Hanforce and the closenessof this view to his own
thoughts, General van Deventer assented to O'Grady's wishes and authorised the
667
his
command.
reinforcement of
On the German side, despitehis inability to attack, von Lettow choseto remain in
the Lutende area for a number of days, using the opportunity to gather supplies and
difficulties
learning
Finally,
the
the
on
of
rich stocks as possible.
consume as much of
faced by Abt von Debermann further north, he marched off with five companiesand two
668
batteries,leaving Wahle a single companyas reinforcement.
The latter part of the month of July was spent ferrying troops and the allimportant carriers from Dar-es-Salaarninto Lindi. The reinforcements were substantial,
battery
Stokes
battalions,
infantry
the
mortar
and
three
mountain
a
consisting of
battery.669 While all the organisational.and administrative preparationswere underway,
Brigadier-GeneralO'Grady conductedhis reconnaissancesand developedhis plan.
As van Deventer's operationsprogressedthroughout July and the Belgian buildThese
began
Belgian
to
friction
between
British
the
commands
mount.
and
up continued,
in
Iringa
due
lack
British
the
troops
to
area and the slow arrival of
were chiefly
a
of
battalion
By
that
Belgian
the
was under
end of
month, one
planned
reinforcements.
670
interposed.
direct British commandand the various columns were
Originally, both governmentshad agreedthat the British would set the strategic
direction of the campaign and would set the mission for the Belgians within a distinct
zone of operations. Within this framework, the Belgian commander-in-chief was
responsiblefor the detailed deployment and operation of his own units and columns. He
had been specifically ordered not to allow the formation of mixed allied columns and to
6" WO 95/5293,War Diary GHQ, 13 July 1917,Appendix AS and A6, Note on Situation at Lind! and Note on the relative importance
of the Kilwa (A) and Lindi (R) lines ofadvance by Brigadier-General OGrady, 13 July 1917.
667CAB 44/10, p. 7.
"s Boell, Die Operationen, p. 339.
669CAB 44/10, p. 7.

670CampagnesColontalesBeiges,111,3,p. 141.

241

maintain a distinctly Belgian area of command. Given the pace of deployment and the
had
been
ignored
Colonel
Huyghd
this
and
now came
pressure of operations,
principle
under national pressureto rectify this situation. This took place at a meeting in the British
GHQ in Dar-es-Salaarnfrom 20 to 23 July 1917, whose aim was both to resolve these
disputesand to settlethe more prosaic questionsof supply and transport.
After much discussion,it was agreedthat Mahenge would remain the immediate
Belgian objective. To that end, the Colonne Hubert, presently en route from Iringa to the
Kilombero Valley was instructed to reinforce Colonel Tytler's column along the line of
the River Ruipa.

The Brigade Sud, having concentrated at Uleia was then to advance to

Kidatu and start preparations to cross the River Ruaha. Once in position, Tytler was to
withdraw his column and Hubert was to come under Belgian orders once again. It was
also agreed that the British-installed
Belgians. 671

telegraph line would

be turned over to the

As ever the question of transport loomed large, with the British agreeingto supply
African rations to the Belgians, while the purchaseof livestock and rice was left to their
Royal Commissioner at Kigoma. British MT would move supplies forward from the
railheads at Dodoma and Kilossa to the forward line of the Ruipa. From there, the
supplieswould be moved forward by British carriers and mules. Owing to a shortageof
Belgian personnel, the British agreed also to provide overseersto supervise Belgianrecruited carriers.
For their part, the Belgians agreedto take responsibility for the improvement of
the road leading from Kilosa to Kidoti and later extendedthis onward to Kidatu on the
Ruaha. They also had to meet the challengesof bridging the many rivers en-route, with
the greatestbeing the requirement to allow eight-tonne vehicles to cross the Ruaha and
Ulanga Rivers. As these were, respectively, 70 feet wide and 10 feet deep and 300 feet
672
20
feet
deep,
they representeda major task.
wide and

671CampagnesColoniales Belges,III,
pp. 140-141.

242

THE ADVANCE

IN AUGUST 1917

By early August, the situation was as follows. The Germans remained divided
into two major groups under Tafel and von Lettow. The Kommandeur kept a strong force
Liwale
it
Mihambia
or Massassi with columns
either
at
where
could easily reinforce
facing both Hanforce and Linforce.

The British were still unsure as to his ultimate

intentions; whereas he had initially appeared to be headed for Portuguese East Africa, he
Furthermore,
in
Mahenge-Liwale
be
the
they believed that
to
area.
now appeared
staying
the reinforced and aggressive operations from Lindi
673
be

resistancewould

would draw him in and that

substantial.

Van Deventer was taking no chancesand considered the clearance of AN von

Stuemer from Portuguese territory and the denial of the rich southern districts of the
German colony to be essential.

By early August, the first stage had been achieved.

Shorthose's Column was now moving on Tunduru, an important fanning area and road
junction.

Also germane to van Devnter's plan was the reintroduction of mounted troops

into the campaign as the reformed South African mounted regiment began to arrive in
East Africa. 674

Van Deventer paid a personalvisit to Lindi in early August, conferring at length

with O'Grady and setting out his policy for a further advance. After the lossesin Juneand
July, the commander-in-chiefordered that more detailed reconnaissancesbe carried out
beforehandand that a frontal attack was not to be repeated. Insteadhe wanted a turning
movement,using the fordable right bank of the River Lukuledi, againstthe strong enemy
675
To achieve this, O'Grady had just received substantial reinforcements
positions.
togetherwith a large number of individual replacementofficers and NCOs and eight new
Lewis gunsper battalion.676

67'CampagnesColoniales Belges,III,
pp. 142-143.
673WO 95/5294, War Diary GHQ, 7 August 1917,Appendix A4, Telegram G 650,
van Deventerto CIGS, 7 August.
674WO 95/5294, War Diary GHQ, 8 August 1917,Appendix B, Telegram G 675,
van Deventerto CIGS, 8 August.
675CAB 44/10, p. 19. Van Deventer left Lindi on 5 August
676CAB 44/10,
p. 7.

243

The new plan requireda numberof changesin the British deploymentsand O'Grady
decided to thin out his forward positions with the bulk of his troops being concentratedin
two mobile columnsand only weak garrisontroops holding Lindi. Colonel Taylor's column
of 1,200 rifles was to assembleat Mkwaya to make the outflanking march and to seize a
position on the German line of retreat "so as to compel the enemy to leave his
entrenchmentsto fight his way southwards." O'Grady also planned to lead a column
himself, about 1200strong,that would provide the main punch for the attack,which was set
for 9 August.

The arrival of the last troops on 6 August meant that all was in readinessfor the
advanceto begin three days later. Intermittent aerial bombing of Tandamuti Hill was used
in an attempt to focus Gen-nanattention on that place. Taylor's Column set off in midafternoonof 9 August, crossingthe Lukuledi near Mkwaya, and bivouackednear Kiwambi
Hill. The next morning the fire plan, which included all the availablegunsincluding the two
monitors, openedup on TandamutiHill.

O'Grady's Column then marchedwest between

the hill and the trolley line, hoping to cut off the retreat of the defenders. Opposition was
slight, but the densebush made for slow going and it was not until dusk on 10 August that
the leadingbattalion reachedthe trolley line abouttwo miles west of the Loop. The column
then haltedand dug in for the night. In the meantime,Taylor's Column had continuedon to
its objective,a point one mile eastof Narunyu, and beganto entrenchitself. The plan was
now to wait for the Germansto try and break out. At about 1800 hours, a probe from
Narunyu was driven off but it was the only engagementof the night and the next morning
the defendersfound the enemyhad withdrawn around their trenchesand were now astride
677
line
Narunyu.
the trolley
at
The advance to Narunyu had been successfullyaccomplishedand the Germans
forced out of their strongposition at TandamutiHill. However,the next phasewould prove
more difficult as the Lindi force was beyond the rangeof the ships' guns and sicknesswas
rising after heavy rains. Van Deventer signalled O'Grady on 14 August again to avoid
frontal attacksand use flanking movementson the Narunyu position. However, this was
677CAB 44/10, pp. 19-21; WO 106/273,History
of3 KAR, pp. 19-20.

244

--

I er
p

rl.
CIN

LV
.

CV,

-M

7T

.
,

;z

fv

cp

-7...

ll

q%I

C4

Im

" -.

W.
. ....

'X ..

-7z
IMP
_u

lh

Id

2N

-k
%

I-

ib

%%
47,
)b

06
-1

ol"It

4a

11

easier said than done, as the ground was exceptionallydifficult since the southernside was
protectedby wide and impassableswampsformed by the River Lukuledi and the Narunyu
in
dense
largely
The
face
sisal plantation, while the small
covered
a
stream.
eastern
was
stretch of open ground near the trolley line was lined with trenchesprotected by a sisal
abattis. The trencheson the low ground were flanked on the northern side by a steep
escarpment,and on the top of the high ground and sheerslope were further entrenchments
amid thick bush. However, in the north the country was more open and held the possibility
of a flank march.
While these operationswere underway, von Lettow had been very active. On
hearingof the British attackon TandamutiHill on 3 August, he had left Mihambia with five
companiesand severalguns. He had:
"decided to join General Wahle with some of the companies from Ndessa and
failed
had
bring
the
that
at Narungombe;a decisivesuccessby
perhaps
off
operation
678
,
an unexpectedreinforcement.
Marching by way of Mawerenye, he crossed the Mbennkuru at Nakiu below
Nahungu, thence to Nahungu and Nyangao, reaching about eight miles to the rear of
Wahle's positions at Nyengedi on 15 August. By the morning of 18 August, the German
positionswere: Abt Kraut on the right with three companies;Abt Rotheon the left with two
companies; and Abt von Chappuis with two more in reserve. Von Lettow with two
companiesremainedbehind Wahle's troops.
O'Grady decidedto use the northern option and issuedorders for the move on 13
August. However, continuousheavyrainfall forced the postponementof the attackuntil the
evening of 17 August. His plan was straightforward,a feint towards the marshesto the
south coupledwith an artillery bombardmentof the main position to cover the move of the

('78Lettow, Reminiscences,p. 204.

245

679
detached
O'Grady
his
Retaining
body
a
two-column
to
the
organisation,
north.
main
demonstration
the
line
hold
battalion
the
marshes.
against
to
the
make
and
single
As previously, Taylor led the advancebefore dark on 17 August. After halting for
bearing.
day
dawn
before
the
along
a
compass
the
next
the night,
advancewas resumed
Emerging onto a newly made road that ran east-west,the lead battalion ran into minor
Despite
the
the
that
using
road
as
axis,
a
new
cleared.
quickly
opposition
was
deployed
impossible
bush
became
to
thick
movement
and this
make
as
so
surrounding
lead
18
August,
By
the
the
on
unit reacheda
afternoon
slowed
advance considerably.
brisk
Kraut.
fire-fight
Abt
A
by
huts
held
securedthe
of
elements
small collection of
battalion
in
halt
dig
induced
bush
from
the
but
increased
firing
to
the
until
and
position,
German
be
This
to
precaution
as
reinforcements
a
wise
proved
reinforcements arrived.
having
Wahle,
the
through
already committed one
to
afternoon.
continued
arrive
hours
but
it
1500
last
in
his
in
the
at
was unable to make
reserve
company
morning, sent
his
forward
fire.
Taylor
heavy
headway
troops
to
the
pushed
as quickly
significant
owing
facing
his
battalions
both
by
1600
an enemy outflanking movement.
were
as possible and
Owing to the ground, he pulled back about 600 yards,just as von Lettow arrived with two
680
companies. A close-rangebattle raged through the night and neither side gave way.
Finally, von Lettow had had enoughand orderedWahle to hold his existing positionsandto
681
refonn an adequatereserve.
While this confusedfighting was underway,GeneralO'Grady was following Taylor
dangerously
his
it
difficult
to
active. Indeed,
move and patrols were
with
own column; was
O'Grady narrowly escapedcaptureby a Germanpatrol Whilst moving betweenunits. In
this situation,he decidedto wait until morning before making any further moves. Early on
19 August, he pushed forward, driving back a number of German parties before linking
up with Taylor. They found that the enemy had withdrawn less than half a mile to the
fire
by
the
west and was strongly entrenched,supported
of a Konigsberg gun. It seemed
clear that substantial reinforcements had arrived; a view that was confirmed by GHQ
679CAB 44/10, pp. 22-23.
60 Boell, Die Operationen,pp. 346-348; CAB 44/10, pp. 23-24. The 1'/2"d KAR under Lt Col Giffard was a much steadierand
reliable unit than the battalion that had broken and run in May.

246

682

intelligence reports that von Lettow had brought half of his main body to the south.
The situation had now changedmaterially, with Linforce stalled some 20 miles inland and
in
had
Germans
The
from
losing
numbers
superiority
troops
a
marked
sickness.
continually
back
O'Grady
For
this
ftirther
the
pulled
reason,
question.
advanceout of
and put any
Taylor's Column from its exposedforward position and establisheddefenceswith a strong
defensive
Linforce
to
the
back
River.
Lukuledi
over
until support
the
went
now
on
reserve
683
finther
be
from
Hanforce,still
north.
could obtained
NORTHEY'S OPERATIONS AUGUST TO SEPTEMBER 1917
Northey continued to be very concerned about the dispersion of his force,
it
deal
left
him
Mpepo,
to
detachment
with
the
as
nothing
operating against
particularly
Tunduru on his right flank. He askedvan Deventer to sendthe promised KAR battalion
to Mpepo so that he could in turn concentratehis Rhodesian units further south on the
Liwale-Tunduru axis. Despite having a force of nearly 4,000 effective rifles, Northey
five
forced
front
300
columns, which meant that only a small
to
with
miles
of
was
cover a
684
he
in
However,
bear
brought
be
to
also noted that the
one
place.
proportion could ever
had
force
Deventer
already ground to a standstill, with a start strength of
van
main
under
5,000 now being reduced to 2,000. Unhappy with events, he arranged a meeting with
685
future
discuss
GeneralSheppard,the BGGS, at Iringa, to
the
conduct of operations.
Shorthose's battalion continued to make good progress having pushed von
Stuemer back through 250 miles of Portugueseterritory. He had out-manoeuvredhis
opponent, crossing the rapidly flowing and crocodile-infested Rovuma without boats by
improvising rafts and ropes out of tree bark and grass. He had to live off the country for
two weeks before switching his base from Fort Johnstonto Wiedhafen, some 300 miles
apart. After a hiatus in communicationsowing to Shorthose'sisolation and the weather,

681Lettow, Reminiscences,p. 205.

612CAB 44110,pp. 25-26.
"' Boell, Die Operationen, p. 348; CAB 44/10, pp. 27-28.
6" WO 95/5329, War Diary Norforce, Entries 5 and 8 August 1917.
"s WO 95/5329,War Diary Norforce, Entries 9 and 10 August 1917. The meeting was set up for 16 August and facilitated by
Northey's completion of a motor road from Songeato Iringa, nearly 300 miles long.

247

Tunduru
three
him
Northey
4
August
towards
to
while
retaining
continue
ordered
on
686
other battalions in reservein Nyasaland.
In the meantime,Northey met Sheppardat Iringa and was able to clear up a large
in
him
Nyasaland
to
differences.
The
the
crisis
enabled
passingof
number of queriesand
687
further
building
he
As
Colonel
Hawthorn
to
north.
up
was
commandoperations
return
his strengthfor an attack on A bt Lincke still heavily-fortified at Mpondas, he used Murray
investment
The
'communication.
Aumann
lines
his
Rhodesians
the
to
take
of
of
over
and
by
heavy
firing
dug-in
Fair's
Mpepo
taking place.
troops
close
and
at
continued with
The Germans struck back on 22 August, launching counter-attacks both at Mpepo
impression
British
failed
Mpondas
the
to
that
on
positions.
any
make
and

It was

becoming apparent that the stubborn defences of both places were being used to cover the
withdrawal of the forces north of Mahenge to the east. At the same time, van Deventer
issued fresh instructions to Northey with the aim of increasing the pressure from the west.
He was now ordered to attack both Mpepo and Mpondas until they had fallen while also
denying the enemy the Tunduru district and any attempted re-entry of Portuguese East
Africa. 688

Eventsbeganto gatherspeedin late Augustas Shorthosemadea daringmoveon

Tunduru,takingit on 23 Augustwhile Fair continuedto tightenthe nooseon Aumannat
Mpepo. That placewas evacuatedduring the night of 27/28 August,but a hot pursuit
followed in which the Germanssufferedheavily, with nearly a third of their force as
689
casualties. By the end of the month,Norforcehad inflicted significantlosseson the
Westtruppen,
with 22 Germansand 241 Askaris being confirmedcasualtiesand three
690
gunstaken. Augusthadbroughtsuccessat relativelylow cost.

'" WO95/5329,WarDiaryNorforce,Entry 14August1917.

6S7
WO95/5329,WarDiaryNorforce,Entry 15and16August1917.
6nWO95/5329,WarDiaryNorforce,Entries21 and22August1917.
"9 WO95/5329,WarDiaryNorforce,Entries28 and30August1917.TheGermans
the
suffered26 killed and55takenprisoners;
Britishlosteightdeadand15wounded.
690
WO95/5329,WarDiaryNorforce,Entry31 August1917.

248

B
I
cts--s
4
1:

COPYRIGHT

Ll

1-1 11 Ill II1

WITHOUT PERMISSION

TO BE it EP ROOLICEO PHOTOGRAPHICALLY

NOT

11 11

r4
AN

ill,,

11

91
A,

mp
L
I
Z

ul

(.j

\\:3
yl

.4,

1'1

bd

I,

/I.

In

'1

\"4
SI,

It

f
01,31,

cv

i%,

-"

40a'

'K .

Ift

15.
V-gRM I

Ld
4

Z'

V) to

I--

P,

-%I

Ia

1z'

:3

'4
01

-14

"\

05.

IA//"/.

.4%-,

1-b
I

CL

14

>
At I
Ir

9-84-

As these operations were in full swing, communications between the respective

September,
By
liaison
limited
telegrams
to
officers.
early
and
commanders-in-chiefwere
during the pauseto regroup before renewing the attack, Huyghd and van Deventer met
for a secondtime at the British headquartersin Dar-es-Salaarnfrom 6 to 9 September.
Van Deventer explained that the coming rainy season in December would make
impossible
Mahenge
to
Kilossa
between
virtually
and
proposed
and
communications
keep
bulk
Belgian
latter
to
the
the
to
the
the
and
of
the
minimum
reduce
garrison of
forces as a reserve on the railway. This information was unwelcome to Huyghd as his
forces were now at maximum strength and the carriers recruited in the Belgian Congo
idea
front.
While
British
forward
the
in
to
the
was not taken
the
were
processof moving
Mahenge
Deventer
the
area cleared and an overland
the
time,
wanted
up at
as van
fruit
in
bear
it
the medium term.
Liwale
towards
would
advance
undertaken,
A modified campaigndirective came out of the meeting, in which the advancesof
the Kilwa force on Nahungu, the Lindi force towards Massassi, and Norforce on
Mponda/Liwale, would be co-ordinated with the Belgians moving on Mahenge. Once
that place was taken and cleared of the enemy, a blocking position was to be taken up
from Mponda to Luwegu River in order to prevent any breakout to the west while making
preparations to send a substantial force to occupy the rich food-growing area around
Liwale. Should the combined actions prove insufficient to wind up the fighting before
December,then Huyghd was to be ready for a further campaign operating either from
Ifakara on the Central Railway or possibly from Liwale or Kilwa.

While no dates or

bring
Deventer
Belgians,
them
to
the Germans
to
the
timings
urged
van
exact
were given
691
decisive
battle
to
and to act as speedily as possible.
To that end, Tytler's Column would only remain in the area of the River Ruipa
until the Brigade Sud could move forward from Kilossa and start preparations for the
crossing of the River Ruaha. The Hubert Column, which was presently en route from

"' CampagnesColonialesBelges,III, pp. 364-366,AnnexeNo. 36,7 September1916,Letter van Deventerto Huygh6.

249

Iringa towards Makuas-Mpanga to reinforce Tytler would remain in the area and rejoin
692
body
Belgian
the
main
on MiVa,

Meanwhile in the north, the problem of Abt Naumann remained unsolved

although he had been gradually losing strength. It had reached the Kilimanjaro area in
late August and threatenedto enter British East Africa. Finally, faced by reinforcements
had
Deventer
Van
hastily
by
it
turned
again.
reinforced the Belgians
south
moved
rail,
began
infantry
battalion
to outpace the raiders. Wishing to
who
with a
plus mounted
he
12
August
further
Belgian
the
on
south,
made the elimination
speedup
concentration
"'
Germans
British
the
of
responsibility.
a solely
Finally, after much doggedpursuit and effort, the remnantsof Abt Naumann were
trapped at Luita, north of the Central Railway. It had been run down by the superior
mobility of the mounted infantry and was seriously short of supplies. Surroundedin a
"'
force
had
September.
The
2
last
Germans
the
on
small
surrendered
waterlessposition,
its
but
disruption
to
size,
created
now the British could
all out of all proportion
body.
destroying
the
concentrateon
enemymain
By the middle of Septemberthe time was ripe again for the advance; both the
Kilwa and Lindi forces had been substantially reinforced and motor transport was now
available to the entire force. Van Deventer wished to push the Kilwa force south with a
concurrent move south-westerly by the Lindi force; his aim was to catch the German
main body whatever course of action it took.

The western troops, both British and

Belgian, were to carry on aggressively in order to contain the troops in the Mahenge
695
area.

HANFORCE

PUSHES SOUTH TO THE LUKULEDI

692Campagnes Colontales Belges III,

141.
P.
,
693

CampagnesColoniales Belges, III, pp. 117-118;CAB 44/8, pp. 73-78..

694WO 33/953, TelegramsD 2, No. 2116,3 September1917,TelegramG 095,
van Deventerto CIGS, p. 168; CAB 44/8, p. 120.
69'Van Deventer,Despatch1, pp. 183-185.

250

in
Schutztruppe
hard
the
While the advancing columns were pushing
against
GermanEast Africa, British intelligence had gatheredmuch useful information about the
border.
Captured
from
Portuguese
the
the
prisoners
of
side
strength and state of morale
field,
in
Europeans
1800
the
July
of whom
remained
that
the
some
end of
at
revealed
lines
600
the
300
on
of communication
were employed
some
were sick and another
696
information
This
front
line
leaving less than 1,000 in the
remained
good.
units; morale
intelligence
by
that
later
showing
resistancewould continue
a
week
was supplemented
Allied
by
the
being
the
German
enormous
reports
of
sustained
morale was
and that
shipping losses:
"... Prisoners estimate total European force in field including L of C and sick
1200. GW in firing line and out hospital thinks they will fight on owing
Enemy
Von
LETTOW.
influence
not recruiting askaris or
personality
enormous
697
,,
to fight
Consider

rugaruga.

askarisremaininggoodandprepared

on...

Reinforced with the Nigerian Brigade and Colonel Dyke's reservebrigade, Force
HQ was well forward, co-locatedat Mssindy with the bulk of its units. It facedtwo enemy
facing
No
2
Column
Mihambia
Ndessa
larger
the
at
the
and
weaker
at
groupings,with
defenders
The
Column.
the
two
to
turn
the
No
I
to
columns
out
use
plan was
opposite
from their strong positions, while the Nigerians were used to make a deep outflanking
lack
The
Mbemkuru.
the
River
to
the
of
their
problem
was
to
greatest
retreat
move cut off
best
develop
find,
the
to
the
as
suitable
sources
ever
positions
secureand
need
water and
were usually defended.
The advanceof Hanforcebeganearly on 19 September,with No I Column moving
towardsMihambia and a small flanking force making for Nitshi, some 12 miles south-west.
A local outflanking action, despite the presence of unexpected depth defences, was
successfuland the all-important waterholeswere capturedby early afternoon. A further
it
Nitshi
day
that
the
unlikely
appeared
advance next
achievedonly slight success,and, as
and its water supply could be reached,Hanforce HQ orderedOff to return to Mihambia on
6% Wo

158/478,7 September1917,TelegramNo 127/5, Capt Cohen to Force Intelligence, 5 September.

251

in
difficulties
moving rations
Development
the
21 September.
water supplies and
of
forward delayedprogressuntil late on the next day. Opposition was slight and two days of
hard marching in hot sun with little food followed. By 24 September,the whole of No I
698
in
Chini.
Column was Bweho
The Nigerian Brigade,lessone battalion,had beencollectedand reachedKilwa by 8
Septemberand beganthe long move forward to Mssindy, some 80 miles distant. Leaving
20
developing
intermediate
task
it
some
sources
18
September,
water
the
of
was given
on
20
September,
Mihambia
the
hearing
On
Mssindy.
the
at
on
action
miles south-westof
leadingbattalionmoved southtrying to find the track that led to the Mbemkuru River before
699
the oppositionarrived. On 22 September,the small village of Bweho Chini was reached,
lay
directly
As
the
hastily
this
fight,
most probable
on
occupied.
was
sharp
and after a short,
its
defences.
A
began
brigade
Nahungu,
strong
preparing
the
and
closedup
escaperoute to
it
left
had
Bweho
but
hardly
the
to
north-east,
two
out
companieswas sent
of
reconnaissance
At
the
time,
heavily
it
a strong attack was
off.
same
cut
than
nearly
and
engaged
chini
was
launchedagainstthe easternside of the hastily dug perimeter. The advancedforce managed
heavy
losses,
German
having
the
boma,
back
fight
its
attack
to
the
while
suffered
to
way
intensified. Chargeswere led againstthe northern and easternfaces of the defencesand
heavily
brigade
hours,
1700
By
the
the
and
was
encircled
other sides.
patrolswere probing
just
before
fierce
Fighting
and,
sundown, a very strong attack,
was
engagedon all sides.
defenders
into
defences.
The
fire,
the
heavy
by
crashed
machine gun and rifle
supported
just held and then a companycounter-attackwith fixed bayonetsdrove back the Germans.
During the night a lull developed,and the time was usedto collect the wounded,improve
the trenchesand to coral the frightened carriers. Ammunition was beginning to run short
andtherewas no sign of the other columns.
The battle was resumedat 2130 hours by moonlight and continuedthroughout the
impetus
had
been
but
faing
followed
the
A
main
and
continued,
evening. numberof attacks
lost so that by midnight the action had beenreducedto sniping while the attackerscollected
'97WO 158/478,14 September1917,TelegramNo 144-1, Cohen to Force Intelligence, 12 September.
' Orr, "Rumbo to the Rovuma7',pp. 116-117.
6" CAB 45/19, Nigeria RegimentRecord,pp. 33-35.

252

their deadand wounded. At dawn, it becameclear that the battle was over and the Germans
had broken up into small parties, moving toward Nahungu. Considerablestores and
ammunition were left behind while patrols later roundedup a number of stragglers. Pursuit
was impossible owing to the lack of food and water, while the remaining porters were
insufficient to carry either the wounded or first line supplies. Finally, the dead had to be
collectedand buried; a taskthat lastedall day.
On the morning of 24 September,a patrol linked up with the batteredcolumn and
cable was re-laid. The woundedwere evacuatedup the road to Mihambia and the troops
marchedoff to Bweho Juu and rejoinedthe rest of the brigade. It had beena severetest for
both sides,but the Nigerianshad retainedthe field of battle.700
In the meantime, No 2 Column and Dyke's Reserve Column were preparing to
move on Ndessa. In order to avoid a frontal assault,Dyke and elementsof No 2 Column
would attack from the north, while the bulk of No 2 Column would move around the
With
four
1,400
to
to
the
only
and
south-west.
rifles
and
guns
position cut off escape
west
,
to cover a frontageof over four miles, it was not an easytask.
In order to achievesuccessagainstthe inevitable counter-attacks,a modification of
lines of communicationdefencewas used. As was usual,the Germanshad a seriesof three
to four defensivelines sited in depth and centredon the main road. To counterthe strength
of the defences,a new road, cut to a point some3 miles west of Ndessa,was clearedby the
flanking force. Protectedby a small garrison,the remainderof the column marchedup this
track and then deployedfacing the direction of the enemy. After a pre-arrangedadvance,
the left hand battalion peeled off half left, moving on a compassbearing for a specified
distance,then halted, and re-aligned itself with the original direction of advance. It then
beganconstructinga boma as silently as possible. In the meantime,the remainderof the
column kept advancingto a pre-determineddistance,and then repeatedthe actions of the
first battalion. Finally, the fourth andreservebattalionpreparedits position to the rear of the
right forward unit. After four and a half hours of marchingand construction,the boma was

253

from
front
hours.
Assailed
defences
1030
the
and
the
the
at
artillery openedup on
ready and
taken from an unexpectedleft flank, the defenceswere not held long and the Germans
hand
launched
A
the
break
to
the
against
right
to
strong attack was
west.
attempted
out
battalion of the boma but was repulsed,followed by anotherattempt, this time on the left
hand unit at 1900hours. It too failed, and a final drive on the centre of the British line was
Ndessa
This
the
the
end
of
position and a numberof
marked
unsuccessfulabout midnight.
701
hastily
Germans
to the south-east.
withdrew
storeswere abandonedasthe
The loss of both the Ndessaand Mihambia positions increasedthe pressureon the
Germans. Retreatingrapidly to the south-eastand then turning south-westalong the track
betweenMihambia and Nahungu, the combined detachmentsuffered heavily when it ran
into the block at Bweho Chini and the Nigerian Brigade. As a consequenceof the hasty
food
Mawerenje
together
No
2
Column
the
at
stocks,
with
captured empty position
retreat,
10,000roundsof ammunitionand a bomb factory.
The next objective was Nahungu. 'Me defencesin the north were collapsing and
linked
Chini
Bweho
After
British.
the
the
columns
main
up
around
on
momentumwas with
24 September,the next stepswere taken. No 2 Column went south towards Nakiu on the
Mbemkuru while No 1 Column and the Nigerian Brigade pressedsouth-westto Nahungu.
After several days of marching and pushing back rearguards, Nahungu fell without
difficulty.

Once Nahungu was secured,attention turned to pushing south of the Mbemkuru.

The enemyhad taken up a strongposition at Ngarnbururu,some 15 miles to the south-west
of Nahungu. Von Lettow had marched up with his reserve, reinforcing the somewhat
demoralisedforce and had developeda very strong position along the road. The British,
realising that such defenceswould be difficult to take as well as wanting to link up with
Linforce as quickly as possible,decidedto launch a deepright flanking movementaimed at
seizingRupondaandthen Lukuledi. No 2 Column was given the task of locating and fixing
700CAB 45/19, Nigeria RegimentRecord, pp. 36-39. Casualtieswere eight British and 127 Africans killed or wounded. A total of 16
Germansand 87 Askaris were found deadon the field of battle.
70'Ridgway, "With No 2 Column", II, pp. 248-249.

254

Again,
the tactic of cutting a
in
Ngambuniru
the
to
the opposition at
others.
order cover
intensive
together
1,000
to
the
with
used,
was
existing one
yards parallel
road about a
recomaissances,including aeroplanes.
The tactic worked, for as soon as von Lettow beganto withdraw his detachmentto
deal with the pressing matters in the south, No 2 Column was able to drive out the
for
30
It
in
October
pressed
another
miles, going via
and push south. was
rearguards early
Ruangwaand ending at IRoe, a small village on the edge of the waterlessplain that led
down to the Mkemburu Valley. Ultimately, the advancewas too successfid,as the forward
October
No
I
No
2
Columns
10
lines
literally
were
their
and
on
supply;
of
outran
columns
food.
build
in
to
Ruangwa
Ruponda
stocks
of
order
up
sufficient
respectively
and
recalled to
Water, too was in very short supply and would be a major constraint on the final link-up
702
Linforce.
with

THE ADVANCE UP THE LUKULEDI

In early September,Generalvan Deventerissuedgeneralinstructionsto his principal

belief
he
In
that the powers of the enemy's resistance
the
them,
expressed
commanders.
bring
two
the end of the
that
serious
or
reverses
would
one
were perhapsover-estimatedand
factor
be
limiting
He
the
that
transport
and that
would
critical
recognised
campaignnearer.
he
instructed
front,
On
improve
be
tactical
the
to
communications.
every effort would made
that frontal attacks should be avoided as being slow, costly and indecisive. He wanted
formationsto seizepositionsin the enemy'sflanks and rear andto defendthem. This would
force the Germansto attackand return the advantageto the British. However, as a precursor
to any operationshe required extensive reconnaissanceto be carried out by long range
he
Finally,
Scouts.
Intelligence
the
expected the
patrols, aircraft observation and
forthcon-drigoperationsto be continuous and without pause. By making great sacrifices
703
he
hoped
draw
to
the campaignto an endwithin two or threemonths.
now,

702Ridgway, "With No 2 Column", II, pp. 250-253; Off, "Rumbo to the Rovuma7,p. 120; WO 95/5294, War Diary GHQ, 14 October
1917,Appendix B, Telegram 7 WO, van Deventerto CIGS, 14 October; and 15 October 1917,Appendix B, Telegram 6 WO, van
Deventerto CIGS, 15 October.

255

With the advanceof Hanforce southward,attention again turned to Lindi. It had

been on the defensive since 21 August and had received some rest and reinforcement.
Equally important was the belief that the Germanshad reducedtheir forces in the area in
held
British
Intelligence
that
the
deal
now
to
the
threat
estimated
north.
with the
order to
local numerical superiority. On 4 September,GHQ instructed Linforce that the next
involve
This
from
Narunyu.
be
the
to
turn
would
a strong column
south.
advanceshould
Mtua
Lukuledi,
the
River
bank
seizing
the
and
a
crossing
near
with
the
of
moving up
right
increasing
However,
lines
the
German
given
of communication.
of a position astride the
Major
General
Beves
decided
to
take
it
force,
to
importance
send
the
was
of
size and
in
He
Lindi
9
September.
by
then
he
left
Dar-es-Salaam
arrived
on
and
ship
command,and
force
based
The
four
two
into
force
columns,
on
striking
was
main
the
groups.
reorganised
O'Grady
General
Brigadier
brigade,
the
and
of
command
the
under
eachroughly
size of a
Colonel Taylor. In keeping with the organisationof Hanforce, the columns were renamed
No 3 Col andNo 4 Col respectively. A ForceReservewas createdand the remainingtroops
704
dutieS.
were placedon lines of communication

Bevessubmittedhis plansto GHQ for approval. The proposed

On 18 September,
have
battalions
1,300
Column
follows:
No
3
three
totalling
rifles,
would
organisationwas as
17 machine guns, 18 Lewis guns, two mountain guns and four Stokes mortars; No 4
Column would have four battalions and a machine gun company with 1,250 rifles, 29
four
Stokes
Force
Reserve
four
Lewis
29
guns,
and
mortars;
mountain
guns,
machineguns,
hadtwo battalionseachof 800 rifles, 12 machineguns, 16 Lewis gunsand six guns.705Asa
preliminary to the planned move, a number of reconnaissanceand fighting patrols were
launched around the Narunyu position supplementedby aerial observation and artillery
bombardments.On 8 Septembera patrol had found a good crossingon the River Lukuledi
for
became
first
Mtua
the
this
objective the advance.
near
and
Operation orders were issued on 20 September. No 3 Column was instructed to
No
4
flanking
in
Column was
Lukuledi
the
to
the
the
while
make
valley
move
crossingsite
703WO 95/5294, War Diary GHQ, 2 September1917,Appendix A3, TelegramG 059, BGGS to Hanforcc and Lincol, 2 September.
7" CAB 44/10, pp. 29-30. Two battalionswere sentto Lindi on &'9 and 19/20September.
70-'CAB 44/10, p. 3 1.

256

GEWAN EAST ARICA

Sketch M04.
Eastcro

Ilimire

r,

MY

V-1

M-ot

3,r

vn-L-W..

HANFoACC

AHO

61TEMJEA

Q'6A"
OCTO, (EA

AtipolLt

71
.

0-1

Tm%ti.

i4-L
sm-w

-T-.

-v

&"m

ll-A

1-te.
.

4...
0

., Vcc-r

kmi

leading
The
in
its
in
elementsof
Narunyu,
to
support.
to remain opposite
readiness advance
No 3 Column set off before dawn on the morning of 23 September. Lying up to avoid
The
the
bridge
day,
advance
during
river.
over
temporary
the
was constructed
a
observation
hours
in
halting
Finally
the
early
was resumedafter nightfall, althoughprogresswas slow.
day.
Again
lay
for
his
had
the
O'Grady
troops
24
September,
up
again
of the morning of
bush
hack
had
No
3
Column
trail
the
to
dark
and
of
virgin
a
out
that evening,
starting after
Withoutpausing,
25
September.
dawn
just
before
designated
on
crossingpoint
reachedthe
the two leadingcompanieswadedacrossthe river and pushedonto the main road.
The defenders,At Rothe, initially assessedthis move as being merely a patrol.
Quickly proven wrong by heavy firing, Rothe sent his reservesforward to prevent further
706
in
dig
to
then
and
the
started
to
who
These
attackers,
stop
sufficient
were
penetration.
fade
began
dislodging
hopes
to
hours,
the
fire.
1300
By
in
heavy
crossing
of
artillery
called
Narunju
decided
to
Wahle
to
the
and
despite
position
abandon
the
arrival of reserves,
and
707
In the evening, the two guns followed by At
fall back on the Nambalika's position.
Kr6ger were pulled back while Rothe actedasrearguard.
While No 3 Column was underway,No 4 Column had been cutting paths through
the densesisalplantationsin front of Narunyu- Artillery fire on the main defensiveposition
in
No
3
due
September;
to
25
both
the
24
once
go
was
attack
and
coveredthis slow work on
Column had closedthe Germans' escaperoute. This was achievedon 26 September,but
3
during
No
the
had
defenders
to
the
late;
night,
evading
west
the
slipped away
was too
Column on the main road."'
A patrol soon located the new positions along the line of the Nengedi Streamand
Beves ordered his troops to close in on the enemy. Advancing up the thickly vegetated
Chirumaka Hill and along the main road, No 3 Column ran into a number of strong
positions and counter-attacks. By nightfall, the whole Column was committed as was the

706CAB 44/10, pp. 32-33.

707MS Boell, N14/32,28. Kapitel, p. 2582.
703WO 95/5294, War Diary GHQ, 28 September1917,Appendix B, Telegram 0 378, van Deventerto CIGS, 28 September.

257

lead battalion of No 4. Casualties were relatively heavy, but the ridge was cleared and a
709
ff
final counter-attack was successfully beaten 0 .

Beves
General
the
further
for
ordered
day
After a
planning,
reorganisationand
Column
the
No
3
September.
30
existing road around
along
moved
to
advance resumeon
Hill
Chirumaka
as well as along the newly cut track.
the northern end and western side of
Troops on the latter route pushedback German outposts,launching a battalion attack at
had
battalion
Nigerian
On
defenders.
the
back
hours,
driving
1400
the
existing road, a
faced
but
difficulty
Chirumaka
no
with considerable
worked aroundthe northern slopesof
in,
dug
junction,
track
and rapidly
effective opposition. They reachedtheir objective, a
hours,
heavily
1100
It
forward.
further
and
attacked
at
was
sent
was
company
while one
followed
body.
The
the
back
forced
to
the
it
hours
later
enemy
main
was
several
defensive
it
the
position, whereupona
main
withdrawing company closely until reached
Germans
to
defences
the
The
launched.
unable
were
sound
and
were
vigorous attack was
battalion
dark.
O'Grady
to
firing
heavy
a
sent
headway
until
continued
although
make any
dark
to
the
it
but
to
Nigerians,
owing
hard-pressed
arrive
after
only managed
reinforce the
difficult terrain. By this time, however, the enemyhad broken off contact completely and
had slippedoff in the darkness.
Overall, the day's immediateobjective,the occupationof ChirumakaHill, had been
for
Valley
Nengedi
the
it
many miles.
of
achieved and offered excellent observation
However, as so often happenedin attacksthrough thick bush, casualtieshad not beenlight
further
delay
decided
Beves
to
General
had
advance
cautiously
the
got away.
and
enemy
710
until Hanforcewas in closerproximity.
THE ADVANCE TO NYANGAO

If the British were to catch and destroy the Germanmain body in battle, the LindiMassassiroad that ran east-west, largely following the course of the River Lukeledi,
7'9CAB 44/10,p. 38. The British casualtieswere 13 killed, 107wounded,and six missing. 'Me Germanslost five killed, 32 woundedand
one missing.
711CAB 44/10, pp. 36-37. The British casualtieswere 30 killed, 75 wounded and five missing. The Germanslost four dead and 23
wounded.

258

important
Massassi
favourable
the
supply centre and
was
an
opportunity.
offered
most
711
in
The
Newala
linked
Songea
the
to
the
that
and
south.
west
communication node
difficulty of the terrain and lack of alternative routes made the road of considerable
in
had
both
Hanforce
importance
the north and
to
made
good
progress
sides.
operational
the time for a co-ordinatedmove with Linforce now seemedto have arrived.
General Beves was in local command of the operations and believed that the

betweentheNengediRiverandMahiwa,of whichsix
Germanshadeightto ninecompanies
frontal
from
Column
have
No
3
His
Mtama.
the east,
to
a
make
push
werearound
planwas
bank
Lukuledi
River, crossingat Mputo
No
4
Column
the
the
of
north
while
advancedup
Nigerians
in
from
In
Mtama
the
the north were
the
then
meantime,
south.
and
engaging
to cut the Lindi-Massassiroad at Mahiwa and then move north-eastwardsinto the
Germanrear. They were to preventthe enemyfrom withdrawing,while Nos I and 2
Columnsmovedsouthon Massassi.Concurrently,cavalrypatrolswould rangefreely to
712The plan wasa
find anddestroyasmanyfood suppliesandsupplydepotsaspossible.
bold one, but required a speedyadvance,close co-operationbetweenwidely-spread
columns,andaggressiveness.
In the first week of October,a largenumberof patrols had beensentout to locatethe
enemy positions. By 7 October, O'Grady had gained a foothold on the far bank of the
NengediRiver, launchinga reinforcedbattalion onto the high ground overlooking the river.
Supportedby artillery fire, the infantry reachedtheir objective, but were hit by a strong
counter-attack. Sensingthe situation's importance,anotherunit was pushedforward. This
was hit by a strongGermancounter-attackand the high groundwas lost. Fighting stabilised
with the arrival of further British reinforcementsand the bridgeheadon the westernside of
the Nengedi River was retained. This was followed by a resumption of the advanceon
Mtarna,the next day. )While O'Grady's troops continued to progress,the resistancewas

711CAB 45n2, Map 9508, "The Strategic Situation 20* December 1917"; WO 95/5294, War Diary GHQ, 23 September1917,
Appendix A25, "Instructions to O.C. 25' Cavalry", 16 September.
712Downes, With the Nigerians, p. 197.

259

increasingly stiff and it was becoming clear that the Germans intended to fight hard and
713
delay.
cause

In light of the strong defences,Beves now decided not to press a direct attack on
Mtama and to encircle it instead. On 12 October, he sent No 4 Column, now under the
command of Colonel Tytler, marching west and crossing the Lukuledi below the
lasting
days,
After
the
Nengedi
the
arduous
march
several
an
stream.
confluence with
but
in
15
October,
the
Mtama
the
afternoon
of
entered
was
river was crossed and
defendershad withdrawn. At the sametime, No 3 Column had been moving west along
the road, pushing back the German rearguardsuntil they linked up with No 4 Column
his
kept
Mtama.
Van
to
Deventer
opponents
pressing
give
no
rest
and
anxious
was
near
714
Bevesto maintain unrelenting pressure.
On 13 October, Beves made a major alteration to the plan, issuing his orders by

Mahiwa
Nigerian
Brigade
The
to
north-west
of
while sendingone
march
was
wireless.
battalionto cut the Lindi-Massassiroad south-westof that village with the remaining
unitsto attackMahiwaon 15 October.At the sametime, No 3 andNo 4 Columnswould
continuetheir convergence
on Mtama. The die was set for the major, possiblydecisive,
been
had
Deventer
that
seekingsinceresumingoffensiveoperations.
engagement van
THE BATTLE OF MAHIWA

In his hasteto cut-off the Germans,GeneralBeves had greatly underestimatedthe

time required for the Nigerian Brigade to reach its objectives. They had only reached
Namupa Mission by late on 14 October and lost direction in the thick bush and darkness.
Exhausted,the column dug-in for the night unawareof the exact location of the eneMy.715
Colonel Mann, now in temporary command,was concernedabout the vulnerability of his
baggagetrain, and decidedto leave it at NamupaMission. He also had seriousmisgivings
713CAB 44/10, pp. 3940.
714Van Deventer,DespatchI, p. 187;CAB 44/10, pp. 41-42. Cites Telegram552, BGGS to Linforce 15 October 1917;"Chief
wishesyou
to pushon now asfast aspossibletowardsMassassi.Give enemyno time to makedefensivepositionsandmakeeveryendeavourto capture
his guns. Enemy is now much shaken,and a determinedadvanceon your line combinedwith operationsoutlined in my 549 will probably
havedecisiveeffect."
715CAB 44/10, pp, 42-43.

260

Mahiwa
than
intelligence
his
his
to
at
enemy
stronger
a
much
pointed
own
about
role as
716
by
General
Beves.
that assumed
Early on 15 October, a single battalion left the brigade camp to carry out its task
fire
Coming
Nyangao.
blocking
Lindi-Massassi
shortly
the
under
sporadic
road at
of
just
back
leaving,
battalion
two miles short of
the
until
the
rear
guards
pushed
after
Nyangao. After a serious fight, supported by artillery, the Nigerians drove back their
dug
in
There,
to
they
meet
their
to
and
prepared
objective.
opponents and pushed on
Wahle's force. In the meantime, the brigade's main body, two battalions strong, left at
717
The advancedguard of two companies
0530 hours marching south towards Mahiwa.
from
the outset, although the enemy rearguards
heavier
encountered
resistancealmost
River.
Mahiwa
forced
back
the
toward
were
Despite the evidenceof opposition, Beves instructed Mann to press on as quickly
River
had
Mahiwa
leading
the
hours,
I
100
By
the
nearly
reached
company
as possible.
but
German
fire.
This
it
heavy
a
substantial
was clearly no rearguard,
when came under
in
immediately,
began
digging
lead
battalion
At
the
this
and
consolidated
point,
position.
However,
delay
further
this
was not received
as
reinforcements.
company
receiving a
frontal
forward.
instructed
Realising
by
Mann
Beves
to
that
press
a
again
well
was
and
find
he
to
try
off
and
a route on the eastern
sent a company
advancewas not possible,
flank of the enemy. Leaving at 1645 hours, it was only able to advancebefore being hit
by a strong frontal attack which forced it to withdraw. The cost had been heavy with
718
NCOs.
becoming
40
casualties,especially officers and
some per cent of the company

The Nigerianswere now in a difficult situationwith the Germansholding the

The newly-arrived Abt von Ruckteschell
dominating feature of Mremba Hill.
immediately launcheda heavy attack on the main defensivepositions from the south and
south-east. The battalion desperatelyclung to its ground and fended off the attacks until

716Downes, With the Nigerians, p. 197.

717CAB 44/10, p. 44. The main body totalled 1,100rifles, 16 machine and 16 Lewis guns,two Stokesmortars and two 2.95" guns.
7'8Downes, With the Nigerians, pp. 198-201. British casualtiestotalled 63; CAB 45/19, Nigeria RegimentRecord, pp. 474 8; Boell,
Die Operationen,pp. 373-374.

261

flank
lull
dusk,
hours.
1800
Just
the
threatened
eastern
sudden
attack
on
a
around
after
a
to overrun the defenders,but again they held on with both sides suffering heavily.
By the evening of 15 October, it was clear to Mann that a large German force was
moving forward to attack the Nigerian positions. Furthermore, his battalion at Nyangao
and baggagecolumn at Namupa were also under threat. With the heavy expenditureof
small arms ammunition, supplies were running low and concentration seemedessential;
he pulled his troops together onto a hill overlooking the Mahiwa River. He then radioed
Beves to tell him that both water and ammunition were running short and to ask for the
return of his detachedunit. This was agreedand the battalion was instructed to rejoin the
719
body
first
light
October.
main
at
on 16
The situation worsenedovernight as AN G6ring with two companieshad moved
720
line
leading
Nigerians,
Namupa
Mission.
telephone
to
around the
cutting the road and
On the afternoon of 16 October, two guns were brought onto to Mremba Hill and began
bombardingthe trenches. The fire raked the rear of the trenches,forcing the defendersto
abandonthem after suffering heavy casualties. After the Nigerians had been under the
relentlessbombardmentfor some time, the artillery lifted its fire and the infantry surged
towards the defencesfrom the east and south. The dazedNigerians managedto stagger
back to re-occupy the empty trenches moments before the Germans arrived. A heavy
fire-fight ensued with AN von Ruckteschell being halted about 100 yards from the
forward positions. Unable to break in, the attackerswithdrew after an hour.721
As ordered, the blocking battalion returned to Namupa Mission on 16 October,
where it took control of the transport,numbering some 3,000 non-combatantsin all. The
battalionthen marchedto the relief of its comrades. Hearing the firing, the advancedguard
ran straight into At Goring, which was on its flanking move. Heavily outnumberedand
reducedby seriouscasualties,it was driven back towardsNamupa Mission. The rest of the
battalion could do little as it was split between escorting the carriers and providing the
719Downes, With the Nigerians,
pp. 202-203; CAB 45119,Nigeria RegimentRecord, pp. 49-51; CAB 44/10, pp. 4546.
7" MS Boell, N14/36,30. Kapitel, 2802.
p.
721CAB 44/10, pp. 4647; Boell, Die Operationen, 374.
p.

262

At
by
Kohl,
At
GOring
and numberingnearly sevencompanies
was nowi0ined
rearguard.
722
bitter
fighting
The
Nigerians.
launched
the
and
was
they
extended
a strong attack onto
blank.
At
firing
the sametime, the
two
point
the
at
guns
mountain
very close range,with
terrified carrierspanickedand attemptedto flee only to be held by the fixed bayonetsof the
brigade's
to
lose
the
troops
Desperate
the
to
managed
ammunition,
reserve
not
rear guard.
back
Narnupa
Mission.
loads
dropped
towards
to
the
slowly
move
and
collect
majority of
This place was reachedafter dark and order was gradually restored. It had been a bad day
723
lost.
loads
its
with the battalion losing a quarterof strengthand a numberof

Meanwhile,Linforcewasstill someway to the east. No 3 Columnhadleft Mtama

lunchtime.
find
by
Unable
Nyangao
the expected
to
16
October
and reached
early on
battalion, O'Grady heardthe distant soundsof battle that marked a heavy engagement.He
then receiveda messagefrom General Beves informing him that three enemy companies
judgement
he
be
his
it
Nyangao,
to
attacked
up
whether
and would
were reportedwest of
immediatelyor awaited the arrival of No 4 Column. Without hesitating,O'Grady ordered
No 4 Column to close with the enemy along the road, while No 3 Column moved north
againstthe Gennanflank positions.
Thesemovestook place in the late afternoonwith No 4 Column making good initial
in
forward
his
Wahle
However,
troops,
sending
a counter-attackthat
reinforced
progress.
from
Beves
Back
Mtua,
the Nigerians at 1600
the
a
received report
stopped
advance.
at
hoursand despitetheir situation,he orderedthem to attack and drive the enemysouth along
the road. He wanteda co-ordinatedattackwith Linforce, but the Nigerians' critical shortage
724
food
it
impossible.
from
lack
of
made
of ammunitionandphysical weakness
O'Grady's columns were straining to reach the beleagueredNigerians. They had
marchedthrough the night while great teams of carriers had dragged two 5" howitzers
forward.

He had been reinforced by a battalion from Force Reserve which was

722CAB 44/10, pp.48- 49.

72'CAB 44/10, pp. 49-50. The British casualtieswere 217 killed and wounded;CAB 45/19,Nigerian ReginwittRecord,pp. 48-49.
724CAB 44/10. pp. 51-52 W033/953, TelegramsD2, No. 2224,17 October 1917,Telegram 10 WO, van Deventerto CIGS.

263

immediatelysentto relieve NamupaMission. He then followed briefly up the Namuparoad

725
before
battalions,
his
turning south-westand moving crosscountry.
three
with
own
By mid-morning on 17 October,No 4 Column was attackingAbt Wahlealong the
fire.
howitzer
The
Germans
Stokes
by
retaliated
and
mortar
road, supported machine gun,
developed.
heavy
O'Grady
heard
hot
battle
howitzer
fire
from
the
their
a
and
own
with
firing and orderedNo 3 Column to move in supportof its comrades. Extremely densebush
hours.
leading
I
100
The
into
infantry
the
units
the
at
enemy
ran
slowed movement and
were rapidly drawn into the firing line with severalattacksand counter-attacksraging over
just
battalion
his
O'Grady
in
late
hours.
Finally,
the next three
committed
reserve
afternoon,
in time to thwart a major German attack and restore the situation. His forces now fully
726
ftu-ther.
committed,No 3 Column could advanceno
Back on the main road, the fortunesof No 4 Column waxed and waned. On hearing
the firing of O'Grady's troops, Colonel Tytler had renewed the attack on the German
hit
by
leading
their
both
his
hours,
1530
By
on
a
strong
counter-attack
were
units
positions.
front and left flank. They were unable to hold againstthe fresh opponentsand the day's
by
his
had
be
A
to
reservethen stoppedthe enemyattack
counter-attack
gains
abandoned.
dark,
4
Column
its
No
After
to
the
the
returned
area
of
start
and
situation stabilised.
727
positionsof that morning.
Elsewhereon 17 October, von Lettow renewedhis efforts to destroy the Nigerian
Brigade. Heavy rifle and machine gun fire was directed on the trenches, eventually
destroyingthe vital wirelessmast. Now without communications,the only indication of the
defenders
No
firing
the
to
the
the
and
situationwas
south-east. attackwas pressed
soundof
clung on with grim determination. A break-outwas out of the questionand the only option
was for Linforce to continue its drive through the encircling force. However, with Force
HQ still far in the rear at Mtua, Beveswas unableto exert detailedcontrol of the operations

725CAB 44/10, pp. 53-54.

726CAB 44/10, pp. 55-56.
727CAB 44/10,p. 57. During the 17thLettow hadsentforward reinforcementsat intervalsto strengthenWahles forcebut
pulled mostof
themback into reserveafterdarlaiess;Boell, Die Operationen,pp. 375-376.

264

and O'Grady was againgiven local tactical control. He then issuedinstructionsfor an attack
728
dawn
by
No
3
No
4
Columns.
the next morning all units of
at
and
Early on 18 October,O'Grady launchedthe main body of No 3 Column through the
bush while the attachedbattalion was sentto link up with the Nigerians. Concurrently,No
4 Column surgedforward with two battalions leading. Engagedheavily by the defenders,
the assaulttroops drove them out and began to dig in rapidly. The British howitzers and
mountain guns soon came into action, suppressingthe Germansto their front. However,
the advanceof No 3 Column was soon checkedby a well-sited machine gun to its front.
O'Grady made a personal reconnaissanceto his left flank, guiding several units in the
confusion and rallying some shakentroops. Now seriously weakened,the left flank of his
column was rolled up by an enemy counter-attackand pushedback some distance. While
the situation was far from happy, the relievers brought the good news that the Nigerian
Brigadehad beensuccessfullyextracted.729
To the south, No 4 Column was not having much greater successin breaking
through the defenders. The initial advancehad resulted in German reinforcementsand a
heavy fire-fight developed. The lead units had lost heavily in officers, and at I 100 hours a
strong attack forced both battalions once more back to its starting positions. As they
withdrew, the British howitzerscameinto action and inflicted a number of casualtieson the
attackerswho abandonedany pursuit. The two sidesthen took up hasty defensivepositions
730
banks,
being
having
fought
on opposite
everyone
exhaustedand
eachotherto a standstill.
Unexpectedly,the main opportunity of 18 October was with the Nigerian Brigade.
Its patrolshad discovereda gap in the Germans'encirclementand Mann ordereda breakout
when the battalion from No 3 Column linked up with his patrols. It promptly moved
forward, taking over the trencheswhile the defendersmarchedout, harassedonly by
sniper
fire, andreturnedto Nyangadunimpeded. The relief was successfuland crisis was
over.

729CAB 44/10,p. 59, footnote 1.

729CAB 44/10, pp. 60-63; WO 95/5330, History 1'/2"dKM
25-26.
of
pp.
730CAB 44/10, pp. 64-66; Boell, Die Operationen, 375; WO 95/5330, History
p.
of 1`1204KAIpp. 27-28.

265

be
far
lines,
battle
to
too
from
the
still
By late afternoon,Beveswas some six miles
his
left
No
4
Column
the
O'Grady
in direct touch with events.
retirementof
adjudgedthat
left flank exposedwhile all his troops were much weakenedby the strenuousfighting and
Column
No
4
into
back
him
at
to
with
Beves
conformity
move
ordered
suffered.
casualties
1700 hours and to set up a defensiveline. The move had barely startedwhen a German
by
infantry,
However,
the
mountain guns,
well
supported
attack threatenedthe situation.
held off the attack with little problem and the withdrawal continued without ftirther
interference.
This final counter-attackmarkedthe end of the battle of Nyangao or Mahiwa, as it
battle
had
been
bloodiest
It
the
the
Germans
of
by
British
the
respectively.
and
was called
losses
between
40
fighting
and
cent
British
per
of
the
suffering
units
entire campaign,with
The
50 per cent of their strength while ammunition supplies were nearly exhausted.
Germansdid not get off lightly either asAt Wahlelost 30 per cent of its strength,while A bt
Gbring, Kbhl, and von Ruckteschell all lost about 16 per cent."' The evacuation of
for
the
time,
take
and
some
would
material
casualties,reorganisationand resupply of
immediatefuture the advanceof Linforce had halted. Beves himself arrived at the forward
failure
by
been
had
the
the
his
but
October,
19
of
seriously
weakened
position
positionson
in
day
Dar-es-Salaarn
to
He
the
order to make a personal
same
offensive.
was recalled
"'
report to the commander-in-chiefandwas subsequentlysacked.
Mahiwa had been an opportunity for both sides to achieve a great success,but
heavy,
losses
indecisive
it.
It
with each side
tactically
very
were
and
was
neither attained
losing nearly a third of its fighting strength. Von Lettow had his long-awaitedchanceto
defeatthe British in detail, as they had fought with insufficient co-ordination and were too
far apartfor mutual support. Beveshad ignoredthe dangersof too hastyan advancewithout
knowledge of the enemy positions and blundered into a trap. Then at the end, the
Beves
defenders
food
for
Nigerians
to
the
the
escape
unscathed
when
enabled
withdrawal of
battle,
he
his
Tbroughout
have
been
hit
the
to
them
was too
other columns.
with
might
able
731CAB 44/10, p. 72; Boell, Die Operationen,p. 377. The British suffered 1,455casualtiesas comparedto 611 Germans,

266

far in the rear and had a very imperfect graspof the conditions facing his troops. On the
deal
had
losses
their
had
hand,
Germans
of
a
great
consumed
and
the
major
suffered
other
had
been
initial
to
having
despite
Von
Lettow
unable
advantage,
strong
a
very
ammunition.
destroy either the trappedNigerians or No 3 Column. His desire to inflict a heavy defeat
had drawn him into an attritional sluggingmatchthat the British desiredand that he could ill
his
his
later
his
the
Despite
victory,
strategyof prolonging
scale
of
commentsabout
afford.
the campaignwould not survive anothersuch battle. At the operationallevel, von Lettow
had disruptedvan Deventer'splannedadvanceand sloweddown his concentrationof forces.
Linforce was essentiallyparalysedfor three weeks at a time when its presencemight have
been of vital importance and the battle therefore enabled von Lettow to march away to
733
reinforceMajor Kraut much ftirffier to the west.
THE PUSH TO THE PORTUGUESE BORDER

Whilethe seriousfightingaroundMahiwawasunderway,the restof vanDeventer's

food
his
the
Following
supplies,
off
enemy's
plan of cutting
columnswerepushingahead.
the newly arrivedcavalrywas causinga greatdealof troublefor the Germans.Hanforce
hadreachedRupondaon 11 October,capturinga numberof sick whitesandaskarisin the
hospitalaswell assome2,000loadsof grain.734TheBelgianshadtakenMahengeandwere
landed
Kilwa
battalion
had
Liwale,
in
towards
the
at
while one
now
processof advancing
its
joining
Portuguese,
before
The
from
Liwale
too,
Mssindy
to
comrades.
andcut a road
Rovuma
despatch
towards
help,
three
the
the
to
through
columns
across
of
resolved
Newala.735 On 17 October,No 2 Column had driven At Kohl back to Mkoe while No I

ColumnhadreachedChingwea,en-routefor LukulediMission.
One of the reasonsfor the advanceon Mahiwa had been logistical; Kilwa was too
distantto be effective as a supply centreand it was intendedto switch Hanforceto the Lindi
lines of communicationas soon as the road to Massassiwas cleared. The check at Mahiwa
732CAB 44/10, p. 67; WO 33/953, TelegramsD2, No. 2229,20 October 1917,TelegramOA 559, van Deventer to CIGS. Van
Deventerwas dissatisfiedwith Beves's performanceand removed him from command,although he did acknowledgethat the latter
had beensuffering from malaria which may have affected his judgemenL
733CAB 44/10, pp. 70-7 1.
734WO 95/5294, War Diary GHQ, II October 1917,Appendix B. Telegram 4 WO, van Deventerto CIGS, II October.
733WO 95/5294, War Diary GHQ, 13 October 1917,Appendix B, Telegram 6 WO, van Deventerto CIGS, 13 October.

267

forced van Deventerto reconsiderhis plans. In the short term, No I Column had to return
battle
Mission
Lukuledi
the
its
forward
from
Ruponda
was
shortly
after
to
at
position
north
broken off. An operationalpauseof abouttwo weeksnow ensuedas the British broughtup
736
food and ammunitionpreparatoryfor the next phaseof the advance.
Water difficulties were a major constrainton operationsas sourceswere scarcein
the regionsinland. However,the constructionof roadsand rail links continuedat top speed
737 with
by
26
October.
Mtua
with the railhead reaching some six miles south-westof
it was up to Norforce and
Linforce still slowed down by reorganisationand reconnaissance,
the Belgians to maintain the pressure. This was consistently applied, although the
bulk
forced
in
Mahenge
the
the
the
of
of
phased
withdrawal
area
approachingrainy season
the Belgiansfrom there. By 30 October,Norforce had taken Liwale and the Belgianswere
between
Liwale
Mssindy
Importantly,
Ligombazi.
the
road
and
motor
still attackingnear
738
Liwale.
was nearingcompletionand a Belgian column was approaching
At the sametime, and againstGeneralvan Deventer's wishes, the new Portuguese
ExpeditionaryForce was beginning a preliminary move north of the Rovuma, with the aim
The
because
GHQ
Nangyadi739
Newala,
Matua
taking
move
was
unwelcome
mainly
of
and
had a completelack of confidencein the ability of the Portugueseto succeed.Nonetheless,
the British commander-in-chieftried to co-ordinate the Portuguesemoves with his own
advance,following the pauseimposedby the battle of Mahiwa.
Throughout October, General Northey in the south and west had maintained
from
detachment
Tunduru
Shorthose
a
strong
north-east
sent
steadypressureas ordered.
to the Mohesi River. By mid-October, it had moved forward to Abdallah-Kwa-Nanga,
which was in the centre of a fertile district some 50 miles south of Liwale. From there, it
moved up to Liwale itself, taking the village on 29 October together with some24 whites.
Colonel Hawthorn continued to push along the Luwegu River and by 5 October had
establishedhimself on the north bank east of the enemy's position. He attacked on 16
736Off, Colonel G M, "Von Lettow's Escapeinto PortugueseEast Africa, 1917". Army Quarterly, MII, (1925), pp. 50-5 1.
737WO 95/5294, War Diary GHQ, 26 October 1917.Appendix B, Telegram 17 WO, van Deventerto CIGS, 26 October.
733WO 95/5294, War Diary GHQ, 30 October 1917,Appendix B, Telegram21 WO,
van Deventerto CIGS, 30 October.

268

October, but the enemy was forewarned and withdrew down the Luwegu River, having
first destroyedhis two remaining field guns. Hawthorn followed up relentlessly and by
740
both
Liganduka's.
the end of the month
sideswere along the river at a point south of
The combined British-Belgian pressurewas felt most strongly by the Germans,
and by 28 October the local commander,Captain Tafel, was reported to be withdrawing
all forces from the Mahenge area towards Kahambu. Desertions were numerous and
many startedto surrenderto the Belgians in large numbers. It was evident that the whole
enemy front in the west was giving way, but Allied supply difficulties and long lines of
communicationpreventedthe full exploitation of the situation.
While the pursuit to the south of Mahengewas underway, Colonel Huyghd arrived
at Kilwa on 27 October to discuss future operationswith General van Deventer. With
Mahenge secure and his troops in pursuit of the Germans to the south, the discussions
centredon supply problems during the coming rainy season. Having learnt the lessonsof
the previous year, it was decidedto abandonthe advancefrom Mahengeto Liwale and to
withdraw the bulk of the Belgian forces before the roads becameimpassableat the end of
November. This would leave a garrison of only two Belgian battalions in Mahenge and
speltthe end of offensive operationsin the west. Furthermore, if the troops there ran out of
food or were attackedin strength,they were given advanceapproval to withdraw north of
the River Ulanga; van Deventerwas not taking any chances.
However, he was keen to use the Belgian RO to supporthis advancein the east. It
was to be reinforced to a strengthof about 2,000 rifles with artillery and services. Under
previousarrangements,the RO, led by CommandantHdrion, had left Dar-es-Salaamby sea
for Kilwa, landing there on 13 October. It was then to move forward to Mitondo where it
would halt and assistwith route maintenanceand the clearanceof any German stragglers.
If the areawere clear, then the RO was to push onto Liwale where it would receive fin-ther
instructions from the British in dealing with Abt Tafel. Finally, as soon as the rains

739WO 158/478,13 October 1917,Telegram92, Colonel Macdonell to Genstaff, 13 October.

740Van Deventer,Despatch1, 189.
p.

269

be
line
Kilossa-Mahenge
the
would
to
transport
support
the
used
motor
commenced,
R0.741
in
the
to
Kilwa-Liwale
sustain
route order
transferredto work the
I
began
body
Belgian
the
on
Following the meeting, the retirement of
main
November and only a regiment remained south of Mahenge. By mid-November, only a
RO
Mahenge.
At
in
the
left
this
time,
force
2,100
around
and
rifles was
of some
small
had been sent to Liwale to help push At

Tafel into the British net. On 25 November, all

Kilossa
the
to
Mahenge
forces
occupied
and
Belgian
recalled
were
around
remaining
bulk
2
December,
By
the
the
British
there.
of
handed
their
to
allies
territory was
over
forward western troops had reached the railway where they would shelter for the wet
742
season.

THE SITUATION IN NOVEMBER 1917

forward
his
Deventer
troops
with all
ordered
Anxiousto wind up the campaign,van
Lindithe
lines
with
continued
At
communication
time,
the same
of
work on
speed.
Similarly,
Kilwa.
between
link
Hanforce
200
from
Massassi
the
and
mile
routetakingover
The
Mtama.
improved,
the
tramline
Lukuledi-Mwiti
while
now reached
the
roadwasmuch
in
No
4
Column
battalions
Belgian
the
engaged
were
whole of
and
sappersand miners,two
fixther road building. The newly formed Mounted Column was also hard at work in
food
depots.
destroying
lines
Germans'
the
of supply and
attemptingto cut
By early November,the situationwas as follows. At Otto with about 6 companies
in
Tafel
in
Liganduka's.
Belgian
the
facing
Norforce
remained
areaof
column
and a
was
In
15
the
in
forces
the
companies.
the
some
area,
controlling
northern
overall commandof
holding
Mahiwa
25
the
disposed
Lettow
area,
with
seven
of about companies,
south,von
body
Newala
the
Mnacho,
two
and
main
of
to
two
mainly convalescentcompaniesat
one
at
743
Tafel
Indications
Lukuledi.
between
Nangoo
that
held
12-13
was moving
and
centrally
bulk
Norforce
British;
the
join
Lettow
to
the
to
of
accordingly
were evident
southwards
von
7"' CampagnesColoniales Belges,111,pp. 219-221. Seealso AnnexeNo 48, Confirence de Kilwa, d Kilwa de is. <Pemba>, le 27
octoher 1917,p.387-388; WO 33/953, TelegramsD 2, No. 2311,25 November 1917,Telegram42 WO, van Deventerto CIGS.
742CampagnesColoniales Belges,III, pp. 221-222.

270

Portuguese
head
towards
in
to
move
off any
an attempt
now swung to the south-east
744
territory.

its
Hanforce
to
November
The advance resumed on
marching southward
with
former positions around Lukuledi Mission, arriving there on 8 November. Further east,
Linforce, now suitably reorganisedand brought up to strength, began pushing Wahle's
Columns
2
the
I
day.
No
following
on
south-east
wheeled
and
the
rearguardsouth-west
Mission-Chiwata
track
Lukuledi
down
day,
the
and making good progress.
same
moving
On the flank, the Mounted Column was ordered to seize Mwiti about 5 miles south of
Chiwata in order to deprive von Lettow of a great deal of food and supplies,but also to cut
745
However, No 1 Column was then orderedto
Rovuma.
the tracks leading south to the
Plateau
Makonde
Ndanda,
the
the
northernedgeof
on
move north-eastto take the village of
in the belief that Wahle's rearguardwould haveto passthrough there.
The movementsof the Germansremained unclear to their opponents- Abt von
Liebermanwas fighting aroundNdanda while Wahle was moving south via a little-known
had
infonnation
detailed
lack
topographic
Once
from
Chiwata.
the
Nangoo
to
of
trail
again,
its
GHQ
Realising
the
down
British
recast
the
changedcircumstances,
advance.
slowed
drive
issued
for
Orders
November.
10/11
on
were
a
strong
northerly
the
plans on
night of
Chiwata, using No 3 Column from Ndanda and the Nigerian Brigade from Nangoo,
No
Mounted
I
by
No
2
Column,
from
the
by
the
and
while
west
supported an advance
Column swept south-eastwardstowards Kitangari, deep in the enemy's rear. This bold
being
Kitangari
Lettow's
indicated
intelligence
based
that
as
von
next
on
move was
British
Kitangari
""
the
However,
on
maps
struck;
misunderstanding
once again
objective.
for
headed
Simba's,
to
Gen-nans',
the
some
nine
miles
the
that
who
were
was not
on
sameas
the north-eastof the former place.
In the meantime,the pressurecontinued. Late on 14 November,No I Column had
seizedthe Mwiti defile and blocked any escaperoute to the south-west. The Mounted
7" WO 95/5294,War Diary GHQ, 2 November 1917,Appendix B, Telegram 23 WO, van Deventerto CIGS, 2 November.
7" WO 95/5294, War Diary GHQ, 9 November 1917,Appendix B, Telegram 29 WO, van Deventerto CIGS, 9 November.
7"sOrr, "Von Lettow's Retreaf', pp. 52-53.

271

Column had reachedMkunde and was pushing on to Kitangari. No 2 Column was about a
3
Column
No
just
Brigade
town,
Nigerian
the
Chiwata
and
the
of
north
with
mile west of
Plateau,
Makonde
to
the
the
German
fighting
hard
north-east
the
on
rearguards
against
was
747
Schutztruppe.
of that place. The net was closing tightly aroundthe
Chiwata,fell unopposedon 15 November to No 2 Column and the Nigerians. No 3
Column continuedits fight with At Wahleon the plateauwhile the Mounted Column was
day,
The
Kitangari
Lulindi.
Newala
between
next
at
and
orderedto cut the communications
it was clear that the enemywas moving south-eastwards
and the Nigerian Brigade was sent
I
feature
No
Hill
Mwiti
Makonde
just
below
the
the
and
while
to clear the ground
scarpof
Column was sent over the hill towards Lutshemi. No 2 Column was placed in reserve,
having cut a number of new roads, and moved to Mwiti, while, in order to deal with any
between
improve
track
the east edge of the
No
4
Column
to
a
was ordered
eventuality,
Makondeandthe sea.748
As this was happening,Wahle kept to the high ground, while von Lettow's main
body was moving just below the plateauand aheadof the Nigerians towardsNambindinga.
Again, the British believedthat placeto be Lutshemi,which was actually about a mile away
Lutshemi,
November,
No
I
Column
had
17
lower
By
occupied
on
nightfall
and on
ground.
from
its
Lettow's
difficult
rearguard.
source
von
water
after a very
march and secured
Again, using darkness,the Germanswere able to slip away to Nambindinga that evening.
Deciding to leave over 300 Germansand 700 askaris behind, the Kommandeur led his
troops neatly out of the closely-drawnnet and made his way east towards Simba's. The
British were now aware that Kitangari could not be the new objective and that the main
body was moving away from them.
The occupationof the Makonde Plateaunow beganin earnestand was notable for
the difficulty of movementrather than the ferocity of the defence. The major hospital and
baseat Chiwata was threatenedby a flanking move of No 3 Column to the north-eastand
746WO 95/5294, War Diary GHQ, 10 November 1917,Annex Bjelegram 30 WO, van Deventerto CIGS, 10 November.
747Off, "Von Lettow's Retreat7,pp. 53-54; WO 106/273,History of3 KAR, p. 84.
748Off, "Von Lettow's Retreat", p. 55.

272

No 2 Column togetherwith two Nigerian battalionsto the west. It was taken after a sharp
fight on 15 November. Nearly 500 prisonerswere taken and some70 prisonersof war were
beginning
the
figure
the
This
since
of
casualties,
the
of
enemy
aggregate
made
released.
4.1
killed,
two
1072
463
Europeans,
captured
or
surrendered,
plus
total
askaris
some
month,
749
1,000
inch guns,33 machineguns,and over
rifles.
The next objective was Lushemi which was held in some force. On 16 November,

No
half
the
drove
back
Nigerians
to
while
therearguards within a
mainposition
the
mile of
3 Columnadvancedsomefive miles to the south-eastof Chiwata,fighting a numberof
from
in
Mwiti
to
Column
No
1
an
attempt
north-east
and
east
moved
small engagements.
750
days
difficult
Two
the
force,
but
Lushemi
SJOW.
the
and
going
groundwas very
cut of the
later, the British were in possessionof Lushemi, the enemy having pulled back to Simba's
during the night of 17/18 November. A further large haul of prisonerswas taken from a
14
NCOs
700
242
20
few
civilian
and
askaris
as
as
well
camp a
miles away, some officers,
32
172
the
by
This
of
the
release
undamagedrifles and
seizureof
males.
was accompanied
751
allied officer prisoners.
Elsewhere,the march of the Belgian RO had progressedaccording to plan, and
its
blistering
Liwale
the
two
sun,
advance
guard
reached
under
after
weeks of marching
day
had
Column,
Shorthose's
linking
the
30
October,
arrived
of
who
on
up with elements
previously. The British handed over control of the village and then returned towards
752
At the same time, the transfer of a second
Abdallah-Kwa-Nanga on 2 November.
Belgian battalion together with mountain guns was underway, with the troops arriving in
Kilwa on 30 October; a third had been planned, but its participation had been cancelled
after its abortive march from Mahenge.
On the night of 1/2 November, Hawthorn attacked and seized the position at
Liganduka's, taking 24 prisoners and a machine gun. The remaining enemy on the
Luwegu retired to the east; the British followed, with Hawthorn moving to Kabati Mzee
7" WO 95/5294, War Diary GHQ, 15 November 1917,Appendix B, Telegram34 WO, van Deventerto CIGS, 15 November.
71 WO 95/5294,War Diary GHQ, 16 November 1917,Appendix B, Telegram 35 WO, van Deventerto CIGS, 16 November.
751WO 95/5294, War Diary GHQ, IS November 1917,Appendix B, Telegram36 WO, van Deventerto CIGS, 18 November.

273

having
Hawthorn
They
the
Moro.
Fair
Kabati
with
pressure,
towards
maintained
and
driven the enemy rearguards from both banks of the Luwegu by the evening of 5
November and to within a mile of Kabati Mzee. Fair reachedKabati Moro the next day
finding
140
Askaris
Germans
three
142
as
well
as
the
and
of
some
surrender
receiving
753
day,
On
82
damaged
hundreds
the
more
same
rifles.
of
machine guns and some
day
Column;
the
Belgian
to
the
or two more sick soldiers
over
next
askaris surrendered
it
from
Germans
Apart
Mlembwe.
this,
that
the
Kahambu
clear
was
and
surrenderedat
Hawthorn
Dapate.
turned to the south-east while
through
were now moving south
Murray, with 250 rifles, pushedeastalong the Songea-Liwaleroad.
At Liwale, there was no major activity until 14 November when a party of 16
Germans,22 Askaris and 85 porters from At Tafel surrenderedto their advancedposts.
Patrol contactsand desertershad already made it clear that this force was not far away.
At Abdallah-kwa-Nanga,Major Hawkins held on with a half battalion of KAR, anxiously
from
Tunduru,
Colonel
Shorthose
while
the
the
under
remainder
awaiting
arrival of
Murray, now with 400 rifles, pushedsouth-easttowards his position.
The Westtruppen,who were now on top of Hawkins, attacked shortly thereafter.
Heavily outnumberingthe KAR, they cut-off off the garrison and threatenedto destroy it.
Fortunately,the Belgians came to the rescue,sending a reinforced company from Liwale
754
November.
16
on the morning of 14 November and marching until early on
Surprising and capturing a water party, they learnt that seven companieswere attacking
the KAR; despitethe disparity in numbersthey moved forward, hitting the German main
body in the rear and scatteringthem and taking a number of prisoners. Apart from saving
Hawkins from destruction, the Belgian intervention also resulted in the capture of
documentsshowing Tafel's routes and order of march.
Huyghe met with van Deventer at Ndanda on 18 November to discussthe future
use of the RO. They agreedthat it would move to Massassiwhere it would assist in the
712CampagnesColoniales Belges, III, pp. 222-223.
75'Van Deventer,DespatchI p. 190.
754CampagnesColoniaks Belges, III, pp. 223-224 and p. 389,13 November 1917,TelegramNo 10873,Huyghd to Hdrion.

274

forces
Belgian
body
the
German
the
would
of
remainder
while
main
reduction of the
battalions
However,
Kilossa.
two
Dodoma
were already
additional
and
remain at
November
Mnero
join
RO,
the
and subsequently
to
the
of
end
at
reaching
marching
Liwale a few days later. 755

Van Deventer was ready to finish off the fighting. On 5 December,he informed

GeneralRobertsonthat he would takevigorousactionto roundup von Lettow beforethe

left,
he
With
two
of
campaigning
weather
months
an estimated
rains recommenced.
Nigerian
Rovuma
border
hold
the
the
to
the
strongly,
pushing
along
proposed
southern
Brigadeasfar southassupplieswouldpermit while GeneralNortheywouldmakethemain
he
if
Germans
However,
Mwembe.
from
the
Nyasaland
that
towards
recognised
advance
district
Medo
River
that they would effectively
Lujenda
the
towards
the
movedsouthof
756
placethemselves
out of reachin theperiodbeforetherains.
The pressurecontinuedandthe isolationof Tafel now assumedmajorproportions.
Fromcaptureddocuments,
the sizeof his forcewasnow knownto be doublethatpreviously
imagined. With over 1700rifles, it was a seriousforceto be reckonedwith althoughits
British
On
intentions
the
side,threecolumnswerecontinuously
werestill unclear.
ultimate
in
fourth
in
the rear areas. The
the
the
working
on
road
making
engaged
advancewith
difficult
for
from
Ndanda
Newala
the
and
wasstill very
situationwasstill not goodas road
757
improve
insufficient
light carsandtherewere
the situation.
technicaltroopsto
On 18 November, a liaison officer sent by Tafel find von Lettow was capturedby

the British. His papersrevealedthat At Tafel had left Abdallah-kwa-Nanga

on 16
758Consequently,
GHQ held
Novemberandthat it wasnow headedtowardsthe Makonde.
No 2 Column and the Nigerian Brigadeready to deal with this new threat. On 23
November,Tafelwasreportedto be on theupperBangallaRiverandthetwo columnswere
sentin pursuit.
75-'LesCampagnesColoniales Beiges,III, pp. 224-227; WO 33/953, TelegramsD 2, No. 2311,25 November 1917,Telegram 42 WO,
van Deventerto CIGS, p. 238.
756WO 95/5294, War Diary GHQ, 5 December 1917,Appendix A6, Telegram402, van Deventerto CIGS, 4 December.
757WO 95/5294, War Diary GHQ, 19 November 1917,Appendix B, Telegram 37 WO, van Deventerto CIGS, 19 November.
751Ridgway, "With No 2 Column", 11,pp. 256-257.

275

Concurrently,No 3 Column was following von Lettow down the track to Simba's,
Column
Mounted
Lulindi,
No
Column
the
I
towards
was reconnoitringnorthmoving
with
18
On
Simba's
No
4
Column
on
work.
reaching
on
road
was still
east and east, while
November,von Lettow was desperatelyshort of food and ammunition; it was now time to
he
into
Turning
Portuguese
territory.
German
East
Africa
south,
reached
and move
abandon
Newala on 20 November and conducteda final reorganisationof his bedraggledforce. All
1700
300
Germans,
behind
left
the
infin-n
as
remaining
the sick,
and uncommitted were
bank
River
Mpili
the
the
3000
to
on
north
of
askaris and
porters marched south-west
759
November.
21
Rovuma,arriving thereearly on

No I ColumnenteredNewalaat 1630hourson the sameday,just a daybehindtheir

Luatala,
Column
The
Mounted
sometenmilesnorthof the
and
reached
waspursuing
prey.
Rovuma,but the chancesof capturingvon Lettowwerefadingrapidlyandattentionturned
there
to Tafel. Simba'swastakenunopposed
with 52 Germansand75 askarissurrendering
further
126
Gennans
Newala
No
taking
I
Column
and
a
undefended
alsooccupiedan
while
78 askarisprisoner.Moralewasreportedto bevery low andthe troopsmuchtired,but von
760
force
Lettowwasstill holdinghis
togetherwith a combinationof exhortationandthreats.
The tension increasedas everybodynearedthe Rovuma River. On 23 November,
Tafel was locatedmoving south along the BangalaRiver towards the confluencewith the
Rovuma, which indicated that a link up was imminent. Two days later, von Lettow was
intelligence
indicated
in
hot
Tafel
No
I
Column
that
while
pursuit,
was
moving rapidly with
761
he
Newala.
was still moving towards
unawareof the main body's move south and that
Depite the difficulties of terrain and supply,the trap was closing. Tbree columnswere now
moving on Tafel while a fourth was attempting to catch von Lettow, who was now at

7" Orr, "Von Lettow's Retreat", pp. 56-57.

7'0WO 95/5294, War Diary GHQ, 21 November, 1917,Appendix B, Telegram 38 WO, van Deventerto CIGS, 21 November. The
statementon morale was "His force, both Europeanand African is reportedto be very tired of the campaign;but VON LETTOW
keepsthem going by exaggeratedtales of Germansuccessesin Europe and by assuringthe askaristhat G.E.A. will very soon be
handedback to the Germansand that every deserterwill then be hung. This is the enemy's trump card now, and is probably the only
thing that enableVON LETTOW to keep his askariswith him."
761WO 95/5295, War Diary GHQ, 25 November 1917,Appendix B, Telegram 43 WO, van Deventer to CIGS, 25 November.

276

Ngomano. Further to the west, Northey was manoeuvring to prevent any move northward
762
Tunduru
towards
.

On 24 November,it was clear that Tafel was still marching south along the Bangala.
River and the various columns were unleashed. No 2 Column moved south-west from
Massassi and No I Column left Luatala at 1600 hours. The weather was very hot and
763
It was here that the
trying, but the British columns were now at both ends of the river.
in
big
Having
Major
Pretorious,
captured a
the
off
measure.
paid
chief scout,
efforts of
letter from Tafel to von Lettow, Pretorious moved to the nominate junction, and cleared all
food and inhabitants from the area. On 27 November Tafel reached his destination, and,
less
Lettow's
than
a mile
to
marched
north-westwards
von
unbeknownst
either party,
764
away.

No I Column then hit the Wesuruppenin the front while No 2 Column ran into their
decided
On
27
November
food
Tafel
to
Now
surrender.
rear.
surroundedand very short of
1112
that
the
27
Germans
178
while
of
surrendered
rearguard
porters
askaris
with
some
and
The
day,
he
13
Tafel's
the
next
and
remaining
evening
offer of capitulation was received.
companiessurrendered,comprising some 1115 Europeans,3382 askaris, 43 machine guns
765
had
had
been
Schutziruppe
halved,
Tafel
four
At
the
also
while
and
a stroke,
guns.
destroyed1200-1400rifles, 25-30 machinegunsand two guns.766
Despite failing to catch von Lettow, this was a major successand marked the end
of the campaignin German East Africa. Considering the major fighting over, the British
likewise.
force
did
The
Belgians
RO
in
the
their
the
time
and
of
wastedno
size
reducing
returnedto the coastwhere it was shipped from Lindi to Dar-es-Salaarnprior to returning
to the Belgian Congo. By early 1918 all Belgian forces were ready to return to the Congo
767
Belgian
or the
zone of occupation. Henceforth, the campaign would be conductedby
the British and Portugueseonly.
762
WO95/5294,WarDiaryGHQ,27November1917,AppendixB, Telegram44 WO,vanDeventerto CIGS,26November.
7`3Ridgway,"With No 2 Column",11,p. 258.
764
Pretorious,
Maj P J,JungleMan,London:GeorgeG Haffap& Co, 1947,pp. 182-185.
765
Off, "Rumboto theRovurne,pp. 128-129.
7'AWO95/5294,WarDiaryGHQ,30November1917,AppendixB, Telegram46 WO,vanDeventerto CIGS,28 November.
767
Campagnes
ColonialesBeiges,111,
pp.227-228.

277

LETTOWZ
ESCAPE.
1910r.

If

k
0.
M

fl. CLg,

-!

V,
.
. Clo

4-II

CL u

WWI

CA

i'fi

5,

/1

/'1/I/I

Z,

IN
-%'v

if

Ljvow, -

e.
*'

'i/f'

s0.5

.9

SCCLL6
q Nuss.

.f's

do

2'T?,o

15

[Tofac-ep. 58.

be
known
Von
Lettow,
to
The political and military situation was evolving rapidly.
East
interior
into
Portuguese
Ngomano
the
had
disappeared
food,
of
and
south of
short of
Africa. On 2 December,van Deventersent a letter to Dr Schnee,informing him of Tafel's
further
in
to
troops,
for
the
avoid
order
the
remaining
surrenderof
surrenderand asking
768
in
he
At
time,
the
African
close consultationwith
was
same
the
populations.
suffering of
the War Office over the future of the campaign.
London sooncameto a decision.Given the feeblenessof Portuguesemilitary power
let
Lettow
deemed
dangerous
it
too
to
form
for
hatred
von
their
of colonial rule, was
and the
However,
the
the
Africa.
East
Portuguese
remaining
small
of
given
numbers
roam around
be
Indian
to
division
the
be
to
EAEF
were
all
of
units
to
size and
reduced
enemy,the
was
homes
be
to
their
West
African
to
Similarly,
the
were
returned
units
to
theatres.
sent other
both
in
being
terms
time
too
finther
of
Belgian
consuming,
co-operationwas seenas
while
land
desired
The
decision-making.
a
was
option
necessaryreorganisationand political
based
force,
the
Norforce
on
the
with
operations
striking
main
as
campaign, using
Mozambique coast to be avoided if possible owing to the shipping shortage. Finally,
largely
become
to
the
in
an
campaign
was
and
troops
were agreed
reductions numbersof
769
The continuation of the campaign
African one supportedby Imperial technical troops.
fighting
had
lost
bulk
Schuatruppe
least
but
the
their
blow
British,
the
to
the
of
at
was a
power.
TACTICAL INNOVATION

Tactics had moved on considerably since Smuts' departure. The policy of

attrition through a constant search for battle combined with operations against the rear
its
Innovation
in
African
East
also
played
part, as
the
setting.
areasmade good sense
defender's
the
to
advantageswith new methods.
column commanderssought neutralise
The Germanshad long basedtheir centresof resistancearound the sole track in an area;
the responsewas the construction of new roads in the forward areas by the fighting
76'WO95/5294,WarDiaryGHQ,2 December1917,AppendixA2a,LettervanDeventerto Schnee,
2 December.

278

follows.
defenders.
One
Pioneers
technique
to
the
worked
as
columns prior
engaging
beganworking on a track as the leading troops of the column marchedpast them, hacking
a way through the bush. This continued until the entire column had moved forward for a
fixed time and then halted. All troops then spent seven minutes of the 10-minute halt
cutting back the undergrowth on either side of the new track before resting. The pioneers
then moved up the now 5-10 foot wide track and beganto improve the surface for animal
The column then resumed the process. This simple method,
although tiring for the fighting troops, enabled columns to march on the enemy while
and vehicular traffic.

allowing the vital supply columns to follow up relatively closely. It representeda major
improvement in both tactical mobility and sustainability of operations as the previous
method of hacking only footpaths through the bush made linking up very slow and
difficult. Subsequently,once the areahad been cleared, engineersand road gangswould
develop the surfacesfurther.
Owing to the thick bush, thesemethodsenabledthe attackersto move close to the
Germanswithout being detected. Once ready to attack, flanking columns would blaze
tracks to a point about three miles parallel to the enemy position. From there, they would
strike inward towards the vulnerable flanks and rear. A further refinement to this
technique was the construction of a strong boma at the end of the new track. Now in
close proximity to the enemy, the column would establish itself behind thom and brush
defences,which could be erected in several hours. Once the Germans detected the
British presenceor tried to withdraw, they would be forced to attack the well-ensconced
flanking column. This neatly turned the relative advantageof attackers and defenders
aroundwhile also disrupting the evacuationof the position.
Van Deventer also graspedthe importance of good lines of communication and
devoted considerable efforts to building and improving the road network. Apart from
easingthe problems of supplies, they enabledhim to shift forces much more rapidly and
healthily. New equipmentwas usedto good advantage;the Lewis gun was introduced in
early 1917 to battalions on the same scale as machine guns. It was much more mobile
769WO 95/52394,War Diary GHQ, 3 December 1917,Appendix A4, Telegram 47022, CIGS
to van Deventer, 2 December.

279

than the Maxim and added considerable flexibility to the attacker who now used the
heavier machine guns in the firing line keeping the Lewis guns available for the assault
group. They increased considerably the firepower available to the attack considerably
and were very useful in fighting off counter-attacks.
Another weapon of considerableimportance was the Stokesmortar. While it was
heavy and its ammunition required a considerable baggage train, its high angle of fire
made it especially useful in overcoming strong frontal defencesas it could fire over trees
and small hills. It complementedthe Indian mountain guns and could be brought into
action quickly.

The Germans found it difficult to counter and disliked its effects

intensely. Finally, the Mills bomb, a hand grenade,enhancedclose range firepower and
was effective in both the attack and defence. All in all, the British forces had been much
more aggressiveand successfulin their aim of defeating the Germansin battle.

280

CHAPTER 8- THE OPERATIONS OF 1918

As soon as von Lettow left German territory, the various British columns were
it
long
for
However,
broken
the
the
phase.
was
not
next
units reorganised
rapidly
up and
forced a modification to theseplans. The advanceof At
before Portugueseunpreparedness
KdhI towards MontepuezthreatenedPort Amelia on the coast and British reinforcements
Colonel
formed
force
(Rosecol)
Rose
A
12
December.
to
under
was
were requestedon
landed
day.
intended
have
He
Coast
Regiment
Gold
there
the
to
the
the
next
secure port and
the column readyto move inland by the end of December,but much dependedon the water
770
be
supply and the stateof the port, which was reportedto primitive.
The supply situation causedconcernas ever. On 23 December,the needto support
the Port Amelia force forced a redeploymentof the Lindi forces. Ngomano,on the Rovuma
River, was held by the Nigerian Brigade, with a battalion in reserveat Massassi,and the
back
In
Mtarna.
fighting
to
the
the west,
troops
the
railhead
at
moved
were
remainderof
Hawthorn's Column was in the areaof Namwerawith a battalion operatingtowardsUniago.
However, an unlucky accidentwas to hinder the actions of Port Amelia force; a troopship
carrying the bulk of the Gold Coast Regiment ran agroundnear Mikindani and the troops
771
hadto evacuatedandreturnedto Lindi to await more shipping.
At the end of December 1917 the situation was largely quiet, apart from near the
Nyasaland border. Rosecol was completed at Port Amelia by 29 December while patrols
from the Rovuma River had vainly searched up to 110 miles south for the enemy. 'Me
Rovuma River was beginning to rise rapidly and the Nigerian Brigade was moved back to
the north. In the west, Northey's troops were in contact and were pushing east and north
from Mtengula and Namwera. Everywhere, the rain was beginning to fall and the roads
772
dissolve.
to
started
7" WO 95/5294, War Diary GHQ, 12 December 1917,Appendix B, Telegram 52 WO, van Deventer to CIGS, II December; 13
December1917,Appendix B, Telegram 53 WO, van Deventer to CIGS, 13 December.
771WO 95/5294, War Diary GHQ, 23 December1917,Appendix B, Telegram 62 WO, van Deventerto CIGS, 22 December; 25
December1917,Appendix B, Telegram 63 WO, van Deventerto CIGS, 24 December.
772WO 95/5294, War Diary GHQ, 29 December 1917,Appendix B, Telegram 68 WO, van Deventerto CIGS, 28 December; I
January 1918,Appendix B, Telegram 71 WO, van Deventerto CIGS, 31 December.

281

The main body of the Schuatruppe had narrowly escapedthe tightening British
Rovuma
Moving
fate
Abt
Tafel.
the
the
rapidly
along
valley,
of
coraon and was unawareof
their immediateproblem was lack of food as there were only sufficient suppliesto last until
773
in
I December. Ammunition was also very short supply and urgent action was required.
Von Lettow realisedthat the imminent rainy season,due to start at the end of December,
Furthermore,
halt
British
the almost totally
operations.
offensive
any major
would
deal
East
Africa
Portuguese
that
take
time
would
meant
and
a
great
undevelopednature of
force.
infrastructure
This
build
to
to
the
a
substantial
support
enemy
and
of effort
stocks
up
774
his
for
foraging.
first
deal
Portuguese
then
he
troops
the
split
that
and
with
meant
could
At the end of November 1917, the German intelligence assessedthe Portuguese as
having a strong force of some 6500 rifles, 42 machine guns and 14 guns with which to guard
775
400 k-m. Of these,they believed
border
River, distance

their

along Rovuma

of some

that Colonel Rosahad deployednearly 3,000 in a numberof postsbetweenthe coastand the

776
boma
The fort at Ngomano,which commandedthe confluenceof the
Ngomano.
major
at
Lugendaand Rovumarivers, was estimatedto supportbetween 1,800to 2000 rifles. Mile
thesewere substantialnumbersand well-armed,patrolshad deterrninedthat most postswere
777
had
Germans
Past
laid
the
experience
made
poorly
sited.
carelessly
out and
contemptuousof their putative foes and von Lettow actually viewed their garrisons as
isolated
Ngomano
bases
threat.
than
was
enoughfrom the
potential supply
rather
a serious
Portuguesemain body and sufficiently far away from the British basesto offer a tempting
target.
Although weakenedby hunger,the force deployedby von Lettow was still sizeable.
Led by Abt Goring, with three companies,as the advanceguard, next camethe main body
of 15 companiestogether with all the support troops, followed by a single company
rearguard. Altogether this made for 19 companies,totalling 268 Europeans,1700 Askari,
77'Boelt, Die Operationen,p. 399.
774Lettow, Reminiscences,pp. 236-237.
775MS Boell, N14/36,1. Kapitel, p. 1.
77"Boell, Die Operationen, p. 399.
7" MS Boell, N 14/36,1. Kapitel, 2. They had the 3,21,25,26,27,28
p.

282

Native Companiesand 2 machine gun batteries.

6,. rLy

"A4

AoitA
flit

c.... a

(.

II

Ae4ANA

L
\

__...

........

-,
*is .

't'*&"4

-. -/

'to

";

1 S.-"

.eis

.,

A.. A

. ".
! o# 0.4
ot, tA,

tj, Aff&

an

es

CiAlW4o

AA00.4

04.46

S<f

Lc4c\"

"W0i")

%l

11

.,.

\..

-.

0'-

'.

AKA I ALL
jp* "C-M&Avt

1.6it t

PAOM14

pmo

(AWA

%".
%,

%,

A.
.6%,

1.011

14

VI

02; AA; Ivlk

CIAORW
60116 of

2tZ4

600
in
included
Not
these
"boys".
370
were
over
numbers
and about 3,900 carriers and
778
find
had
their
"boys"
to
Askari
fin-ther
700
own rations.
who
wives and children,plus a
In order to preventingany reinforcementof Ngomano, von Lettow sentA bt Goring
body
boma
Nampakesho.
The
Portuguese
isolate
main
the
at
to
smaller
off separately
25
November
It
Ngomano.
Rovuma
and
early
towards
reached
on
was
continued up the
in
being
Despite
line
the
the
decided
Lettow
the
crossingof
river
to attackoff
of march.
von
full sight of the garrison,the latter remainedpassiveand did nothing to disrupt the attackers.
A mountain gun beganshelling the fort from the west while 4bt von Ruckteschelllaunched
Abt
Kohl
In
fixing
from
the
was sent on a southerly
the north-west.
meantime,
a
attack
by
it
fort
done
This
the
flank.
hit
Portuguese
stormed
supported
and
the
was
march to
its
the
fire.
The
aim at a cost
of
attack
achieved
shockand violence
artillery and small arms
killed
187
Portuguese
Schutziruppe
just
30
the
and capturedwell over
casualties;
of
over
779
The seizure also yielded six heavy machine
500 prisoners,the majority unwounded.
fighting
This
the
600
restored
guns,
rifles and a quarterof a million roundsof ammunition.
780
in
food
African
remained very short supply.
power of the Germansat a stroke,although
After a brief halt to treat casualtiesand collect their booty, the main column set off
down the Lugenda valley on 27 November. Von Stuemer's raid earlier in the year had
781
ftirther
but
in
food,
the area
south was much more promising.
shown it to be very sparse
However,it was somedistanceaway, and consciousof the precariousnessof his stocks,von
Lettow split up the force to forage independently. Fortime, and poor Portuguese
dispositions,led to the discovery by Abt Kohl of a small platoon-sizedboma at Nanguare
by
haul
300,000
70
It
it
duly
2
December.
and
rewarded
a
of
rifles
was
which
stormedon
food.
find
African
They
important
858
More
the
of
was
of
packs
roundsof ammunition.
journey
for
have
days'
to
the
through
coming
suffice
providedabouteight
rations and would
the largely uninhabitedand foodlessstretchthat lay ahead. Despitethe find, K6hl's carriers

77'MS Boell, N 14/36,1. Kapitel, p.3.

"9 Lenow, Reminiscences,pp. 229-232;.Schnee,Deutsch-Ostaftika, p. 311. He statesthat German casualtieswere eight dead and 25
wounded;MS Boell, N14/36, Kapitel 1, p. 3. About 400 PortugueseAskaris also ran away. The German battle strength was 675
soldiers, II machine guns and two guns; their losseswere sevendeadand 26 wounded.
"" Boell, Die Operationen, p. 399.
7" MS Boell, N14/36,1. Kapitel, p. 5.

283

in
local
deal
his
Askaris
of
effort
half
a
great
were spending
rations and
were surviving on
.
forag ing. 782

In the meantime,At Gdring was isolated and out of touch with the main body. It
had capturedsomePortuguesepapersshowing the existenceof a magazineat Nampakesho.
Marching rapidly, the garrisonwas able to bum the fort with the petrol stocksheld there and
his
G6ring
to
then
Concerned
few
marched
own
survival,
about
supplieswere saved.
only a
Ngomano where he expectedto find the rearguard. Arriving on 1 Decemberto find the
British there,he waited in concealmentfor two daysbefore setting south down the Lugenda
783
he
to
the
12
December
that
rearguard,
catch
up
It
able
with
was
until
was not
valley.
784
for
days.
link-up
learn
did
the
Lettow
another six
about
not
although von
Fortified by these local successes,the main body set off for the south on 4
December,trying to pass the foodless belt as quickly as possible. It reachedthe post at
Chirumbaon II December,surprisingthe occupantsand taking it without a fight. This was
flour
80
food
loads
African
800
another
of
were seized,easing
and
of
anothermajor coup as
followed
15,000
More
food
as
a
patrol
seized
another
success
the
situation considerably.
Wahle,
Finally,
Abt
300
which
mountain gun shells.
roundsof small arms ammunition and
had beenmoving slowly as it foraged,cameacrossa dug-in Portugueseposition at Ndimbo
in the Mkula Hills. Despitethe near parity of numbers,the position was easily overrun and
18
days'
food,
8
including
taken
large
was
on
quinine
and
valuable
quantity of supplies,
a
December.785This was followed by minor capturesof food and weaponsat anotherpost in
786
killed.
the Oizulo Hills on 27 December,in which 34 Portugueseand Askaris were

to completehis marchto the fertile areasin the

Von Lettownow hadthe resources
food,
him
from
Medo,
Lusinje
the
Chirumba
Mwembe
put
also
at
apart
which,
areaof
787
En-route,
detachments.
British
from
distance
a party underthe
the
various
maximum
792Boell, Die Operationen, pp. 399-400; Schnee,Deutsch-Ostafrika, p. 314.
7" MS Boell, N14/36,1. Kapitel, pp. 8-9.
7" MS Boell, N 14/36,1. Kapitel, pp. 20-21.
"' Schnee,Deutsch-Ostafrika,p. 320; MS Boell, N14/36,1. Kapitel, pp. 14-15.
7" MS Boell, N14/36,1. Kapitel, p. 30. Six Portugueseand 28 Askaris were buried there while 41 rifles, 4,700 rounds and 130 loads
of food were taken.
787MS Boell, N14/36,1. Kapitel, p. 10.

284

General
Deventer's
bearing
Ngomano,
to
flag
his
call
van
at
rearguard
reached
white
formally
8
January
Lettow
finally
It
dated
4
December.
and
on
von
reached
surrender
informed him of Tafel's surrender,although it appearshe had learnedof that fact following
788
It was too late to have any effect, for the
November.
29
the capture of Ngomano on
Germanswere now buoyedby their considerablesuccesses
againstthe Portugueseand were
approachinga land of plenty.
The Germanshad usedtheir first two weeks in Portugueseterritory very fruitfully,
had
They
680,000
from
to
taken
comfort.
relative
some
shortage
a
state
of
extreme
passing
12
The
750
together
machine
guns
and
rifles.
with
rounds of small arms ammunition

least
in
by
food
had
been
the
term,
the very
at
short
ameliorated,
greatly
critical
situation
789
blow
it
bitter
General
Deventer,
For
foraging.
local
van
as
wasa
substantial
capturesand
hadeffectivelycollapsedat nearlyeverystagewhile presentingthe enemy
the Portuguese
be
be
He
to
the
to
not
said
surprisedat such a
could
campaign.
with the meansof carry on
dismal performance,but it was highly disappointing,especially as he would be unable to
deploy substantialforcessouthof the Rovumafor severalmonths.

ALLIED PLANS FOR 1918

The end of 1917 saw politicians, and most especially Lloyd George, concerned
aboutthree major areas:manpower,casualtiesand Ireland. The heavy fighting in Flanders
had temporarily drainedthe BEF of its offensive power and the prime minister looked for a
successin Palestineto sustainpublic morale. With the slow build-up of American forces
and the collapseof Russia,the war seemedunlikely to end before 1919.790The need to
conserve manpower for the perceived final effort meant that reinforcements were
791
been
had
for
the shipping
available. For while the U-Boat
unavailable East Africa, even
threat had beencontained,it was far from being eliminated,with merchantshipping losses
in the latter part of 1917running at 241,260tons or 75 shipsper month.792
7UMS Boell, N14/36,1. Kapitel, pp. 6 and 11-12.
7"9Boelt, Die Operationen,p. 400.
7" French,Lloyd George Coalition, p. 169; Hughes,Brush Strategy in the Middle East, p. 27.
791French,Lloyd George Coalition, p. 180-181.
792Marder, 1917 Year
ofCrisis, IV, p. 277.

285

Another areaconcerningpolicy-makerswas that of the fate of the capturedcolonies.

From the beginning of the war, the British Governmenthad been careful not to declareany
it
had
Initially,
to
formal
territory.
wanted
add
not
on
captured
claims
annexationsor make
to the size of the Empire and had rebuffed attempts by South Africa and Belgium to
Africa
South-West
their
over
proclaim
suzerainity

and Ruanda/Urundi

respectively.

However, as the war progressed and the complexion of government had changed, attitudes
began to harden 793 Imperialists, such as Curzon, Milner and Amery, and sub-imperialists,
.
demand
Germany's
the
Botha
Smuts,
to
permanent
of
retention
such as
continued
and
794
overseas possessions.

An influential report in April 1917, presentedby a committee chaired by Lord

Curzon, looked for substantialgains in African and elsewhere. Supportedby the Imperial
War Cabinet, although not binding on the British Government,it showed the strength of
feeling on the issue.795The prolonged resistancein East Africa played its part, as lack of
indigenous
Africans,
the
had
impelled
British
to
the
use
of
and
widespread
make
resources
former Germansubjects,in their war machine. In Germaneyes,this constitutedtreasonous
betrayal,and widespreadretribution was threatenedagainstperceivedcollaborators. Smuts
noted:
"The Germans also assertedthat any native who had desertedduring the war
...
Gennan
Governmentwhen that war was over."
be
hanged
by
the
the
order of
would
This view was certainly sharedby Generalvan Deventer:
"No German deserterwill return to the enemy lines as he knows that he will be
forthwith shot or hung Last and most important is the questionof a guaranteeand
...
lose
German
East
The
Germans
they
assert
even
granting
assiduously
protection...
793Smith, Gaddis, "The British Governmentand the Disposition of the German Colonies in Africa, 1914-1918".Giffard and Louis
Britain and Germany,pp. 283-285.
794Louis, W R, "Great Britain and GermanExpansion in Africa, 1884-1919", Giffard and Louis, Britain and Germany, pp. 4042;
French,Lloyd George Coalition, pp. 63-64; CAB 24/4,8 December 1917,PaperG 182.German and Turkish Territories Captured in
the War, by Lord Curzon, p. 10.
791Smith, "Disposition of the German Colonies", pp. 288-289.

286

Affica now it will assuredlybe given back to them at the end of the war, and that
in
infidlibly
be
helPed
has
deserted
has
any
way
will
us
or who
every man who
hung. And this certainly would be the caseunlesswe can authoritatively refute the
""'
statement.
At a local level, the British realisedthat the Schutztruppewas vulnerablealthoughit
it.
left
in
had
still
considerable power and resilience

However, there were a number of

important factors to consider for the campaign of 1918. The first was the perennial problem
heavily
by
depleted
health.
Many
out
and
sickness.
now
worn
were
of manpower and
units
This was by no means restricted to the British and South Africans, as the Indians and West
Africans were also suffering badly. Believing that the native East Africans were better able
to withstand both the rigours of the climate and the ravages of disease, the bulk of non-KAR
fighting troops were sent to other theatres or returned home. However, substantial numbers
of Imperial technical and administrative soldiers were retained to support the new
campaign.797

Shippingwas a major limitation as heavylossesfrom submarinescontinuedto

be
East
Africa
both
that
to
could
as
the
supplies
sent
of
constrain
numbersand quantities
well as the intra-theatremovement of supplies. Becauseof limitations of vessels,port
capacitiesand unloading facilities, all arrivals into theatre had to come through Dar-esSalaam. It was uneconomicaland virtually impractical to sendpersonneland storesdirect to
the sea-baseof destination. Instead,a complicatedprocessof unloading at Dar-es-Salaam,
waiting and then loading before sailing to Kilwa, Lindi or Mombasahad to be followed. If
major operations south of the Rovuma River were to be undertaken,then it would be
necessaryto set-upand operateone or more ports in PEA, with all the extra effort and delay
that that implied.

796WO 33/953, TelegramsD2, No. 2203,6 October 1917,Telegram 380, van Deventer to CIGS; Louis, "Great Britain
and German
Expansion", pp. 40 and 4546; CAB 24/4,8 December1917, PaperG 182, German and Turkish Territories Captured in the War, by
Lord Curzon.
'" WO 33/953, TelegramsD 2, No. 2412,29 December 1917,Telegram X 9842, GOC to War Office, pp. 271-274. Out
of a total
strengthof 52,000 over 11,000 Indian, 10,000British and 3,700 South African troops remainedin theatre.

287

The commander-in-chiefwas well aware that the area south of the Rovuma was
from
But,
food
had
little
the previous year's operationsagainst
to
offer.
sparselysettledand
At von Stuemer,he also knew that the central districts of the Nyassa Company's holdings
he
for
Lettow
ideal
This
both
fertile
healthy.
as
could stay there
von
was
were
and
indefinitely, while his own approachesfrom either the coastor Nyasaland were long and illif
for
left
dilemma,
Germans
homs
by
He
the
the
undisturbed
of
a
served roads.
was on
could build up strength and recruit more Askaris, posing a potentially greaterthreat than
hitherto. On the other hand,a British advancecould disrupt and possibly destroythe enemy,
but at the cost of immense difficulties in supply and reinforcement. However, his
instructionsfrom London were quite clear and PEA would have to be cleared.79'
Van Deventer had three aims: the first was to fight the enemy wherever and
whenever possible in order to cause maximum attrition; the second was to prevent the
invasion of Nyasaland;and the third was to prevent the enemy from re-entering German
East Africa.799 In the short term, the approachingrains would make it difficult to achieve
the first goal although the others would be helped by the impasse. Once the dry season
be
needed.
would
policy
returned,an aggressiveand offensive
The regroupingof the British force had beenunderwaysincethe clearanceof
GermanEastAfrica, but had taken sometime owing to the needto extractunits from
forward positions,columnshad to be completelyreorganizedand fresh battalionshad to be
broughtforward. At the sametime, scarceshippingwas taken up with the withdrawal of the
Belgian troops,the removal of the substantialnumbersof prisonersof war, and the
evacuationof the many sick and woundedback to the baseat Dar-es-Salaam.As the
RovumaRiver would flood from Januaryonwards,it would be impossibleto maintain any
substantialforcesacrossit until the dry season.This meantthat the bulk of British forces
would haveto remain in the Germancolony. For the wet weather,van Deventerdeployed
the Nigerian Brigadenorth of Ngomano,with a KAR battalion at eachof Massassiand

7" WO 33/953, TelegramsD 2, No. 2334,2 December 1917,Telegram 47022, CIGS to

van Deventer,p. 246.
7" Moyse-BartleM TheKing's Aftican RV7,
es,p. 390.

288

Tunduru. The remainderof the main force was concentratedat Ndanda,for easeof supply
800

and training.

The new striking force would be largely East African and two brigade-sized
formations were createdat Ndanda. The columns were basedon the 2ndand 3rdRegiments
KAR, known as Kartucol and Kartrecol, each deploying three battalions of the respective
key
from
Songea,and southern
Norforce
to
a
role
operating
play
regiments.
was also
Nyasaland. Its principal column was based on the three battalions of the I" Regiment
KAR commanded by the redoubtable Colonel Hawthorn. Northey also had three
battalions of 4h Regiment KAR, together with various Rhodesianand South African units
deployed as garrisonsand a reserve.
Northey's lines of communications would remain essentially the same as those
East
Portuguese
Africa.
during
the
employed
earlier clearance of

Those for the main

force would be more difficult as the Rovuma was a significant barrier to movement in the
rainy seasonand the land to its south was barren. Any supplies moved overland would
have to come from Lindi via Massassito Ngomano before going into the largely roadless
Portugueseterritory. It was a long and tortuous route that promised to be difficult. On
the other hand, the sea offered the best possibilities for supply with the harbour at Port
Amelia being one of the best on the entire East coast. It also offered a relatively direct
route to the main body of the German forces, and, almost uniquely for the coast, it was
801
virtually malaria-free.
At the same time, the Portuguese continued to be alarmed by the progress of the
German incursion and especially about Port Amelia.

Van Deventer had already

despatched Rosecol to secure the port and now decided to develop it as a new advanced
base. However, a lack of porters meant that mobile operations beyond the immediate
802
be impossible for
the town
time.

areaof

would

some

8'0VanDeventer,Lt-GenSir J 1, DespatchftomtheLieutenant-General
Commanding-in-Chief
East4#tcan Force,30 September
1918publishedin LondonGazette,"Supplement",
No. 31069,16December1918,p. 121.HenceforthvanDeventerDespatch11.
go'Admiralty(NID),A Handbook
Nyasaland,
London:1920,p. 51.
ofPortuguese

289

Van Deventer was determined to bring the Schutztruppe to battle as quickly as

he
General
lack
by
delays
the
ordered
carriers,
of
and
a
rain
caused
possible and with
Northey to start pressuring the Germans from the west. His intention was to drive in the
803
harass
his
tired units.
German outposts in order to restrict the enemy's foraging and to
He was equally sure that operations with the Portuguese would be unsuccessful without a
War
Office
he
to
their
the
January
1
seek
requested
on
unified command and
804
him.
subordination to

The Portuguese at this time held a general line Mocimboa-do-Rovuma-Chomba,

805
fielded
Porto
Amelia.
The
Portuguese
Rovuma
between
the
a
the
and
area
securing
battalion
European
Askari
least,
force,
ten
and
companies
one
of
on paper at
substantial
and a battery of mountain guns.

They were deployed with the battalion and three

Molokwe,
in
Alto
towards
the
Nampula,
three
south
moving
companies
companies near
The
dispersed
four
a
of
stations.
amongst
number
companies
and the remaining
lack
immobile
but
Nakature,
to
due
battery
to
to
a
of
was
owing
move
was
mountain
806
carriers.

The uneasy relationship between the British and Portuguesehad to be kept on

track, for, despite their private misgivings, the British neededclose co-operation if there
between
in
Accordingly,
1918.
fighting
be
the
a
conference
was to
any chanceof ending
held
Marques
29
Lourenco
Governor-General
Deventer
on
at
the
was
acting
and
van
January. The national governmentshad agreedthat military operations in the Portuguese
807
lead.
British
This
be
the
taking
the
with
was
command
colony should
under unified
had
but
it
involved
January
disasters
the
shown,
a more tactful
of
absolutely essentialas
and sensitiveapproachto Portuguesenational pride.

$02WO 95/5294, War Diary GHQ, 29 December1917,Appendix B, Telegram 68 WO, van Deventerto CIGS, 28 December.
303Van DeventerDespatch11,p. 122.
8'4WO 95/5295, War Diary GHQ, 3 January 1918,Appendix A2, Telegram I B, van Deventerto War Office, I January.
603WO 95/5294, War Diary GHQ, 4 December 1917,Appendix B, Telegram49 WO, van Deventerto CIGS, 2 December.
3" Boell, Die Operationen,pp. 406-407.
m WO 95/5295, War Diary GHQ, 29 January 1918,Appendix A5(l), "Minutes of Conferencebetweenthe Acting Governor General
of Mozambiqueand Generalvan Deventer", 29 January.

290

for
Beira
had
to
Northey,
a conference, and
come
Van Deventer then saw
who
12
he
Colonel
Rosa
Mocimboa-da-Praia,
met
on
where
subsequently went north to
in
forces
Rosa's
trying
Northey's
both
co-operate
would
February. It was agreed that
and
drive
if
locations,
to
them south of
in
defeat
Germans
necessary,
or,
their present
to
the
Portuguese
desert.
Further
the
Askari
in
hopes
their
River
Lurio
south,
that
the
would
the
1,500
Chomba
the
to
attack
of
some
rifles
to
at
column
provide a mobile
were asked
battalions
hold
KAR
British.
to
in
from
the
were
sent
the north
conjunction with
enemy
Mozambique and the fertile Namule district (60-70 miles east of Lake Shirwa), while
between
line
lying
Ribaue-Inagu
those
the
forces
to
Portuguese
secure
asked
were
other
808

units.

An inspection of the main body at Mocimboa left van Deventer distinctly

to
bulk
troops
the
he
opinion
worthless;
an
the
militarily
of
unimpressedand considered
809
be vindicated during the courseof the campaign.
GERMAN PLANS FOR 1918
By January 1918, von Lettow had achievedhis immediate goal of positioning his
force in the more fertile portions betweenthe Rovuma and Lurio Rivers. His efforts were
from
food
the
troops
the
further
sheltering
effects
while
of
supplies
now set on obtaining
their
technique
to
He
the
of concentric
resume
the
enemy
expected
of
wet season.
810
had
dried
out.
movementsas soon as the rains stoppedand the country
His plans were to occupy the area between the River Lurio and the line
Montepuesi - Msalu - Mtende for as long as possible, and at least until the new crop
The
in
March-April.
harvested
be
protection of these supplies was a
could
sometime
high priority while he also knew that the longer his troops stayed in an area, stripping it
British.
As
it
denied
this
to
the
the
well,
position maximized
support
of supplies,
more
the length of the enemy's lines of communications with a concomitant weakening of
gooVan DeventerDespatch11,p. 123.
a" WO 95/5295, War Diary GHQ, 12 February 1918,Appendix A5, Telegram 36 S, van Deventer to CIGS, 12 February.
"0 Lettow, Reminiscences,p. 242.

291

forward fighting power.811 Thereafter, he would have to react to the actions of his
opponents.
For supply reasons, the Germans were broken into five groupings, with At
Wahle, three companies, in the area Likopolwe-Mwembe; At Goring, also with three
companies, around Muabala; At Otto, with two companies, en-route from Chirumba to
Luambala; Kommando,controlling a single company, at Chirumba; and Abt Kohl of five
812
Port
Amelia.
from
Namunyo
Muo
to
opposite
companiesstretching
OPERATIONS

JANUARY

APRIL
-

1918

Norforce was due to strike first in 1918, and by early January it was divided into
two main components:Colonel Murray and the Rhodesianelementsand two battalions of
the 4th KAR were garrisoning the Wiedhafen-Songea-Tunduruline and its vital food
supplies, while Colonel Hawthorn with three battalions of I" KAR, and the Cape Corps
813
into
Portuguese
territory.
were operating out of Nyasaland
Hawthorn's column was sentnorth on 3 Januaryto dislodge Abt Wahlewhich was
the closest force to the Nyasaland border and was now gathering food around the River
Lugenda. Arriving there on 7 January,he had two battalions with him, as the others were
further west, unable to move further owing to a shortage of transport.814 Von Lettow
received word of this two days later, and despatchedAt Otto as reinforcements. Wahle
met Hawthorn's troops above Luambala, causing some delay, but he fell back easton the
night of I 1/12 to protect his magazines. In the meantime,.4bt Otto had been instructed to
reinforce Wahle's threatened position and marched from Chirumba on 12 January.
However, difficult going and the rising river meant that he was only able to cross the
Lugenda River on 15 and 16 January;too late to effect a junction and increasingly short
815
boma
Wahle
16/17
the
the
on supplies.
on
night of
abandoned
and withdrew east. The
8" MS Boell, NI4/36,2. Kapitel, 72.
p.
812Boell, Die Operationen, 403.
p.
8'3WO 95/5330, War Diary Norforce, Entry 1 January 1918.
8'4WO 95/5330, War Diary Norforce, Entries 4
and 7 January 1918.
813MS Boell, 1414/36.2. Kapitel, 49; Boell, Die Operationen, 403.
p.
p.

292

J-t I P4

C-4W'rfk- R
t4 0.4cj

CAAosAAAq r-cj44A&I
af

f-Afr

$In

4"acA
.

L.

'.

41.

L1,01

OP(W-4

090co-40.4

Z*a
-01

lPdg4

Oct$
MOL

r\

lb

Ony

etib

A
IV
4

tv 3.04
LA A

"i

2/
f*

/-

10

.%%..

i.

%I,

%,

, A#V. 40

W-4,

N.

1401A)4;1"k
copippAq
CA,%

2? L&.

British occupied Likopolwe on 14 January and Luambala four days later, thus achieving
8
16
heavy
increasingly
by
hampered
the
rainfall.
their immediate goal, but now

As theseactionswereunderway,the rains continuedto fall with their customary

violence.

Further north, Colonel Fitzgerald's Kartrecol was engaged in its task of

had
It
line
Ndanda-Massassi.
GEA
border
the secondary aim
the
the
along
securing
with
banks
Rovuma
River
being
the
to
of
and of moving the
along
crops
of
ready
raze
population

northward

should the enemy re-emerge.

Finally,

it

was given the

Lugenda
Valley
Nanguare
building
but
job
the
to
and also
up
roads
of
vital,
unglamorous,
817
Tunduru.
west to

The Kommandeur also learned of the British activity in the Mkula Hills, further
food-growing
that
Hawthorn's
This
of
an
expansion
areaswas
advancemeant
north.
and

impossibleandthe expectedstocksof food in Mwembe- Chirumba- Luambalawould

decided
Lettow
body
by
days.
last
few
Faced
to
the
von
move
main
shortages,
only
a
backto Nanungu,via Mtende,while reinforcingChirumbawith a companyfrom eachof
WahleandK6hl.818
Northey's advancecame as a distinct surpriseto von Lettow, coming as it did
from the least expecteddirection.819Lettow decidedto intensify his food-gathering
despatched
Abt
Wahle
hold
to
the
to
trying
was
columns.
enemy
operationswhile
off
Chirumba to overseethe transportationof food stocks there while Abt Otto maintained a
barrier to the west. However, the local African population was now struggling to feed
itself and beganto hide food suppliesfrom the askaris.The rising waters and
deterioratingtracks meant that transport was considerably more difficult and slow than
previously. Operationally, the Germancommanderwas hamstrunguntil he could collect
820
food
to carry on elsewhere.
enough

916W095/5330, War Diary Norforce, Entries 14 and IS January 1918; Van Deventer Despatch11,p. 121, statesthat the latter place
fell on 14 January. This is contradictedby the War Diary.
317Moyse-Bartlett, TheKing's Aftican Riflcs, p. 390.
"a MS Boell, N14/36,2. Kapitel, p. 49.
819MS Boell, N14/36,2. Kapitel, pp. 32-33.
"'0 Boell, Die Operationen, p. 404.

293

Matters were not helped by a clever British campaign against his African troops.
Leaflets and rumours were spreadabout the hopelessnessof the German causeand
112
150
It
deserters.
as
askaris
and
carriers
to
treatment
effective,
was
offering good
desertedin the month of January;since this represented10 per cent of the nine western
821
had
Portuguese
The
the
ill-afforded.
it
be
success
against
of
effects
companies could
been offset by the unexpectedspeedof the follow-up and growing war-weariness. By
dismissed
2
February,
Lettow's
the
the
read
out
and
publicly
commanders
von
order of
822
deserters
British argumentswhile some
were caught and executed.
This, coupled with the Hawthorn's advance,led von Lettow to move his forces
further east,starting on 27 January. A bt Wahlewas ordered to act as rearguardin the
West while Abt K6hI fulfilled the samefunction in the east,with the main body heading
823
for the large supply depot at Nanungu. At the sametime, Rosecol had pushedinland
Pamune
24
January,
from
Amelia,
50
Port
the
of
on
where
village
reaching
nearly miles
it drove back the defendersand captured5,000 kg of food. Food was beginning to run
had
Kdhl
JanuaryAbt
by
in
forward
the
the
concentrated at
end of
out
area, and
Montepusi. 824 Despite the rain, Rosecol was able to occupy Ankuabe by the end of
825
further
The
January and maintained contact with At Kohl some 23 miles
west.
importance of guarding the ripening crops, due for harvest in mid-February, kept K6hl
firmly in position and prevented any withdrawal to help relieve the pressure in the
826
west.

This pressurewas maintainedthroughout the rest of Januaryand well into

February 1918. Operationswere heavily constrainedby the rain as many bridges were
827
being
drowned.
Wahle's
In the
askaris and carriers
washedaway with a number of
west, Hawthorn had clearedthe left bank of the Lujenda River and was forcing the

$21Boell, Die Operationen, p. 404.

822MS Boell, N 14/36,2. Kapitel, pp. 61-62;
$23Boell, Die Operationen,p. 404; WO 95/5330, War Diary Norforce, Entry 20 January 1917.
224MS Boell, N14/36.2. Kapitel, pp. 66-67.
$25Van DeventerDespatch11,p. 122.
226Lettow, Reminiscences,p. 245.
127MS Boell, N14/36,2. Kapitel, p. 62.

294

German rearguardeast,while Northey conferred with the commander-in-chief about

future strategyat Beira on 4 February."'
By early February,the British had four battalions facing the five enemy
On
7
February,
Nanungu.
K6hl
Otto
Mtende
Chirumba
the
and
companieson
road
829
were orderedto give up a company eachto support the threatenedarea. The
exhaustionof suppliesaround Mtende forced a contraction on Nanungu with Abt Goring
in
Wahle,
Abt
by
February.
latter
14
turn, was sent off to reconnoitre
the
retiring on
place
and to gather as many suppliesas possible. By 22 February, Hawthorn was at Mtende,
some 75 miles to the eastof Luambala while Rosecol occupied Meza a week later, it
having been abandonedby K6hl owing to lack of food.830

This renewedpressurewasworrisomeandlate in February,von Lettow madethe

protection of his food suppliesthe highest priority. He assessedthat he faced three
least
30 in the east. He believed
in
20-30
the
at
west with
columnsof some
companies
thattheseforceswould encountermajordifficulties in maintainingsufficientsuppliesand
that he could counterthemby concentratingthe maximumnumberof companiesto hit a
singleBritish column. Thereweretwo possibilities:eitherto marchnorth to the Lujenda
831
Hawella-Malema
Luarnbala
towards
or to go southto
TheBritish hadcontinuedtheir convergingmovement,with Northeyreportingthe
areanorth-eastof Luambalaclearfor 100miles andhis troopsoccupyingMalokoteraon
5 March. The sarneday, Rosecolhadclashedwith a Germandetachmentsome27 miles
to the eastof MedoBoma,which wasclearlyheld in somestrength.As it wasa key road
junction and led to the main body of the Schutztruppe,van Deventercorrectlysurmised
832
it
be
defended.
He decidedto reinforceRosewith anothercolumn
that would strongly
nd
battalions
2
KAR andhalf a mountainbatteryunderLieutenantColonel
two
of
of

"" WO 95/5330, War Diary Norforce, Entry 4 February 1918.

129MS Boell, N14/36,2. Kapitel,
pp. 63-64.
130Van DeventerDespatch11, 122; Bcmll,Die Operationen, p. 404; CO 691/15, War Diary Norforce, Entry 22 February 1918.
p.
831MS Boell, N14/36,2. Kapitel, 84.
p.
832Van Deventer,DespatchII,
p. 123; CO 691/15, War Diary Norforce, Entry 10 March 191S.

295

Giffard in mid-March.833Both columns were placed under the overall command of

Brigadier-General Edwardsand known collectively as the Port Amelia Force (Pamforce).
The third KAR battalion was sent on to Mozambique to protect the local areaand stiffen
834

Portugueseresolve.

At the end of March, the Schuatruppe were deployed in three main areas: the
Kommando and six companies were south-east of Chizona; At Wahle with two
K6hI
Abt
Nanungu;
with six companies and the mountain guns
and
companies were at
around Montepuezi. With the British closing in onto the new German positions, von
Lettow had been active in ensuring that the newly-ripening crops were harvested. A
considerablenumber of the Askari were employed on this crucial task while others were
had
foraging
been
March,
in
this
By
the
training.
successful enough to
engaged
end of
had
Germans
freedom
The
May.
food
last
their
the
to
to
now
of
end of
garner
enough
835
for
least
action restored,at
severalmonths.
THE BATTLE OF CHIRUMBA HILL

Despite the rain and generally soggy conditions, van Deventer was anxious to

bring his opponentto bay asquickly aspossible. He hadParnforcewith its two columns,
Rosecoland Kartucol, pushingwestwardson the Medo-Mwaliaroad againstAt Kohl,
Otto.
found
Abt
Parnforce
first
in
Hawthorn
the
the
up
with
closed
while
was
west
opportunityto attackwhen it ran into a strongGermanposition of aboutsix companies
aroundChirumbaHill, a long rocky outcropthat paralleledthe road runningfrom Medo
to Mwalia. GeneralEdwardsdeterminedto seizethe hill and to presson to Medo as
quickly aspossible.
The defenders had placed only two companies on the eastern edge of the hill,
where they could block the road. The remainder, under K6hl, were echelonedsouth-east
of the defences,ready to counter-attack or conduct an ambush. This was unknown to
'" Moyse-Bartlett, TheKing's Aftican Rifles, p. 39 1; WO 95/5330, History of 1"12'4KAR, p. 37.
834Van DeventerDespatch11,p. 124.
g'sBoell, Die Operationen, p. 406

296

by
10
the
the
the
evening
of
Rosecol who reached and seized
position
eastern end of
April. The next day, Kartucol was sent on a flanking movement to the south of Chirumba
Hill with the aim of cutting off the Medo-Mloco road and thence to Medo on 12 April,
Hill
Chirumba
to
the
holding
the
launched
Rosecol
cover
move.
along
attack
a
while
Giffard's column moved off successfully at dawn on II April and made
large
it
However,
despite
hill
bypassing
the
a
swamp.
encountering
reasonableprogress
836
depth
lead
in
its
By
K6h1
AN
here
he
the
positions.
that
mid-morning,
would meet
was
battalion was halted by heavy machine gun fire which continued until 1330 hours, when
Kbhl launchedall four companiesinto the counter-attack. In the meantime, Rosecol had
been pushing slowly westwards along the road towards Medo with two battalions. They
did well and began to push back the defenders. As the counter-attack was beginning to
threaten Kartucol, K6hl noticed that the Rosecol had occupied the high ground to his
flank. Despite trying to push them off, he was unsuccessfuland with the main position
battle.
fighting
dark
The
break
lost,
he
decided
the
to
the
continued
until
off
astride
road
left
form
to
the rearguard while the
AN
Spangenberg
two
was
companies
with
when
837
body.
remainderwent on to rejoin the main
This battle, the first major encounterin PortugueseEast Africa, cost both sides a
53,000
his
As
forced
K6hl
to
of
some
precious
rounds.
expend
number of casualtiesand
live
land
in
had
he
the
food
becoming
to
as
as
possible
off
much
short and
well,
was
Hill
his
he
Chirumba
had
loss
The
threatened
to
stores
and
of
order conserve stocks.
them evacuatedfrom Namunyo to Mdalamia, but was forced to fight another action on
16/17 April to cover that move. This action cost him another 46,000 rounds and he
continued to withdraw under pressure,while using fighting patrols to try and disrupt the
British supply system.838

""Clifford, TheGoldCoastRegiment,
pp.238-239;WO 95/5330,Historyof 1'12odKAR,pp.37-38.
$37
Clifford, 7heGoldCoastRegiment,
pp.244-245;MS Boell,N14/36,2.Kapitel, p. 113.Germanlosseswere13killed, 48
wounded,fourtakenprisonerandtwo missingoutof 620. TheBritishlost 18killed, 55wounded,andonetakenprisonerout of 2,000
rifles.
833Boell, Die Operationen, p. 407.

297

The British advance continued slowly against a series of small-scale ambushes

that were greatly enhanced by the very difficult

country and thick bush in the area.

Mwalia was reached on 20 April, with the columns having averaged about 4 miles per
day since leaving Medo. 839 Abt Kohl continued its move toward Nanungu, taking up a
position around Mblama on 24/25 April.

Despite the difficulty

of the advance and the

unfavourable conditions, the British scored a minor coup as Kartucol attacked a German
convoy and seized all the spare rifles, ammunition
840
companies.

and documents for K6hl's

six

Parnforce ran into heavy opposition on I May when a German counter-attack

nearly led to the loss of a mountain battery near Koronie. A flank attack onto Kartucol
including
42 dead and the loss of two
high
British
caused a
casualties
number of
halt
in
forced
This
temporary
the advance, as General
a
mountain guns.
setback
Edwards had to reorganiseafter the battle and bring up much-neededsupplies.841

In the meantime,van Deventerhad sensedvon Lettow's intentionsand ordered

his forcesto convergeon Nanungu.Northeywasinstructedto advancefrom his position
at Mahua while maintaining forces to the north to prevent a breakthroughin that
direction. Edwardswastold to carry on driving along the Mwalia-Nanunguroad while
ColonelRosawasaskedto usehis troopsto coverthe gapbetweenthe two forces.
In the west, Northey's troops had occupied Mahua on 5 May, having overrun and
disperseda German company the day previously. This move had been worrying von
Lettow, who had had moved westward with five companiesto block the advance. The
British columns reactedto the German responseby digging-in near Makoti, less than 25
miles south-westof Nanungu. On arrival, von Lettow launched a ferocious attack, using
two flanking movements. Despite the strength of the blow, it was unable to dislodge the
defenders and von Lettow had to regroup under cover of darkness. Casualties were
heavy on both sides with the Germanshaving over 100 and they had to amalgamatetwo
1"9Clifford, 77jeGold Coast Regiment, 256.
p.
"0 Van Deventer,Despatch11, 125.
p.

298

13.842
For
British
their
the
to
their
part,
suffered
strength
companies, reducing
overall
843
over 200 lossesand were forced to halt to sort out the casualties.
This sharp action led the British to believe that Lettow still considered a move
To
by
fact
that end, van
the
of
prisoners.
statements
north, a
apparently confirmed
Deventer ordered one column to hold the north along the Msalu River while the
remainder of Edwards's and Northey's troops would close in on Nanungu in an attempt
844
battle.
The advance resumed on 17 May with
bring
decisive
Germans
to
the
to a
Kartucol entering Nanungu unopposed two days later. Finding that place abandoned
except for a hospital full of sick soldiers, Edwards despatchedKartucol down the Mahua
road in pursuit of the withdrawing Germans. Rosecol and Grifcol were sent off on
flanking marches to the north and south of the road respectively on 20 May. 845 Von
Lettow had already left in a south-westerly direction along the Mahua Road going in the
direction that the British were least preparedand least able to deal with. He had organised
his force into four Ateilungen with the customary advance and rearguardsready for a
846
first
The
River
Lurio.
the
major stop was at Korewa, some 24 miles
move south of
kept
for
Kommandeur
Nanungu
the
alert
an opportunity to strike at
south-west of
while
an isolated column.
The next day, Kartucol ran into At Kohl, now the rearguard, and quickly started
working around its position. This the Germans to pull back further through the dense
bush and out of British clutches. Despite the break south, the situation was beginning to
look favourable for van Deventer as the bulk of the Schutztruppewere being concentrated
into a fairly tight area, albeit rocky and thickly vegetated, while the flanking columns
were pressing in. However, he was unawarethat von Lettow was now preparing to strike
back at his pursuers.
841Boell, Die Operationen, 407; Moyse-Bartlett, TheKing'sAfrican Rifles, pp. 394-395, WO 95/5330, Histoty
KAR, pp.
p.
of 10121d
3940.
942Van Deventer,DespatchII,
pp. 125-126;Boell, Die Operationen, p. 408. Boell cites German casualtiesas being 34 killed, 106
wounded. This represented26.7 per cent of their strengthwith three companiesreducedto lessthan 35 per cent while the British lost
103 killed, 100 wounded and six taken prisoner.
"3 Boell, Die Operationen, 408; WO 33/953, TelegramsD 2, No. 2650,6 May 1917,Telegram G 269, van Deventer to War Office;
p.
WO 95/5330, War Diary Norforce, Entry 5 May 1918.
"4 Van DeventerDespatch11,p. 126.
"s Clifford, 77teGold Coast Regiment, 269.
p.
' Boell, Die Operationen, p. 409.

299

Having first sent off the baggagetrain to a safe distance, von Lettow instructed
Abt Kbhl to launch a spoiling attack on the morning of 22 May. It was not intended to be
decisive, just enough to disrupt the advance and enable the rearguard to break clear and
into
battle
into
Kartucol
the
back.
0900,
At
as
road
went
a narrow
came
slowly move
its
battalion
K61il's
lead
hills.
The
by
impassable
flanked
troops
engaged
with
gorge
flanking
the
Stokes
to
effect
while
columns moved on.
mortars good
mountain guns and
Grifcol emerged from the south onto the Mahua road to the rear of K61il's main body,
it
driving
Moving
it
its
then captured
off.
and
meal
on,
midday
surprising a company at
Abt K6hl's baggagewhile cutting the German force in half. "" Von Lettow was not far
behind, with both Abt Poppe and G6ring, and immediately advanced on hearing the
firing. In danger of being trapped, K6h1 had begun withdrawing through the gorge when
the Kommandeur arrived. A furious counter-attack was launched onto Grifcol by eight
dislodge
but
Germans
K61il's,
including
to
the
the nowtwo
were
unable
of
companies,
kept
hard
Lettow
Kartucol
Further
pressing
and
von
was
east,
surrounded column.
later
despatch
to
two
a third to protect this flank. His blow had
companiesand
compelled
failed, and, in considerable danger of being destroyed in place, he ordered the battle
broken off. The night provided an opportunity for the weakenedSchuatruppe to escape
through the densebush and regroup to the south."'

Thetwo-daybattlecostboth sidesheavily,but especiallythe Germans.They had

suffered heavy casualties,losing over 100 troops and 300 carriers. Four companieswere
both
batteries
destroyed,
Governor,
the
artillery
and At Kohl losing their
effectively
with
baggage. Over 30,000 rounds had been expendedwhile a further 70,000 were captured
by the British; this left only 613,000 for the entire force. Furthermore, food was running
short and the whole situation had a distinct resemblanceto November 1917 when the
force had barely escapedfrom a similar scrape.This setbackforced von Lettow to break
849
in
contactand march south slowly order to evacuatethe seriously wounded and sick.
"7 WO 95/5330,History of 1`12"dKAR, pp. 4142.
"a WO 95/5330, War Diary Norforce, Entry 22 May 191S.
"' Boell, Die Operationen, p. 4 10; MS Boell, N 14/36,2. Kapitel, pp. 166-167. Of a Germanstrengthof 880,17 were killed, 88
were
wounded,317 taken prisoner (296 were carriers) representingsome 10.5 per cent of their strength; WO 33/953, TelegramsD 2, No.
2674,24 May 1918,Telegram 2 10 WO, van Deventer to War Off ice, p. 3 54. The British reportedlossesof about 70.

300

From the British point of view, the battle had represented a chance to end the
it
difference.
As
have
it
join
battle,
the
been
made
might
able to
campaign. Had Rosecol
to
the
heavily,
columns
advancing
cost
Germans
some
at
the
although
mauled
were
was,
Nevertheless,
fighting
they
in
were still not
their
strength.
and a significant reduction
beaten and the British had run into the limits of their supplies again. Nanungu marked a
both
Germans
to
in
the
severe
pressure
gain
were under
the campaign as
turning point
food supplies and ammunition.
THE MOVE SOUTH TO MOZAMBIQUE

Reorganising on the spot, Rosecol was broken up while Kartrecol deployed a

battalion to securethe road along the line Malema-Mozambique while another was sent
left
It
Ribane-Maleme
to
detachment
the
Portuguese
was
road.
to stiffen the
along
Kartucol and Grifcol to carry on the pursuit of von Lettow in parallel columns. Rearguard
hospital
left
by
broken
the
capture
of
another
up
actions were the norm and were only
1
behind by the Germans.850

Von Lettow remainedaswily as everandquickly movedhis reducedforce south

his
His
to
head
half
day
aim
was
opponents.
now-halted
on
start
s
gaining a
and a
food
live
to
his
diminished
to
off-,goingsouthoffered
stocksand secureenough
replenish
the bestopportunitieswhile putting more strain on his opponents'supply lines. By 27
May, the advanceguardhad reachedthe River Lurio and had reconnoitereda suitable
851
Watiwa.
crossingsitenear
By I June,the Germanshad almost completely crossedthe River Lurio at Watiwa
between
difficult
break
the
the river and Inagu and
through
threatened
to
area
and now
into the more settled areas around Mozambique. The land was exceptionally difficult

few
indigenous
The
little
bush,
hills,
thick
paths.
cultivation
and
with numerous
very
"0 Van Deventer,DespatchII, pp. 126-127. One capturedhospital yielded 15 Germans,47 Askarts, and 31 Porters as prisonerson 28
May; CO 691/15, War Diary Norforce, Entry 27 May 1918.
851Boelt, Die Operationen, p. 412.

301

incoming
Germans.
the
to
assist
willing
population was also strongly anti-Portugueseand
Given the past weaknessof the Portugueseforces, van Deventer had to assumethat they
deploy
had
his
to
be
incapable
therefore
own troops to
would
of serious resistanceand
his
from
troops
their
Rosa
Colonel
to
He
of
some
positions
them.
move
support
asked
just south of the River Rovuma to the port of Quelimane and to be prepared to move
them inland from there. The final arrangementswere agreed at a meeting in Dar-esSalaam on 8 June, although subsequentlythe Portuguese commander's enthusiasm for
Instead,
Rosa
instructed
Deventer.
by
had
be
to await
to
was
van
attacking
curbed
base
fortification
the
the
to
of
supply
at Nhamacurra,at the
reinforcement and was ensure
852
Quelimane.
line
end of a small railway
north of
The German push south forced van Deventer to reconsider the ever-lengthening
lines of communication. As things stood, Port Amelia was too distant, but the shifting of
the entire base by sea to the port of Mozambique was out of the question owing to
from
down
Medo
Nanripo,
Consequently,
to
cut
was
road
a
motor
shortagesof shipping.
853
Mcuburi
finally
Nampula.
and
on to
where the Lurio was crossed,and then south to
Van Deventer was particularly concerned about the ability of the Portugueseto defend
themselves and decided to reinforce their key garrisons with his own troops. A new
column under Colonel Fitzgerald, named Fitzcol, was formed from two KAR battalions
854
Muo Nluku. Hopingto box the

in
he
Malema
the
area,
senta unit to Inagu
enemy
at
columnconcentrated
at
while anothermovedinto Malokoterawith an Anglo-Portuguese
Ribaue. Furthernorth,both Grifcol andKartucolwere still strugglingto crossthe Lurio
in the faceof Germanrearguards,
finally makingthe crossingby 5 June.855
The Germans were wasting little time and managed to evade the converging
columns on the Malema line. The advance guard, At Miller with three companies,
attacked the Portuguese boma at Malema after it had been reinforced by several
companiesof KAR on 31 May. For once, the attack was unsucessfuland the defenders
$52
D2, No. 2688,1June1918,
W09515295,WarDiaryGHQ,21 June1918,EntryQuelimane
Area;W0331953,Telegrams
Telegram218WO,vanDeventerto WarOffice,pp.357-358.
$33
VanDeventerDespatch
II, p. 128.
'-"4
Boell, Die Operationen, p. 412; WO 95/5330,History of 1`12"dKAP, p. 43.
'" Van DeventerDespatchII,
p. 128; CO 69 1/15,War Diary Norforce, Entry I June 1918.

302

1. [Asti

Ar4scA
otuploAll

41014-INO
,

- -iArv
...

ff

rot

40

_"_b.

-.

*'

Ott

e
14,

ell

ew

I
**'
fs.
jwlS

t.AW0

(MCLI

oftv"

p,

La

---'i Z

141-

! '- ,*..,

10

MmOmod.,c

/
&AAR

Z.

J4*M

f'

16
AAA-L

_\

J_.

0.

\
T

_s.

..

'

\\

7
'-:

.0

4+
CAJO

Ct&UA-

4-

A.

At
10'

co &&,A
-I

.-.

00 ACA000 0,

held firm. "" Muller remained there for a few days and then marched off, rejoining von
Lettow and the main body on 8 June. The Kommandeur then decided to march in the
direction of least resistanceand sent off Muller towards Alto Molokwe. Arriving there
Portuguese
having
its
found
two
boma
12
June,
of
companies
the
empty, garrison
was
on
fled without a shot, and a rich haul of food and documentswere taken. Most importantly,
the captured papers mentioned an incoming ammunition column. Muller immediately
followed up this lead and went off searching the local area for its whereabouts.
Occupying Ille, Alto Ligonye, Nampave and Muyeba in turn, he seized 2200 loads of
food and various materials. Furthermore,the tracks of a column were found and Mueller
food
23
kg
June. A subsequent
150,000
supplies
other
on
numerous
of
and
captured
attack on a lone Portuguese company yielded several machine guns, some food and
13,000 rounds of ammunition. Despite these considerable successes,the munitions
857
column could not be located and ammunition remained scarce.
After halting on the Malema line to sort out the supply situation, on 16 June van

Deventerorderedthe three mobile columns to move south, with General Edwards

base
The
Mozambique
Mozambique
theatre.
the
the
new
at
port
of
assumingcommandof
Colonel
landed
lead
Rosa's
day,
troops
the
the
of
units
northern
at
while
openedup
next
neighbouringQuelimaneon 20 June. It wasvan Deventer'sintentionto trap andhammer
the Schutztruppebetweenthe villages of Ille and Alto Molokwe, and then to drive it
towardsthe coast. GeneralHawthorn,now havingtakenoverNorforce,broughtthe bulk
of his troopsto Ille while Philcol movedto NakwaandAlto Ligonye;Fitzcol wasdueto
link up with them near Ribawe,while the Portuguesewere relegatedto securingthe
But, van Deventer miscalculatedas his opponent had already
coastal towns."'
determinedto continuesouth-eastandjust beatthe British to Ille. On 25 June,the lead
unit ran into the Germanadvanceguard,AN Spangenberg,but was unable to push
through. While fighting was underway,the enemymain body and rearguardbypassed
859
fighting
the
andslippedaway.

$56CO 691/15, War Diary Norforce, Entry 2 June 191S.

857Boell, Die Operationen, p. 413.
8" CO 691/15, War Diary Norforce, Entries 15,20 and 24 June 1918.

303

More troops were now brought into the fighting as Kartrecol was taken off its
duties along the Rovuma and brought south. Fitzcol left Mbalama and after a long and
30
June.
Nampula
This
Edwards
General
joined
on
was now more
at
exhausting march,
important as von Lettow threatened the port of Quelimane which supported a major
Nhamacurra,
base
terminus
a number of miles north.
the
at
railway
supply
at
Nhamacurra was essentialfor the support of any columns operating north of that port and
had a large stockpile of food and essential stores. Van Deventer was especially
Colonel
Rosa
it
he
to
its
reinforce
ordered
while also
and
concerned about
security
least
he
On
battalion
KAR.
two
half
at
occasions,
sent explicit
of
providing a
instructions for its defence, charging Rosa with responsibility for the execution of his
information
important
British
This
the
of
were
short
owing to the hostility
as
orders.
was
860
in
Lettow
was the area.
of the local population, but they did realise that von
On paper the garrison was formidable, with three Portuguese infantry and one
defences
laid
The
KAR
two
the
together
companies.
were
out
with
artillery companies
being
the
distance
3,000
three
western
and
central
sectors,
allocated
over a
yards with
of
to the Portugueseand the easternto the British. The railway station formed the right of
the position and was held by the KAR. The Nhamacurra,River marked the left before
861
defences.
swinging around the rear of the

On 27 June,von Lettow sentoff his advanceguard,At Miller, with ordersto

body
day's
intelligence
the
main
with
remaining
a
march
obtain more ammunition and
behind it. Friendly Africans informed MUller of the position at Nhamacurra and he
from
Emerging
bush
it.
decided
the
to
only 30 metres from the
promptly
march on
Portuguesepositions around the sisal factory in the western sector, he was surprised to
run into the enemy. However, they fled at literally the first shots and the position quickly
collapsed, with substantial quantities of weapons and ammunition being captured.
Moving toward the centre of the position, MUller's troops ran into further Portugueseand
but
forced
British
troops,
were
eventually
out.
some
who put up someresistance
859Boell, Die Operationen,p. 414.
80 WO 33/953, TelegramsD 2, No. 2718,17 June 1918,Telegram 234 WO, van Deventer to War Office, pp. 369-370; No. 2729.22
June 1918,Telegram 238 WO, van Deventerto War Office, p. 371.

304

Consolidating on the objective, he awaited the arrival of the main body who duly
MI.
KAR
located
had
the
2
July.
Patrols
the
at
railway
station
positions
and
arrived on
held
KAR
disciplined,
dug-in
it.
the
their
Well
positions and
to
and
now resolved
attack
862
Heavy machine gun and rifle fire continued
inflicted heavy losses on the attackers.
throughout the day as well as most of 3 July.

At 1500 hours, one of the captured

Portuguese mountain guns was made serviceable again and began firing.

The first round

hit the railway station building and drove out a number of Portuguese 4skaris and carriers
.
in a blind panic. They stampeded through the KAR positions causing chaos while the
Germans took advantage of the confusion to infiltrate.

Now overwhelmed and being

began
The
defenders
from
to
tried to withdraw
the
the
collapse.
position
rear,
overrun
drowned
down
deep
Nhamacurra,
their
to
the
and
many
rear
or
were
shot
as
over
river
863
they tried to escape.

The resultwasa disaster.The Portuguesegarrisonhadbeenlargely wiped out as

food,
importantly,
More
KAR
vast
the
quantities
of
weapons,
well as
companies.
hands,
Germans'
into
fell
the
relieving their supply worries at
ammunitionand stores
once. It gavethemthe ability to fight on for someconsiderabletime. They seizedsome
five heavymachineguns,threelight machineguns,484 rifles, 327,000rounds,300,000
kg of food,and,critically, 3.75kg of quinine.864
This disasteralso causedconsiderablepanic to the Portugueseadministrationas
Quelimanewas now directly threatened.On 4 July, Colonel Rosaorderedthe families
evacuatedand the bullion storesplacedon ships as a precaution,while there was wild
talk of abandoningthe town. However,the British resolvedto hold it and von Lettow
had other ideas. He realisedthat the Zambesiwas in flood and thereforeimpassable
while the British strengthlay in their ability to move by sea. He decidedto reversehis
stepsand to headnorth-eastwith the secondaryaim of disruptingtheir newly created
linesof communication.Marchingin five groupswith the standardadvanceguardof Abt
$61

MS Boell, N14/36,3. Kapitel, pp. 216-218; WO 106/273,History of3 KAR, p. 89a.

262WO 106/1460,SchneeDiary, [2-4 July] 1918, pp. 146-147.
163Moyse-Bartleft, TheKing's Aftican Rifles, pp. 399-400; WO 106/273,History of3' KAR, pp. 90-9 1.

305

865
However,
5 July.

Miller and rearguardof Abt KAI, the Gennans set off on

owing to
the loss of contact and active deception by the local population, the British lost track of
these movementsfor a few days.
Finally, Philcol ran into the leading elements of the Schutztruppe near Ociva on
11 July. Thus warned, General Edwards sent Fitzcol north from Nhamacurra towards
Ociva while the redoubtable Kartucol was ordered to try and head off the Germans at
Murrau. Van Deventerreleasedthe last of his battalions along the border of German East
866
Mozambique.
from
Fort
Johnston
battalions
to
Africa, sendingtwo KAR
The chase was back in full flight. A Portuguese post at Mtiba fell easily, but
Philcol and Kartucol were now closing hard, and K6hl had to hold them off as the main
body slipped off north towards the small post of Namirrue, held by two companies of
KAR. Fitzcol reachedthe Namirrue River about 25 miles west of the boma on 20 July.
Setting off the next day, it could hear the sounds of battle from the direction of the
it
heavy
bush.
Pausing
through
the
could
the
as
as
quickly
garrison;
column marched
Alto
Ligonha-Namirrue
Fitzcol
the
July,
briefly
21/22
onto
emerged
the
only
on
night of
by
Lettow's
the
find
boma
to
the
six
companies
attack
of
von
under
road
was already
main body while the three companiesof advanceguard securedthe flank.

Fitzgeraldwaswell awareof the potentialseriousness

of the situationandhis lead
battalionpressedon to reachthe beleagueredgarrison. By 1600,the leadingGerman
later
hour
in,
but
had
been
driven
a strong attack checkedthe column's
outposts
an
progress. The enemywithdrew at nightfall and both battalionsdug-in in line. Von
Lettow hadbeeninformedaboutthe threatto his rear andhad launcheda night flanking
attack. At 1900hours,it hit the right flank of the forwardpositionwhich wasstill under
break
into
Schuatruppe
The
to
the defencesand routed the
construction.
managed
battalioncompletely. The shockwas so completethat the commandingofficer and his

"4 Boell, Die Operationen, p. 415; WO 106/1460,SchneeDiary, [24] July 1918,pp. 146-147. He statesthat over 100 drowned.
165MS Boell, N14/36,3. Kapitel, p. 230; WO 95/5330, War Diary Norforce, Entry 4 July 1918.
'" Boell, Die Operationen, p. 416; CO 691/15, War Diary Norforce, Entry 13 July 1918.

306

headquarters staff were taken prisoner. 867 Stragglers quickly reached the second battalion
desperate
fight
A
2100
hours,
impending
Fitzgerald
threat.
to
the
ensued
until
and alerted
by
battle
broken
the attackers.
the
off
when
was
withdrew to protect its baggage train.

A much-weakened column then

The next day revealed that one battalion had

less
200
to
than
the
to
effectives. Against
was
reduced
other
practically ceased exist and
the main body of the enemy, the column was now helpless and could do little more than
defend itself. 868

The defeat of Fitzcol sealedthe fate of the unlucky garrison holding the boma at
Namirrue. Cut off from its water supply and overwhelmingly outnumbered,the post fell
869
hours
23
1700
July
final
However it did cost the
and surrendered.
on
after a
assaultat
killed
in
dear,
the attack and another severely
was
attackers
as one company commander
870
idle
been
far
from
it
had
had
left
Munevalia
Kartucol
During
time,
this
as
wounded.
on the night of 2/3 July. It had spent some considerable time chasing rumours of the
Germansheading south towards Quelimane when news was received about the disasterat
Nharnacurra. The column returned to Munevalia on 14 July, having marched 187 miles
bush.
Despite
heavy
day
17
their considerablefatigue,
through
or an averageof
miles per
there was no opportunity for rest and they marched straight off towards Tipe in an
attempt to headoff von Lettow before he could crossthe River Molocque.

Kartucol reachedits objective on 20 July only to find At Kohl holding the

crossings. Showinggreatdetermination,crossingswere forced and the column pushed
up againstthe Germansuntil 22 July when firing could be heardfrom the directionof
Namirrue. The enemy'scampwaslocatedon 24 July, but the columndid not attackuntil
871
the arrivalof reinforcements
anda muchneededfood resupply. It thenadvancedto the
site of the campto find it evacuatedand burnt out. On hearingthat the Germanshad
turnedeast,Giffard ordereda forcedmarchto Calipo,north-eastof Namirrue,wherehe
thenhaltedto link up with the remnantsof Fitzcol. A brief rest ensuedas Kartucol had
967

Boell,Die Operationen,
TheKing'sAfticanRiges,p. 402;CO 691/15,WarDiaryNorforce,Entries22 and
p. 417;Moyse-Bartlett,
23 July 1918.
1" Moyse-Bartlett,
TheKing's AfticanRifles,p. 402;WO 106/273,Historyof3d KAR,pp.92-93.
'69VanDeventer
Despatch1,p. 130.
870
WO 106/1460,
Schnee
Diary,Entries22-23July 1918,pp. 152-153.
'71CO691/15,WarDiaryNorforce,Entry24 July 1918.

307

food,
blankets
kit.
little
in
330
no
or
personal
marched
a month with
miles
872
remarkable achievement.

It was a

The unsatisfactorymilitary situation led to another meeting between General van

Deventer and the Governor-General at Quelimane on 22 July. It was decided to give
each nation separatezones of operation, with the Portuguesebeing allocated the defence
Ennes
local
inland operations.
from
Antonio
Quelimane
together
the
to
with
of
seaports
The British took upon themselves the task of maintaining forces south of the line
Mozambique-Malema.

A week later on 29 July, van Deventer met with Edwards at Nampula to decide
future operations. The Germans were now known to have turned again and this time
headedeastto the areaof Chalaua,which was fertile land lying between the Ligonha and
Meluli Rivers. The local population was in a state of near revolt and was actively
its
hated
invaders
the
the
and
colonial
power
allies. Notable too, was the
assisting
against
Malema
line
desertions
the
the
of
crossing
virtual cessationof
on 10 June which
since
may have been attributed to the great distances between the Askaris' homes and the
location of the fighting forces.873

Von Lettow's calculationshad been influenced by documentscaptured at

Namirruethat indicatedthat the areato its north hadfew suppliesof food. The areaeast
of Ligonye had a numberof Portugueseposts,mostly abandoned,and, althoughover
13,000loads of food were discovered,the lack of porters meant that most had to be
destroyed.The Kommandeurdecidedto go eastand marchedto Pekera,arriving on 24
July,beforemovingon to Chalauatwo dayslater.874The forcemanagedto spenda week
in the areaandthe lack of a closepursuit gavethe force time to considersomeessential
had
been
It
measures
possibleto makeup the diminishednumbers
regardingmanpower.
of Askaris through the promotion of the best carriers, but in the eight months in
Portuguese
territory over a third hadbeenlost (1190from 1790)while 310 goodporters
'72Moyse-Bardett,
TheKing'sAfticanRifles,p. 402.;WO 106/273,Historyof3 KAR,pp.93-94.
'73VanDeventerDespatchIf, p. 131.
'74WO 106/1460,
SchneeDiary,24-31July 1918,pp. 154-155;Latow, Reminiscences,
pp.290-291.

308

f. Arr
Ar4itA

*
k

&"' L..,
L. k44

?t;i
A04ASA

R::

0A

"

Li

/ ot"

/ \( .("a

00
.

va

2vo.1V
QCJAV

, %

or

AAIOQ

It. -

I"

001-0--631
-4

)IA-d

OVVA

.1

W'O k1

17

Nb

&at
11

I
No

Ado

4-.

op.

%.

AL

rI

op"

k.
')'?

! f-\

hIAAJIO%-.

-o
4- 4- ++

0'

C"Aim
PAK

lp

eft.

0-4 XOIWA.
a4f7wif
to,.,

had been enlisted to replace them and began immediate training. However, more serious
was the diminition of the irreplaceable German officers and non-commissioned officers
who had dwindled by nearly 30 per cent in the same time (197 remaining out of 281).
The loss rate of 10-11 per month was increasingly serious,particularly when the need for
875
becoming
battles
leaders
Pitched
too expensive and
good
were
was especially acute.
were working in favour of the British.
While these operations were underway, the British continued their work on
improving the communications in PortugueseEast Africa. The extension of the motor
road from Medo to Nampula was completed quickly and large numbers of vehicles were
moved via this route to the Mozambique line rather than by sea. The road leading from
Mnapo to Chinga was extended to Ribaue and Malema, while General Hawthorn's
engineerscontinued the road from Nyasaland to Malokotera. Similarly, the Portuguese
built a road from Ngomano through Chomba to Medo. The net effect of this activity was
that, by the end of August, it was possible to drive from the Rovuma to Medo, through
Nampula and Malema and on to Zomba in Nyasaland.876It was a major achievementand
greatly aided the movement of suppliesand troops. A number of smaller tracks were also
constructed,as were improved telegraphic links from Nyasaland and the major ports of
Quelimaneand Mozambique.

Theseefforts'wereto havebeneficialeffectsin the next stageof events.With a

brief restbeinggrantedto the mosttired units,a numberof smallercolumnsweresent
out to protectthe vulnerablelinesof communication.Then,on 9 August,van Deventer
initiateda convergingmovementon Chalauawith the aim of trappingvon Lettow. The
latter,of course,hadno intentionof beingcaughtandhadalreadystartedto collecthis
dispersedforce,with a preliminarymoveon 7 AugusttowardNamatil asa feint towards
Mozambique.His real aim wasthe British magazinein the rich areaof NyamaroiRegone.Therapidly closingBritish forcedhim to halt on 10Augustandthenhe turned
southfollowedby westandthennorth-west,in a successfulattemptto confusethe
*75Boell, Die Operationen,
pp. 417418.
'176
Van DeventerDespatchII, p. 131.

309

877
fateful
have
to
However,
telegram
consequencesas the
was
a captured
scent.
Kommandeur learnt that the supply centre at Mukubi was being evacuatedto Regone.
878
for
Regone.
This tipped the balanceand he now ordered his columns to make
The threat to Nyasalandwas never far from van Deventer's mind and the latest
his
flank.
him
led
by
the
Schuatruppe
to
of
western
the
reinforcement
order
move
General Hawthorn was orderedto hold Regonewhile Kartucol was sent rapidly to Alto
Molocque; the remaining troops were instructed to follow the Germansas closely as
Abt
Mfiller
into
battalion
been
had
Numarroe
On
24
August,
and
reached
ran
a
possible.
forced
being
back
battalion
fight
KAR.
heavy
A
to the boma at
the
of
ensuedwith
Regone.879It was an advancedsupply baseof some considerablevalue, as it contained
10,000loads of food, 500 casesof small arms ammunition and 200 of StokesMortar

bombs;its losswould haveseverelyhinderedBritish operationsin the areaapartfrom

880
to their resources.
providingthe Germanswith a largeaugmentation
Von Lettow wanted to take Regoneby the use of small side paths, thereby
heavy
fog,
his
through
defences.
However,
the
miscarried
rain and
plan
avoiding main
difficult going. He reachedthe bomaon 26 Augustto find it well defendedandrecently
he
its
In
the
columns,
approaching
madethe decision
and
reinforced. view of
strength
881
instead.
not to attackandto moveoff to the north
While these movements were underway, Hawthorn's troops had dug themselves
in south of Lioma while Kartucol had its three battalions spread between Inagu and
Muanhupa having arrived there from Alto Molocque. A hastily reinforced Fitzcol also
On
30
Schutziruppe
defences.
August,
located and
Muanhupa
the
to
to
the
to
went
add
attacked the Norforce in its entrenchedpositions, with eleven of its twelve companies
being committed to the assault. Despite being hit from all sides, the battalion held on
tightly and was relieved by the arrival of Fitzcol on the same day. That unit sent in a
$77WO 106/1460,SchneeDiary, 10 August 1918,p 3.
87'Boell, Die Operationen,
pp. 418-419.
'79Boell, Die Operationen, p. 419. The Germanslost sevenkilled and 25 missing; the British 51 dead,sevenwounded prisonemand
44 prisoners. Two Lewis guns and 40,000 roundswere seized.
3'0 Moyse-Bartlett, King's. 4ftican Rifles, p. 405.
"' Boell, Die Operationen, 419; WO 106/1460,SchneeDiary, 27 August 1918,p. 5, Lettow, Reminiscences,pp. 294-295.
p,

310

from
Muanhupa
battalion
by
followed
early on
arriving
that
another
counter-attack
was
31 August. This resulted in the taking of Abt Maller's baggage column and the loss of
50,000 sorely needed rounds. Realising that he had stumbled into a strong position, von
882
hours.
from
1500
Kartucol
Lettow changedhis plans and began moving north
about
followed up with its usual vigour as it located the enemy crossing the Muanhupa River on

the next day; its lead battalion launched an attack from the west supported by another
from the east. Although able to deflect these blows, the Germans' lost of two of the
883
field
hospital.
Governor's key staff officers, a medical officer and the
The operations in the Numaffoe-Regone-Lioma area had cost both sides heavily.
For the British, one battalion had taken a heavy pounding and the remnants of a second
Germans
However,
Numarroe.
the
largely
finished
suffered one of their
were
off at
in
detail,
British
defeat
Trying
the
they
Regone.
to
Lioma
were
very
closest escapesat
losses
heavy
the
in
Casualties
of military equipment and
and
were
nearly caught a trap.
further
killed
70
Sixty-three
a
with
wounded, while
were
medical stores were marked.
884
by
British.
some250 valuable carriers were captured the
Prisonerstaken by the Schuatruppe indicated that yet another KAR battalion was
Malema.
Increasingly
desperate
Molocque
from
Alto
to
to
and
the
now
east coming
decided
to move north and recross the River
Kommandeur
British
the
the
trap,
escape
Lurio, which was now a trickle. Forced marches between I and 3 September were
885
by
British
On 5
interrupted only by attacks on the rearguard the persistent
columns.
September,the Germanshad their first rest day in 25, but could only pausebriefly as the
pursuerswere close at hand. Supplies were again growing very short, and the country
had
German
devastated
incursions
from
the
to separatewidely
and
columns
was
previous
in order to seekfood.

912
1918;Moyse-Bardctt,
7heKing'sAfricanR07e$.
CO691/15,WarDiaryNorforce,Entries31 AugustandI September
p. 406.
8" WO 106/1460,
SchneeDiary,[31 August)1918,pp. 10-15.
"'WO 95/5295,WarDiaryGHQ, I September
1918;MS Bocil,N14/36,3.Kapitel,p. 307. On 30 August,theGermanslost27
killed, 58woundedand14asprisonersout of 667troops(14.6percent).A furthersevencarrierswerekilled and78 missing.The
Britishwerereportedto havehad5 killed and133woundedoutof 900(15.4percent). On 31August,theGermanslosta further
threekilled,elevenwoundedand22 missingwhile 197carrierswerecasualties.Furthermore,
some480,00rounds,tenrifles,40 loads
kit werelost.
of medicalsupplies,40 companyloads,30 foodloadsand115loadsof European
"3 WO 106/1460,
Schnee
1918,p. 17.
Diary, I September

311

Shortcol.was sent aheadto try and prevent the crossing of the Lurio, but collided
day
for
its
left
Kartucol
the
5
September.
camp
on
same
and
made
with the rearguard on
into
bush
detour
German
the
But
Germans'
quickly confused
the
a
expected position.
its
(and
front)
it
to
Kartucol
north
when
thought that the enemy was some way
matters;
body
in
it
had
hit
the
Quite
column of
main
while
unexpectedly,
came under contact.
march.

This unexpected collision caused great confusion in both opposing columns as

the rear battalion of Kartucol clashed with the middle of the German line, that of 4bt
Midler.

The engagement was fought in terrain with very limited visibility

and it took

dispersed
both
hours
for
to
the
to
sides,
owing
nature of
arrive on
several
reinforcements
the respective columns. The confusion was exacerbated by the Germans running into the
believed
They
had
that
transport.
they
their
the
chanced
rear of
column and seeing only
The
KAR
heavy
initiated
were
placed
under
attack.
a
rapid
upon a supply convoy and
four
two
the
it
had
the
column
was
some
and
of
to react quickly, while
rest
pressure as
miles ahead respectively.

It took several hours for the second unit to come into action

lead
battalion
Finally,
hours.
1400
the
German
reached the
the
and stop
attack at around
battle at 1700 hours and launched a final counter-attack.

This was successful, and the

battle was broken off and the Germans moved off into the bush. Kartucol had suffered
heavily in the encounter and was unable to continue the advance which was given over to
886
Shortcol. Again, the Germanslost heavily in terms of officers and suffered against the

887
British
defences.
good
The inevitable reorganisation after battle included the collection, treatment and
deal
200
had
Kartucol
to
the
over
with
stretcher cases,including
evacuationof
wounded.
the enemy's, and deliver them to the Lurio where they were met by the MAC. This was
13
September
battered
it
KAR
that
8
September,
the
until
not
completed on
while was
battalion was able to rejoin the column.
The battles at the end of August and early Septemberwere painful to both sides.
The Germanshad lost 39 soldiers killed, 133 wounded and 51 captured, as well as 437
Ateflung
leaders
2
lost.
impact
Telling
the
the
structure
as
command
carriers
on
and
was
"" Moyse-Bartlett, TheKing's Aftican Rifles, pp. 407-408; CO 691/15, War Diary Norforcc, Entry I Septembcr 1918.

312

3 company commandershad been either killed or seriously wounded. The force was
baggage
40
deal
lost
together
had
12
with
of
packs of
a great
reduced to
companieswhich
"'
crucial medical supplies.

The weather was also hurting the Germans' health as the

intense heat of the days was followed by severecold at nights, coupled with little food or
889
For
in
September.
broke
their part, the
influenza
lung
An
out early
water.
epidemic of
British had lost 8 soldiers killed and 71 wounded from one unit alone on 6 September.
THE BREAK TO THE NORTHAWST
By now, von Lettow was anxious to avoid another series of battles and did his
best to break free; Mwembe was reached on 17 September after another exhausting
890
burnt
Shortcol
its
find
for
Germans
or
removed.
to
could
stores
the
all
march only
difficulties.
Durcol
to
in
battalion
reached
supply
the
pursuit owing
only support one
Inagu and Malema in early September,only to be disbanded once the enemy crossedthe
River Lurio. Kartucol lost one of its battalions which went into reserve near Tabora,
September.
28
Rovuma
the
the
two
on
while
crossed
remaining
While the reasons for this loss of momentum were quite clear to General
Hawthom, the commander-in-chief was very displeased. A sharp telegram accused

Norforce of having lost its drive and determination. While Hawthorn'sangry rebuttal
had somejustification, it vergedon the insubordinate.Therewas truth in both views,as
but
it
let
had
the
slip,
equally
was clear that
opportunities
someof
columncommanders
the overall strategy had not been a great successeither. Firmly rebuked, Hawthorn was
given the task of pursuing von Lettow northwards and of bringing him to battle at any
cost.

891

The break northwards naturally gave concernsabout the safety of both Mahenge
familiar
Lettow.
to
Songea,
fertile
von
very
and
and
areas
which were
3" WO 106/1460,SchneeDiary, Entry 6 September,pp 21-22.
us Boell, Die Operationen,p. 420.
ls')WO 106/1460,SchneeDiary, 8 September1918,pp. 24-25.
2'0WO 106/1460,SchneeDiary, 17 September1918,p.25.
891CO 691/15, War Diary Norforce, Entries 25 and 26 September1918.

313

The newly

developed road system came in extremely handy as units were re-deployed throughout
September. Kartucol was broken up, having marched an incredible 1600 miles, crossed
29 large rivers and fought 32 engagements.Its well-worn troops were sent into reserveto
in
Massassi-Ndanda.
Africa
East
Fitzcol
German
the
the
area
secure
southernportion of
was disbanded and the bulk of units transferred to a new formation, Cenforce, to be
commandedby Colonel Fitzgerald. Cenforce was basedon the Central Railway between
Morogoro and Dodoma with columns extended south to Mahenge and Iringa. 892 Van
Deventer's aim was to use his strategic mobility to reinforce threatened areasas required
while keeping up the close pursuit by Shortcol. Hawthorn recalled his forward units to
Fort Johnston, preparatory to a move by steamer up Lake Nyassa and thence to
Sphinxhafen. He wanted to catch the Schutaruppe in the flank before they could reach
the Rhodesia-Nyasalandborder. Orders were given by GHQ to send Norforce to Songea
and Ubena, with battalions sent to garrison Mahenge and Iringa while the remaining
infantry battalions were concentratedaround Massassi.893

However,the age and decrepitudeof the vesselson Lake Nyassameant that

frequentbreakdownsimpededthe speedof deployment. Hawthorn's lead battalionleft
Fort Johnstonon 18 September
at Sphinxhafenby the endof the month.
andconcentrated
It too wastired, havingbeenmarchingandfighting continuallysincethe previousMarch.
The Rhodesianpoliceunits followedup asquickly asresourcespermitted.894
If the Germans had made good their escape from the clutches of General
Edwards, they were not having an easy time of it. Health worries impinged in a major
way as influenza and pneumoniahit the force. Weakenedby the continual marching and
poor rations, all members of the Schutztruppe and its followers were suffering from
serious weight loss and weakened resistance to disease. At times 50 per cent had
influenza and each company had 6
Only
90
pneumonia
cases.
of the worst cases
-8
could be carried and the rest had to walk. Between I and 22 September,12 askaris and
20 carriers had to be abandoneddue to exhaustion, while 24 askaris and 67 carriers had
1192
Moyse-Bardett, TheKing'sAftican Rifles, p. 409.
393WO 95/5295, War Diary GHQ, 19 September1918.
894Moyse-Bartlett, Vie King's Aftican Rifles, 409.
p.

314

deserted.895Matters cameto a head on 20 Septemberwhen Governor Schneesent a letter

to General von Lettow stating that the carrying along of the seriously ill was an
be
by
lives
they
that
the
abandoned
should
as
advised
unjustifiable sacrifice of
and
doctors.896 This went down poorly with the Kommandeur and he retorted that to do so
would irretrievably weaken his combat power and many of the sick would recover in
time. The situation was not helped by the continued reports of the unfavourable military
897
in
situation Europe from the remaining wireless receiver and captured newspapers.
Having rebuffed the Governor, von Lettow then called in all of his Abteilung
leaders and emphasized their mission of tying up as many Allied troops as possible and
898
bad
from
by
Europe.
be
depressed
It was a
telling them not to
the apparently
news
for
being.
been
However,
have
the
time
to
the
that
staunched
crisis of confidence
appears
had
had
had
day's
in
beginning
they
the
to
only
as
one
columns
rest
pressure was
show
increasing
At
African
Kraut
35
Desertions
troops
the
were
out of
amongst
marching.
lost 300 of its experienced carriers in two days while over 200 prisoners of war managed
899
to escape.

Having kept his force together with the strength of his personality, von Lettow's
lans
30
Rovuma
to
the
eastof Mitomani on 29 Septemberand
next move was cross
some
900
It had been a difficult
re-enter German territory after an absenceof some 10 months.
time, particularly from British pressurein the latter months, and the German force had
shrunk from 278 to 168 Europeans,from 1,600-1,700Askaris to just over 1,000, and the
experienced carriers from 4,000 to about 2,000.901 While the fighting power of the
Schutztrupperemained formidable, the constant sufferings and losses were taking a toll
of its morale. At

Wahle was broken up on 2 October and the force became even more
902
concentratedthan ever.

195MS Boell, N 14/36,3. Kapitel

p. 32 1.
'96WO 106/1460,SchneeDiary, 19 September1918,pp. 31-32.
s97MS Boell, N 14/36,3. Kapitel,
pp. 321-322.
39'Boell, Die Operationen, 42 1.
p.
WO 106/1460,SchneeDiary, 14 September1918,pp. 26-27.
WO 106/1460,SchneeDiary, [29] September1918,p. 38.
WO 106/1460,SchneeDiary, 30 September1918,p. 39.
"2 WO 106/1460,SchneeDiary, 2 October 1918, 42.
p.

315

Patrols soon found that the British were holding Mitomani and captured
documents indicated that two further companies were en-route from Sphinxhafen to
Songea. This resulted in an action on 4 October as Abt Spangenberg blocked the
Wiedhafen-Songea
the
the
while
road,
remainder of
these
along
progress of
companies
903
Although successful,
the force bypassedto the west and continued their march north.
this action cost another30-40 casualtieswhile both. 4skaris and carriers were deserting in
904
greater numbers.
The options facing the Germansat this stagewere several, as the British could not
direction
Furthermore,
in
their
the
the
troops
to
choice.
of
move
rapidly enough go off
from
the
due
in
the Central Railway to
December
this
roads
make
would
and
rains were
Iringa and Mahenge unusable. Therefore, if either place could be seized prior to then, it
With
bulk
following
May.
the
dislodge
the
be
them
impossible
of
to
until
would
almost
the British forces in the long process of redeploying from the heart of PortugueseEast
Africa, it would be sometime before a suitably large formation could be assembled.
As a precaution, van Deventer ordered the reinforcement of the Lindi-Tunduru
both
Iringa
Mahenge
in
while roads
to
the
and
garrisons of
area the south with additions

leadingsouthfrom the railway wereimproved. The main reservewaskept on the Central

Railwayand,asTaboraappearedthe most likely objective,GeneralEdwardswasplaced
905
in commandwith his headquarters
there.
Having bypassedthe first blocking force, von Lettow continued his rapid march
Hawthorn
his
his
intentions
to
opponents.
was anxious to stop him
north,
still unknown
battalions
by
Nyassa
two
to
Lake
the
ordered
move
ship up to Alt
and
at
northern end of
Langenburg as quickly as possible. With shortagesof carriers and food, it was only
Furthermore,
in
the
keep
battalion
Lake
the
the
to
poor
condition
pursuit.
possible
of
one

fleet meantthat two of the threeavailableshipsbroke down in the middle of one movc.
90'Boell, Die Operationen, 42 1.
p.
9" WO 106/1460,SchneeDiary, 5-8 October 1918,pp. 44-50. About 30 Askaris and 40 carriers disappearedin this period; MS Well,
N14/36,3. Kapitel, p. 327. The Germanslost eight killed, 53 wounded and II missing out of 283 riflcs. Losseswere II per cent.
90"Van Deventer,Lt-Gen Sir J L, Despatchfrom the Lieutenant-GeneralCommanding-in-Chlef East. Vrican Force, 30 September
1918,published in the London Gazette,"Supplement" 31310,26 April 1919,p. 263.

316

The battalion was finally complete at Alt Langenburg on 18 October and further
reinforcements still en-route.
On 15 October the KAR regained contact with Abt Kohl and maintained it over
the course of the next few days. The Germans were also successul in foraging as some
906
during the
10 days'
On 17 October,
found

engagement.
von Lettow felt able to have a rest at Ubena, the first in 17 days. It was here that the
redoubtable General Wahle was wounded. It was a remarkable effort by a 66 year old
supplies were

and carried off

man, who had previously retired from the Saxon Army and who had suffered a double
hernia and much malaria during the arduous campaign.907 He was left behind with a
number of sick and woundedtroops and their followers.

On 19 October,the columnsmovedonto Gambawanowhere a quantity of food

was collected over the next two days, while the British were unable to pursue and
remainedstatic in Ubena. Here, the Kommandeurdecided to turn west and enter
NorthernRhodesiainsteadof carryingon to Taboraas expectedby van Deventer. His
troops had largely recoveredfrom the epidemicin September,but badly neededrest,
while the low stocksof ammunitionconstrainedtheir ability to fight battles.
Northern Rhodesiahad not been touched heavily by the war since late 1915 and
there were few defencesto slow down an invader. Von Lettow considered going on to
the West Coast and Angola in particular, as neither the Portuguesenor the Belgians could
do anything meaningful to stop his movements. The British would have to redraw their
extensive lines of communication and commence another campaign. All of this time
would enablenew harveststo be gatheredand the Schuatruppe would be able to continue
their actions, suitably refreshed.908 On the other hand, the dire situation in Europe was
apparentto the German element and the desertion of carriers continued unabated. As the
Governor recordedin this diary:

' WO 106/1460,SchneeDiary, 15 October 1918,

pp. 66-67.
"7 Boell, Die Operationen, 422.
p.
90'Boell, Die Operationen, 422.
p.

317

"Everyone hopes for peace deliverance from out situation, which grows ever
...
leave
its
or
unconsciously
will
mark
consciously
more unbearable and which
hardships,
danger
the
The
to
terrible
constant
and
privations
permanently on us...
health from unhygienic and other causes,the perpetual sickness, frequent periods
final
insufficient
the
the
result, and the very
uncertainty of
of
nourishment,
health
have
the
told
on
and spirits of the
unfavourably
unpleasantconditions, all
Europeansand our good blacks have worked in vain (one wonders?).909
The movement had been very good indeed as they had averaged nearly 18 miles
per day since 12 August with only two or three rest days. The long service Askaris and
porters from the Wanyamwezi tribe were capableof great exertions under pressure.
THE END OF THE CAMPAIGN

Now awareof the changedenemydirection, van Deventermade a number of

strategicdeploymentsto containit. He sentonebattalionto Bismarkburgat the southern
from
from
Ndanda
Lindi
to
Lake
Tanganyika,
where
endof
while two othersweremoved
they were sent by ship to Beira and then on by railway to Broken Hill in Southern
haste,
in
25
September,
Brandt
Rhodesia.Hawthorndespatched
to
all
arriving
on
another
910
but only in time to meetwith the rearguard.
Still expecting the Germans to turn north or west, van Deventer was surprised
when they continued south into the relatively foodless areasof Northern Rhodesia. Fife
was occupied only hours before the Germans arrived on I November; an attack the
following day was held off by the defenders. Unable to secure the vital stores there, the
Germans decided to make further south to Kasama which was the major supply depot
from the south. It was a rich sourceand was also the key to the crossing of the Zambesi
River and von Lettow's hopes for a continued campaign. The main body arrived in
Kasamaon 12 November, while K6hl's rearguard collected supplies back in Brandt. In
"9 WO 106/1460,SchneeDiary, I November 198,
pp. 81-82.
910Moyse-Bartlett, TheKing's Aftican Rifles, 420; Broken Hill was 500 miles south Fife
of
and Elizabethville was 400 miles to the
p.
west

318

the meantime, von Spangenberg'sadvanceguard was reconnoitering crossing sites on the

Zambesi.

Desperateto catch the Germansbefore they moved off again, Hawkins with the
KAP, left his supplies far behind and attacked the main body with only his 750 men on
12 November. A number of casualtieswere inflicted, but he could not move further until
his baggagetrain caught up with him. This was to be the last engagementof the war as
the armistice had been already signed in Europe although it was not until 13 November
that von Lettow received formal notification from van Deventer.

It was the endto an extremelyarduousand hard-foughtcampaignin which both

sides had marchedincredibledistancesin virgin bush and on very short rations with
almostno creaturecomforts. The forcethat surrendered
consistedof 20 Germanofficers,
6 medicalofficers,the Governorand32 officials, 3 subordinateofficials and 122German
NCOs, 1168Askaris, 1522 Carriers, 130 Prisonersof War Carriers,428 Agricultural
Carriers/Workers,282 "Boys" for the Europeans,427 Wives, and 392 "Boys" for the
Africans. An overall total of 155 Germansand 4416 Africans remainedtogetheruntil
911
the end.

911Boell, Die Operationen, 424.

p.

319

CHAPTER 9- INTER-ALLIED

COOPERATION

Despite the joint desire to defeat Gen-nany, the British and Belgians found
far
Britain
By
the
in
difficult
Africa
troublesome.
together
stronger
power,
working
and
had required Belgian assistancein taking the western tracts of German East Africa owing
to the distances involved and insufficient local forces. Anglo-Belgian military coin
Tabora
September did little to
during
1916
the
capture of
operation was uneasy
and
reduce tensions. One senior officer describedthe combined advanceon Tabora:
"... want of cordial co-operation between the two elements of the force. As a
described
better
be
fact
as the advance of
almost
the
matter of
movement would
two forces, to some extent in co-operation,and with a similar object."912

The campaignhadbeenmarredby personalityclashesbetweencommandersand

differing national goals. Practicalmilitary co-operationendedon a sour note in the
immediate
had
their
1916,
because
Belgians
political goals
the
achieved
autumnof
partly
and partly over a prolongeddisputeabout the division of the spoils. Above all, their
Governmentwantedthe completereturn of occupiedBelgium and by seizingsufficient
Germanterritory they had strengthened
their hand in any peacesettlement. They were
well awareof Germandesignson the Congo colony and the possessionof RwandaUrundi gave them additional security. Ultimately, they wished to add the captured
territoriesto their Congocolony, an approachwhich clashedwith the British desireto
leave such settlementsto the end of the War.913
Aware of British interests, they offered to hand the town over to British
administration on 8 September,but the Colonial Office did not reply until nearly two
914
later.
In the meantime, they established their own administration and remonths
exertedpolitical control over the African population. As time progressed,they made the
912Fendall, ne East AM= Force, 78.
p
913Louis, Ruanda Orundi,
pp. 208 and 216-217.
914FO 371/2856,Folio 7930,2 January 1917,"Belgian feeling over question of Tabora & other points". *Extract of a letter from Sit F.
Villiers to Mr. Balfour".

320

in
This
further
the
campaign.
their
was
this
participation
retention of
area a condition of
deemed unacceptableby the Colonial Office as it viewed the area as belonging to the
British sphereof influence. A memorandumstated:
"... They propose, as the price of their future military co-operation with General
Smuts, a recognition of their rights over the occupied territories in sensedescribed
above and acceptanceof the principle that Belgium should reap advantages,to be
defined hereafter,from her further participation in the campaign in proportion to the
915
"
measureof her future efforts.
Bonar Law was intransigent and his uncompromising attitude was reinforced by
Smuts's hostility towards Belgian ambitions. Given the disparity of power, the Colonial
Office eventually prevailed.
"It appeared to Mr. Bonar Law quite impossible to discuss further Belgian cothe
is
that
It
therefore
question of
presumed
terms
operation on any such
...
is
Belgians
the
that
further
the
settled,
and
with
seeking
military co-operation
Belgian Government will be informed that His Majesty's Government arc unable
to discuss the question of their co-operating further on the terms which they
916
"
propose.

However, it was at a cost and the Belgians felt very strongly about the rebuff-.
"In short it [giving up Tabora] would be a "humiliation" -I quote the word usedto
me severaltimes over by the Prime Minister ... In order to make the point more clear
I must explain that the King of the Belgians takes a direct interest in theseAfrican
is
line.
It
the
reported that Ministers
questionsand supports
adoption of a strong
917
"
havepositively declaredthat they will not give up Tabora.
915FO 371/2856,Folio 7930,19 January 1917,Colonial office Memorandum, [n.d.) enclosedwith letter Long to Balfour.
9'6FO 371/2856, Folio 7930,19 January 1917,Colonial Office Memorandum, [n.d.) enclosedwith letter Long to Balfour.
917FO 371/2856, Folio 7930,2 January 1917,"Belgian feeling over question of Tabora & other points", "Extract Of 2 letter from Sir F.
Villiers to Mr Balfour"; Albert ler, [Belgium], Les Carnets de Guerre, ed Thielcmans, Marie-Rose, Paris: Duculot, 1991. Entry I
November 1916,pp. 290-291 and 14 February 1917,p. 289.

321

The intervention of the Foreign Office soothed Belgian feelings to an extent, but
in the end they were forced to withdraw to the west. In the circumstances,the Belgians
irrevocable
breach
did
further
the
was
not
and
suspended
although
military operations,
not rule out future co-operation. Only days after the handover, the CIGS announcedthat
Belgian troops would indeed be available for operations without condition and General
Smuts acceptedthe offer.918 This was in fact an error, as the Belgian position had not
919
21
November.
However,
changed and the War Office withdrew the telegram on
Robertson did point out that at no time had the Belgians ever categorically refused further
assistance,and that they could probably be induced to help if necessary. In reply, Smuts
made little effort to retain their services, stating that he did not believe that further
Belgian assistancewould be needed,even if the campaign were to go into 1917, although
he would be happy to have a reservecolumn of 1,500rifles available at Tabora.920
Given the desperatestateof Smuts's own forces in this period, his was a very
tepid response;with his own expansionistambitions, he sought to discouragecompetition
during the post-war settlement. Little action of substancefollowed and by late December
GHQ was officially informed that all negotiationswith the Belgians had been dropped 921
.
The failure to use his allies in the pursuit of the Westtruppenwas soon apparentwith the
breakthroughand near overwhelming of Norforce in October and November. Thus, the
first phaseof Anglo-Belgian cooperationcame to an ungracious end in mid-November
1916, with both sidesseemingly more interestedin arguing over territory rather than
beating the common foe.922
However, by early 1917 the shortageof manpower led the British to re-opcn the
question of Belgian assistance. The first approach came on 17 January 1917, when the
British Governmentmade the Belgians aware of the possibility of asking for more troops.
9'aWO95/5292,WarDiaryGHQ, 19November
1916,EntryGHQ919
WO95/5292,WarDiaryGHQ,21November1916,EntryGHQ.
9" WO95/5292,WarDiaryGHQ,30November1916,Appendix42,TelegramOA 198,Smutsto CIGS,30 November,FO371/2856.
Folio 7930,2January1917,"Belgianfeelingoverquestionof Tabora& otherpoints","Extractof a letterfrom Sit F. Villiers to Mr
Balfour".
921
WO95/5292,WarDiaryGHQ,20 December1916,EntryGHQ.
922
WO 95/5292,WarDiaryGHQ, 10November1916,Appendix13,Telegram6150,BonarLawto Smuts,9 November.

322

A few weeks later, on 9 March 1917, the British formally

approach of At

notified their allies of the

Wintgens towards Bismarckburg and concluded by a request for 300

Belgian soldiers to take over its defence in order to allow British forces to concentrate to
contain the menace. This was easier said than done as the demobilisation of the Belgian
colonial forces had started in January 1917 and only four battalions remained in occupied
German East Africa.

The remainder of the force was either disbanding or in the process

of travelling to their permanent garrison locations, many of which were far in the interior
923
Congo.
of the

In the circumstances, the Belgian response was quite generous, as they not only
prepared to take over Bismarckburg, but also deployed several other units near Lake
Tanganyika
five battalions to concentrate
Kigoma.924

around

and ordered another

However, whatever the new found spirit of co-operation, it would take some time to
rebuild and redeploy forces back into the main theatre of operations.
Other factors were at play, not least of which was the attitude of General Hoskins
who recognisedthat additional transport was essential to his plans. The recruitment of
porters in Belgian territories offered a local solution and he requestedthis measurein his
draft plansto the CIGS.925GeneralNortheyhadalwaysseenthe use Belgiantroops

of

as
essentialand he pressedHoskins to bring them back, a view that was forwardedto
London.926Thesemeasures
days,
few
discussions
the
to
and
again
a
within
wereagreed
927
Belgian
Government.
wereresumedwith the
In mid-March, the British ambassadorformally asked for permission to use the
Belgian-controlled section of the Central Railway as well as lake steamers for the
movement of British forces in an emergency. Two weeks later, he placed a furthcr
requestbefore the Government,this time to recruit African soldiers for the British forces.
According to the Belgians, highly sensitive to the situation in their occupied home
9"CampagnesColoniales
CIGSto Smuts,10
Beiges,III, pp. 13-14;W095/5292,WarDiaryGHQ,10March19I7,TelegTam3075S,
March.
924
WO 95/5292,WarDiaryGHQ,16March1917,AppendixA25,Telegram10077,Huyghdto Hoskins.16March.
92-'WO95/5292,WarDiaryGHQ,I March1917,AppendixAt, TelegramG 337,Hoskinsto CIGS,I March.
9MWO 95/5292,WarDiaryGHQ,28 February1917,AppendixJ, TelegramNF 2410.Norforce Hoskins,27 February.
to
"7 WO 95/5292,WarDiaryGHQ,7 March1917,AppendixA4,Telegram
30514,CIGSto Hoskins,6 MarclL

323

territory, such recruiting clashed with the Hague Convention, whereby occupants of an
928
former
their
forced
be
take
to
against
country.
territory
arms
up
occupied
could not
They were very concernedabout the possibility of retaliation against their own people in
followed
little
on this contentious subject.
occupied-Belgium and
progress
late
first
1916.
The
had
British
and
since
most
substantially
modified
attitude
-The
important reasonwas the growing realisation at the War Office that the campaign was far
6orn over despite Smuts's utterances. The secondwas the sheer lack of resources,both
in terms of soldiers and of transport, with which to prosecutethe fighting. The third was
the strain imposed on Britain's shipping resources,now exacerbatedby the unrestricted
U-Boat campaign. Local Belgian participation, considered a needless encumbrance in
late 1916, now becamean urgent priority. Finally, inter-departmental rivalries in London
played their part.

Bonar Law, the Colonial Secretary, had carried out the earlier

be
described
in
as tactless, verging on the rude.
that
negotiations
can only
a manner
However, the Foreign Office differed both in manner and point of view; in early April,
the War Office sought its help:
"... The military situation in East Africa has, as Mr. Balfour is aware, becn
factors
have
Several
Council.
the
contributcd to
engagingthe anxious attention of
the prolongation of the campaign, notably (1) the rainy season;(2) the necessity
(3)
in
long
Service
by
the abnormally
the
debilitated
country;
of exchangingunits
high sick-rates amongst both white and Indian native troops; (4) the difficulty of
in
African
locally-raised
troops
the field in a
placing newly-recruited and
sufficient time; (5) the ineffectiveness,hitherto of the Portugueseco-operation.
Moreover, as regards projected future operations, the Council havc
experienced difficulty in making available the large quantity of mechanical
transport, and the great number of carriers, neededfor an offensive, whilst in view
Indian
the
the
the
strength
of
the
of
and Wcst
maintenance
of
shortageof shipping

m CampagnesColoniales Belges, III,

pp. 14-1S.

324

African troops representsa diversion of tonnage which it is most desirable to

release..."

929

The War Office highlighted its need for ftu-ther Belgian military co-operation
together with the urgency of the situation. Abt Wintgens was seriously disrupting its
930
for
Norforce
had
its
the offensive and
preparations
sapped
of much of
manpower.
Equally important, there were insufficient troops either to cover the gap in the IringaMahenge area or to clear it of enemy forces. The degree of anxiety may be measuredin
the fact that the Foreign Office arrangeda bilateral meeting in London for II April, being
931
days
by
Belgian
to
Government.
unwilling
wait even the two extra
proposed the
The conferencewas successfulas the two national governmentsagreed to resume

combinedoperations.The main two priorities were established,with the first being the
rounding up of Abt Wintgens,and the secondbeing part of the encirclementand
destructionof the Germanmainbodyin the south-eastof the colony. It wasprovisionally
agreedthat threeBelgiancolumns,totalling some2,000 rifles with supportingservices
and carriers,would be provided,althoughthe exact details were left to the military
commanderson the spot. Refinementof the numberswas then carried out by the two
between
Ujiji
headquarters
18 and 19 April. "'
Belgian
commanders-in-chief
the
at
at
After muchdiscussion,the force levelsweremodifiedto a total of 3,000soldiersdivided
into two mobile columns of 1,200 rifles, each supported by an operational reservc
battalion of 600 rifles, plus a further two battalions of 500 men eachto provide individual
933
for
replacements casualties. The final agreementlaid down the contribution as having a
maximum of 4,000 rifles and that the Belgian troops were not to be mixed with British
forces. Importantly, and with an eye to the future, the Belgian Government reservcd its
right to decide on any extension of the campaign, should it move into PortugueseEast
Africa.

929FO 371/2857,Folio 72473,7 April 1917,Letter Secretary,War Office to Under-Secretary State, Foreign Office.
of
WO 95/5293, War Diary GHQ, 3 April 1917,Appendix A25, Telegram 32025, CIGS to Hoskins, 2 April.
FO 371/2857,Folio 72473,8 April 1917,Telegram3 1, Foreign Office to Sir F. Villiers, Havre.
WO 9515293,War Diary GHQ, 14 April 1917,Appendix A104, Telegram 32670, CIGS to Hoskins, 14 April.
933CampagnesColoniales Belges, III,
pp. 29-30, Map 3.

325

For their part, the British promised to provide supplies and equipment for their
be
by
direction
Overall
the British
the
exercised
would
campaign
allies'. columns.
of

He
be
the
would set out both the
officer.
senior
commander-in-chiefwho would
but
for
the
contingent,
each
within those
objectives and
respective zones of operations
constraints, the Belgian commander-in-chief was free to execute his missions as he saw
fit. 934.

The military plan of operationswas divided into two distinct phasesthat reflected
both the exigenciesof the moment and the difficulties of remobilisation. The first phase
was the most pressing and involved despatching all the available troops in occupied
German East Africa, some 1,200 rifles, against 4bt Wintgens in co-operation with
Colonel Murray's column in the south. The aim of this force was to encircle and destroy
the raiders as quickly as possible. The second and subsequent phase, envisaged a
reinforced Belgian contingent of some 2,000 to 3,000 rifles working with General
Northey's troops in the west and south. Together, they would help to drive the Germans
into the general operational encirclement planned for June and jUly. 935 These
for
back
the
to
approval, with the
arrangementswere presented
national governments
Belgians signalling their assenton 26 April. Despite the irritations of previous disputes,
the Belgians were back in the campaign.
The changeof command between GeneralsHoskins and van Deventer at the end
of May 1917 had little immediate effect on Anglo-Belgian co-operation. The gcncral
strategy devised by Hoskins was little changed and the Belgians continued to build up
their forces in preparation for the planned drive on Mahenge.936 However, before the
main operational objectives could be reached,there were several preliminary operations
that had to be completed beforehand. In the prospective Belgian sector to the west, van
Deventer wanted them to clear the Kilombero valley while also advancing from Iringa
towards Ifakara. This move was aimed at inducing the Germansto weaken the line along
934
WO 95/5293,WarDiaryGHQ,21 April 1917,AppendixA132,TelegramP 16,Hoskinsto CIGS,19April, Campagnes
Coloniaks
Beiges,III, pp.33-35.
'"5 Campagnes
de la BrigadeNM4 ler BureauNo 5811dated25
ColonialesBeiges,111,pp. 294-295.instructionau commandant
avril 17.
m CamPagnes
ColonialesBelges,III, pp. 125-127;WO 95/5293,WarDiaryGHQ,27 April 1917,AppendixB21.TeicgmmalO,
Hoskinsto CIGS,27 April; 10June1917,AppendixA74,TelegramG 843,vanDeventerto CIGS,10June.

326

the Ruaha River, notably at Kidatu, while covering the move forward of the Belgian main
body from Kilossa. Eager to get underway, van Deventer pressed Huyghd to provide a
column of 1,200 rifles to clear Kilombero valley by mid-June, but it could not be readied
until the end of July. However, a reduced bid for 500 men to fill a gap around Mpanga,
south-east of Iringa was agreed, despite breaking the principles of a concentrated Belgian
deployment. 937

Abt Naumann occupied the Belgian forces for much of the period May-August
1917 as they chasedthe ever diminishing raiders through much of the northern part of the
German East Africa colony. Despite several checks, they continued the follow up
until
September,when the British resumedfull control of the operations. Henceforth, the main

effort focusedon the encirclementanddestructionof the main body of the Schulztruppe

in the south-westem
comerof the colony.
For the remainderof the campaign,the biggest irritations between the two nations
were the Belgian inability to supply its own forces and its demand for a separateand
distinct sphereof operations. The British provided a great deal of MT and other transport
assistance,but as the area of operations constricted, the mixing of British and Belgian
units becameincreasingly hard to avoid."' By late 1917 as the RO was operating out of
Kilwa, Huyghd came under heavy pressure for allowing "mixed" columns and van
Deventer sought assistancein resolving the matter. The Foreign Office put pressureon
the Belgian Government to relax its rule on the grounds that it would delay the
939
Finally, it was agreed to allow some intermingling of
completion of the campaign.
units and the remaining Belgian units were able to play their part in the clearance of
German East Africa.

937
Compagnes
20
June
ColonialesBeiges,111,
GHQ,
1917,
War
Diary
95/5293,
AppendixAt 16,Telegram0 976.
WO
126-128;
pp.
BGGSto NortheyandTytler,20 June.
"s WO 95/5293,WarDiaryGHQ,5 August1917,EntryIringa Column,13August1917,AppendixAt 0, Telegram22011
luyghd
-1,1
to vanDeventer,
6 August;27 August1917,AppendixA17,LettervanDeventerto Huyghd;9 September
1917,AppendixA 11.Pricis
Huyghdto Sheppard.
939
WO 95/5293,WarDiaryGHQ,6 November1917,AppendixB, Telegram26 WO,vanDeventerto CIGS,6 November.;WO
33/953,Telegrams
D 2. No. 2271,7November1917,TelegramX 900,vanDeventerto CIGS,p. 225;No. 2272,8November1917.
Telegram44997,CIGSto vanDeventer,p. 225;No. 2291,17November1917,Telegram45875,CIGSto
vanDeventer.p. 232.

327

in
The
December
1917
Anglo-Belgian
came
end
co-operation
of
military
with the
,,
capture of the Wesuruppen. The Belgian Government had reserved its right to consider
continuing the fight in Portugueseterritory while the British now saw the campaign as
largely, over. Faced with the difficulties of working with the Portuguese, it saw the
negotiations for a continuedBelgian presenceas being too time consuming, in terms of both
940
decision
necessaryreorganisationandpolitical
making. In the end, the two nations cnded
on a much more amicablenote than the year previously.
ANGLO-PORTUGUESE

COOPERATION

Despite its age and duration, the Anglo-Portuguese alliance was never evenly
balanced and relations during the war were seldom easy. The British believed the
Portugueseto be incapable of serious military effort and put a great deal of Pressureon
them to remain neutral from the outbreak of the war.941 However, as shipping grew
scarcer, the British attitude changed and by early 1916 they were encouraging the
Portugueseto seize Germanvesselsin their territorial waters.942They realised that such a
course would probably lead to war between the two countries, having previously tried to
discouragePortuguesecolonial aspirations:
".

do not want Portugalto establishtoo great a claim on our gratitude or to be

--We

under obligation to protect their Colonies or divide up German territory VAth

them

In view of the above we have discouraged the Portuguese from assisting us

...

either in South West Africa or East Africa in spite of repeated offers on their

pad... 99943
The British had long suspected that German mail and contraband had bccn
passing clandestinely through PortugueseEast Africa and that many local merchants
940
WO 33/953,Telegrams
D 2, No. 2317,27November1917,Telegram299,vanDeventerto CIGS,p. 240;No. 2325,29November
1917,Telegram46804,CIGSto vanDeventer,p. 243;WO 95/5294,WarDiaryGHQ,3 December1917,AppendixA4. Telegram
47022,CIGSto vanDeventer,2 December.
"' Vincent-Smith,
JohnD, "Britain,PortugalandtheFirstWorldWar,1914-16",EuropeanStudiesReview,IV, No 3. (1974),pp.
210-211.
942
Stone,GlynA, "TheOfficial BritishAttitudeto theAnglo-Portuguese
Alliance,1910-1945".TheJournalof Contemporary
History,X. No 4, (October1975),p. 732.

328

were engaged in cross-bordertrade. The opening of hostilities would probably be the

onlyway in which such exchangescould be stopped, although it would take a major
944
military and naval effort to achieve succesS. On the other hand, as long as Portugal
remained neutral, it would have been politically difficult for the Schutztruppe to move
south of the Rovuma especiallywithout authority from the Imperial Government.
by
dismal failure of the
the
the
seriousness
was
underlined
of
situation
. --Tbe
Portuguese expeditionary forces sent out in 1915 and 1916. Both attempts to seize
German territory had resulted in humiliating reverses and large losses in troops in
equipment. National pride was further affronted by the need to ask for British assistance,
particularly as General Smuts, a man well known for his appetite for the Portuguese
colony, commandedthe forces:
"General Smuts, though not extremely popular in Government circles in
PortugueseEast Africa owing, I am given to understand,to some severecriticisms
he has had occasionto expressin regard to PortugueseEast Africa... " 945

'' .C

On the other hand,the weak and highly decentralisedrule in PortugueseEast

Africa gave rise to a very different situation in Anglo-Portuguese relations. Portuguese

administration was much less developedthan that of other nations and the fiat of the state
946
did
far
beyond
often
the various governorships. Furthermore, the north of
not extend
the colony was under charteredcompanyrule and was virtually unexplored and was quite
undeveloped. As well, the severity and rapacity of colonial rule had led to widespread
opposition by the African population that would shortly lead to open, armed rebellion.
By the middle of 1916, relations between the British and Portugueseforces were
far from cordial. Attempts to improve the situation were unsuccessful,as an attempt to
attach a British political officer to the portugueseheadquarterswas abruptly rebuffed in
"3 WO 106/308,"Memorandum theSituation
whichmayariseastheresultof Portugalcomplyingwith therequestmadeby theBritish
on
", [n.d.] [likelyearly19161,
thatGermanshippingin theTagusshouldbecommandeered.
government
p. 2.
9" WO 106/308."Memorandum theSituation
on
whichmayariseastheresultof Portugalcomplyingwith therequestmadeby theBritish
[nA] [likelyearly19161,
thatGermanshippingin theTagusshouldbecommandeereV.
vernment
pp.3-4.
FO371/2857,Folio55352,26January1917,LetterConsulGeneral(ErrolMacdonell),LourencoMarquesto ForeignSecretary.
"6 Newitt,Malyn,A History
London:Hurst& Company,1995,pp.367-368,
ofMozambique,

329

947
November.
This attemptto supervisePortugueseadministration naturally caused
early
information.
For
their part, it added
of
offence and prevented any meaningful exchange
to the British senseof distrust of their ally's intentions. The feeling was mutual for the
had
inflamed
Portuguese
Anglo-German
the
to
colonies
prewar
negotiations carve up
formers'
had
Portuguese
the
the
rivalry to some advantagesince
passions, although
used
948
Many local officials were opposed to
the Anglo-Boer War at the turn of the century.
supporting Britain while also wishing to maintain their exploitation of the local
population. Furthermore,neutrality permitted an extensive and lucrative, if clandestine,
trade with Gennan East Africa that would be endedby war.
Attempts were made to work together. With the recall of General Gil in 1916,
and despite the planned sending of another expeditionary force in early 1917, the
Governor General, Dr Alvaro de Castro, assumed the role of commander-in-chief.
Although a civilian, he had served in the Army, reaching the rank of captain. fie was
conscious of the weaknessof his position and asked for a meeting with General Smuts in
late 1916. However, this was not possible for a number of reasonsand it was not until 25
January,two days after he had relinquished command in East Africa, that Smuts met with
de, Castro in Portugueseterritory. By rights the new commander-in-chief should have

conductedthe meeting,but it seemshighly likely that Hoskinswastoo busywith his new

j
the
to
the
ourney.
commandandconducting offensive make
It is notcworthy that Gcneral van Deventer, also on his way home, attended the
meeting as did the British Consul General from Lourenco Marques. During the
conference, Smuts pressed for the Portuguese to clear a line 50 miles south of the
Rovuma of all food and to prevent the inhabitants from sowing any crops in January or
February. This reflected his continuedbelief in a German surrender:

"... the General'sargumentbeing that if the enemyare awarethat there are not
food suppliesto be obtainedwithin easyreachof the Portuguesefrontier, they
947WO 95/5292, War Diary GHQ, 7 November 1916,Appendix 8, Telegram OA 968, Genstaff to CIGS, London, 7 November.
"s Newitt, Malyn, Portugal in, 4frica, London: Hurst & Company, 1981,
pp. 33-35; Willequct, Jacques,"Anglo-German Rivalry in
Belgian and PortugueseAfrica? ", Giffard and Louis, Britain and Germany,p. 265.

330

invaded
having
Portuguese
British
troops
the
to
without
will, perforce, surrender
territory...

049

Smutswent on to advocatethat the Rovurnabe held only lightly by a line of

heavy
be
to
withstand
strong enough
observation posts and that any garrisons must
German attacks. Furthermore, he emphasisedboth the quality of the enemy troops as
his
For
de
for
units.
training
raised
part,
newly
the
thorough
of
any
the
well as
need
Castro announcedthat he was in the processof raising the troops and carriers necessary
to make a simultaneousadvancewith the British at the beginning of the dry season.
Overall, the meeting was adjudged a successwith Smuts performing very well
impressed
him
Portuguese
The
diplomacy.
with
and the
were
and showing considerable
Governor General pronounced himself convinced of the need to work closely with the
British in the future. Little concrete action appearsto have been agreed, although one

tangible outcomewas the agreementto swap liaison officers in the near future and
950
The
between
meetingalso set
two
the
commanders-in-chief.
regular communications
the scenefor much closer co-operationunder GeneralHoskins.
If the renewed Belgian assistancewas welcomed, General Northey faced further
from
border.
being
Apart
African
East
in
Nyasaland-Portuguese
problems the south on the
deal
to
troops
the
he
trained
by
Wintgens,
with
the
needed
also
weakened
pursuit of
developing Germanthreat posedthere by AN von Stuemer. Although the Portuguesewere
held in generallylow regard,GeneralHoskins did try to involve them in support of both his
head
lack
Northey's
the
the
military
officer
at
a
professional
of
of
and
operationsalthough
the forceshinderedtheir military planning. Whatever his qualities, the Governor General
1
95
Portuguese
For
the
their
direct
part,
the
remained
to
was not
operations.
man
forces
into
to
their
intentions
their
British
allow
reluctant
very
were
suspicious of
and
colony.
949F0371/2857, Folio 55352,26 January 1917,Letter 227/17/52, No. 13, Macdonell to Foreign Secretary.pp. 2-3. Ilislater
incorporatesthe minutes of the meeting held on 25 Januaryas well as somecommentsby the Consul General.
950FO 371/2857,Folio 55352,26 January 1917,Letter 227/17/52, No. 13, Macdonell to Foreign Secretary,pp. 4-5.
"' WO 158/477,17 April [May] 1917, Letter No 5. Macdonell to Foreign Secretary and Hoskins, covering Despatch dated 16 May
1917.

331

As the campaignmoved increasingly to the south of the German colony in 1917,

the possibility of having to continue operations in Portugueseterritory became more and
As the British held no faith in the ability of the Portuguese to defend
themselves, diplomatic overtures to the Portuguese Government aimed to determine a

more likely.

joint strategy in the event of invasion. Matters were improved by the appointment of the
civilian consul general, Major Macdonell, as military liaison officer to the Portuguese
Headquarters in mid-March 1917.952However, such an apparently straightforward and
logical post was not viewed as such by the incumbent:
"The Portugueseare extraordinarily proud and are convinced that in military and all
They
nation.
are adverseto any
they
other
other matters
are as experiencedas any
sort of extraneousinterference,and on the slightest provocation they consider that
the "brio nacionar' (national honour and prestige)has been offended. A very large
fact
fully
have
to
the
that
they
Portuguese
Officers
alive
number of
up
and men are
to the presentmadea hopelessfiasco of the GermanEast campaign,and though this
has been statedto me in private, if any foreigner were to make a similar statement

he would incur the obloquyof the Military, Pressand public I shall be looked
...
it
be
by
and
the
will
with the utmost
upon with suspicion
military authorities
ingenuitythat I shallbe ableto
difficulty andonly with the exerciseof considerable
obtainexactinfori-nationor be cognizantof the conditions,movements,an-nament,
supplies &c. of the PortugueseExpeditionary Force. It must always be born in
mind that the Portuguesehate and suspectall foreigners, and I regret to say that in
my opinion they hate the British most of all becausethey fear them, and also there
are many who still rememberthe British Ultimatum of 1891... 9053

The appointmentwas shortly reciprocatedin early April with a Portuguese

professionalofficer, Major Azumbuju Martins, being sent to join British GHQ as a
liaison officer. 954He had someexperienceof the conditionsin EastAfrica and German
932WO 158/477,19 March 1917,Telegram 617, Macdonell to Hoskins, 18 March 1917.
933WO 158/477,20 March 1917,Letter 634/17/127,Macdonell to Hoskins, 20 March 1917.
'54WO 158/477,7 April 1917,Telegram 778, Macdonell to Hoskins, 6 April 1917.

332

ill-fated
Staff
to
the
Chief
been
having
expedition under
the
of
capabilities,
until recently
General Gil. Such appointmentswere the vital precursors to effective cooperation; the
first task for Major Macdonell was to ascertainPortuguesedeployments and to find out
955
boundary.
River
Rovuma
defend
the
intended
they
to
try
and
whether
even
Finally, a meeting betweenGeneralHoskins and Dr de Castro was held at Dar-esSalaam on 9 April 1917. To the former's dismay, de Castro proposed to send a force of
into
River
Zambezi
the
Chinde,
700
from
thence
southern
up
the coast to
over
rifles
Nyasaland and into Lake Nyasa where it would land at Mtengula which was believed to
be von Stuemer's objective.

He suggested that it would take place between 15 and 20

May, but this was hopelessly optimistic given Portuguese transport facilities. Although
force
likely
the
to
diplomatically,
it
was
as
suggestion
unwelcome
received
most
a
was
From
British
the
in
Northey's
problems.
get,
way and would cause major supply
blocking
Portuguese
for
to
it
be
far
the
a
strong
establish
perspective,
would
more useful
force along the Rovuma as well as protecting the vulnerable line of communication
tactfully
his
Hoskins
as possible, and,
Zambesi
as
River.
suggestions
the
put
running up
956

left
he
de
the matter unresolved.
Castro's
to
unable change
mind,

The conferencewas successfulin resolving several other issues;the first was that

the British should deal with the Germanthreat to Mwembe and secondthat their
Intelligence Scouts should be allowed to operate freely across the border, receiving full
issues
discussions
thorny
These
from
Portuguese
several
resolved
the
support
authorities.
South
African's
highly
the
Governor
General
of
the
view
positive
and gave
a
intentions. 957 Further British pressure was exerted through their growing intelligence
between
Major
3
July,
in
On
Portuguese
was
reached
territory.
agreement
service
Macdonell and the Portuguese Chief of Staff that all intelligence and interrogation of
prisoners would be conducted by British troops who would also make their dispositions
958
fit.
they
as
saw
955WO 158/477,4 April 1917,Letter Sheppardto Macdonell, 4 April 1917.
936CAB 44/9, pp. 39-40.
"7 WO 158/477,17 April [May] 1917,Letter No 5, Macdonell to Foreign Secretaryand Hoskins, covering Despatchdated 16 May
1917.
9""WO 1581477.9August 1917,"Precis Interview with the Chief of Staff PortugueseExpeditionary Force", Macdonell to van
of
Deventer, 3 July 1917.

333

Portuguese
the
However,
imminent
in
new
commander-inof
the
arrival
view of
,,::,
be
discussed
delayed
the
Colonel
discussions
Sousa
Rosa,
matter
could
until
were
chief,
in detail. His coming did little to inspire confidence, as the liaison officer reported to
GHQ:

"This appointment has causedthe greatest disatisfaction [sic] among officers of

Officers
in
is
Colonel
the
It
the
unpopular
the
that
most
one
of
all arms. appears
Army and has the utmost difficulty in getting Officers to serve under him. He is a
into
his military duties... "959
introduces
Republican
politics
very pronounced
and
The likelihood of a Germanre-entry into PortugueseEast Africa rendereddetailed
instructed
Macdonell
to:
was
than
and
cooperation even more vital
previously

"Try to find out quietly whetherthe direct assistanceof British or Belgiantroops

in the field in P.E.A. would be welcomedor not. It looks as if the Germanswere
it
ROVUMA
MOCIMBOA
E.
into
P.
A.
break
and
may
at or near
going to try to
be essentialto close this door flnnly.,, 960
However, the prospect of any British intervention was quickly quashed, and, at
961
from
Mwembe.
At a
Portugueseinsistence,GeneralNorthey's troops were withdrawn
local level and in order to improve relations, British GHQ came up with the idea of
Portuguese
Viala,
French
to
the
Colonel
commander-insending a
on a mission
officer,
chief. He had been attached to the East African Force for some time as an official
his
The
British.
disposed
the
aim
of
mission was to find
towards
observer and was well
out the detailed plans and intentions of the Portugueseas well the extent of their transport
direct
British
the
Importantly,
he
to
support
whether
of
problems.
ascertain
or
also
was
Belgian troops would be acceptableto Colonel Sousa Rosa. On the surface, it was a

"9 WO 158/477,8 August 1917,Letter No. 10, Macdonell to van Deventer, 14 July 1917, p.9.
"0 WO 158/478,10 September1917,TelegramG190, BGGS to Macdonell. 10 September1917.
"' WO 1581478,5October 1917,Telegram G 450,
van Deventerto Macdonell, 5 October; WO 95/5294, War Diary GHQ, 30
September1917,Appendix A27. Telegram 83, Macdonellto van Deventer,29 September.

334

highly unusual method of liaison, but the state of Anglo-Portuguese relations was poor
Sheppard:
in
General
Brigadier
the
words of
and
"... The Portugueseare probably less suspiciousof the French than ourselves, and
I think VIALA's visit will have excellent results, for he is a first class fellow, and very
,, 962

keen on helping on the campaignin any way possible.

Some three weeks later, the initiative had proved to be a success and Colonel
Sousa - Rosa's attitude had shifted favourably. Colonel Viala, together with the
Scouts
Intelligence
British
drive
the
operating south of the
of
professionalism and
Rovuma, had made a strong impression on the Portugueseabout the competenceof their
963
lead
to
three-pronged
Rosa
Colonel
Even
a
proposed
now
allies.
more radically,
Portuguese
Newala,
the
into
German
Africa
East
of
a
previous
scene
towards
advance
defeat. General Northey was particularly against any such move, having worked with
their forces for nearly two years, while the reports of the British liaison officer also
964
highly
Deventer
General
high
In
was
van
the
alarmed
a private signal,
command.
pessimistic about the chancesof Portuguesesuccess:
"... I think that the Portuguese proposals if carried out will be fatal. The
Portuguesetroops well entrenchedon the line of the ROVUMA may possibly be
but
the
between
the
left
(2)
them
enemy
part
of
to
on
the
able prevent a thrust on
in the open field they have no chancewhatever against the Germans. Everyone
be
for
I
therefore
knows
this
grateful
most
them
shall
any
view
who
agreeswith
...
Portuguese
The
in
this
are most touchy and 1
matter.
assistanceyou can give
965
,,
cannotforcemy opinionsfor fearof giving offence.

Mattershadnot beenhelpedby the tactlessness

of Major Martins,the Portuguese
liaison officer at GHQ,who wasthoughtto be ill disposedtowardsColonelSousaRosa:
%2wo 158/469,24 September1917,TelegramG 345, Sheppardto Macdonell
963WO 158/469,18 October 1917,LetterNo 19, Macdonell to van Deventer.
%4 WO 158/478,6 October 1917, Telegram NF 4662, Norforce to Genstaff, 5 October', 8 October 1917, Telegram 390 S. van
Deventer to Genstaft for onward transmissionto the Portuguese,7 October.
965WO 158/478,8 October 1917,Telegram 391 S, van Deventerto CIGS, 7 October.

335

"... I was also infonned that the question of British and Belgian cooperation had
been telegraphed to the Commander in Chief by Major Azambuja Martins and in
such a peremptory form that Colonel Souza Rosa was offended and unwilling
966
,,
the

consider

to

matter...

Fortunately, the efforts of Colonel Viala were also successful in presenting the
offer of assistancein much more diplomatic terms, thereby neutralising the hostility of
his own liaison officer.

If mattersat the highestlevelswere graduallybeing smoothedout, the situation

on the groundgavethe British ampleworries. Reportsaboutthe weaknessof defensive
General
Deventer
low
leadership
lack
van
on a
reached
preparations,
of
morale, and
basis,
Colonel
do
Rosa to do
but
little
he
to
there
exhort
regular
except
could
was
9"
more. Therewerevery greatlimits on suchleverage,and it would take considerably
more to overcomelong-termand structuraldefectsin the Portuguesecolonial military
structures.
Interestingly,the possibility of introducing the Belgians into PortugueseEast Africa
was quietly explored by the British in September1917. Acting on a suggestionof Colonel
Huyghd, a low key reconnaissanceof the harbour and facilities of Mocimboa da Praia was
carried out by the naval authorities and without the knowledge of either of the other
968
However, for reasonsthat remain unclear, this project never went
prospective parties.
beyond the planning stage.
The balance of power shifted decisively in early 1918, after the disaster at
Ngomano. Then the Portugueseauthorities, now in a stateof great concern, consentedto
the main British force operating in their territory-119 The need to build up a new base at
"6 WO 158/469, Paper 16, Letter No 20, Macdonell to
van Deventer, 18 October 1917.
967WO 158/478,29 August 1917,Appendix to Letter No 11, Macdonell to van Deventer, 14 August.
9"W095/5924, War Diary GHQ, 10 September1917,Appendix A21, BGGS to GSOI, "Establishing Belgian Force Mocimboa
a
at
da Praia7.
969WO 158/478,1 December 1917,Telegram46895, CIGS to
van Deventer,30 November.

336

Port Amelia in late December made for'the first major clash between methods; within a
few days, the British found it difficult to recruit labour as the local Nyassa Company
in
led
British
Ambassador
help.
This
to
the
to
officials were obstructive and unwilling
Lisbon being fonnally instructed to seek:

"... Entire control of everything connectedwith movementsof troops and storesat

Port Amelia, including hiring and managementof dhows. Control, by arrangement
lines
Authorities
Base
local
Portuguese
the
of
and
of communication
with
inland It is important that Portuguese Government should agree that actual
...
be
done
by
&c.
British
Authorities,
of
carriers
should
engagementand payment
070
inefficiency
danger
having
and
of Portuguesemethods...
shown
experience
These wide-ranging demands were accompaniedby private information for the
local
Portuguese
Deventer
General
the
that
suspected
military authorities
van
ambassador
of colluding with the Germans.

This wasfollowedthe very next day by a demandfor furthercontrol:now

the War Office was asking for the Portugueseto place their military forces under General
van Deventer's command. Again the languagewas notable for its directnessand lack of
diplomacy:

"... The course of events since the crossing of the frontier by Major-General Von
Lettow with a small and exhausted force about November 27h last has shown that
the Portuguese command is incapable of conducting military

operations, or is

unwilling to offer effective opposition to the Germans. Detachments have been

left in advanced and isolated posts, despite the representations of the British
Headquarters, and have surrendered after a feeble resistance with their arms
ammunition and stores of supplies, on which the German troops have mainly
9MI
subsisted...
F0371/3128, Folio 655,2 January 1918.TelegramNo. 3, Foreign Office to Sir LCamegie, Lisbon.
FO 371,3128, Folio 2132,3 January 1918,Letter 0165/9270 (MO.2), DMO to Under-Sccretaryof State, Foreign Office, 3 January.
This move had beenrequestedin F0371/3128, Folio 2132,2 January 1918,Telegram Lb., van Deventerto CIGS, I January.

337

in
The pressureand a concurrent changeof government Lisbon soon achievedthe
it
joint
that
was under the
to
desiredresults:the Portugueseacceded a
commandprovided
972
However,
formed.
headquarters
"mixed"
was
highest ranking officer and that a
home
the
the
from
far
assurances of
the surface, whatever
tensions were never
felt
Deventer
1918
in
late
June
head
obliged
when
van
Matters
a
reached
governments.
from
forces
British
the
local
that
formally to reassurethe
withdraw
would
administration
Cabinet
After
been
had
obtaining
successfully concluded.
colony once operations
Britain's
to
Portuguese
informed
he
the
past
adherence
for
of
then
approval the statement,
973
However, even explicit assuranceswere not enough and
treaties and agreements.
differences continued to rankle on both sides. Only a few days later, General van
Deventer, fed up with the continuing Portuguesepractice of leaving small detachments
defences,
sent
a
or
proper
support
adequate
the
without
throughout
country
scattered
his
Instead
to
Office.
War
such
power,
adding
military
of
the
to
critique
withering
for
Lettow,
food,
and
von
and
ammunition
weapons
of
supply
a
as
outpostsacted merely
in
battle.
the
gained
successes
neutralisedmany of
Equally exasperatedfrom constantly pressing Colonel Rosa to draw up a
little,
defence,
Deventer
van
now
and
achieving
of
more concentratedand workable plan
forces
Portuguese
he
to
the
to
drastic
relegate
all
of
military
proposed
sought
measures:
be
to
the
were
civil administration
rear areasor coastal protection, while all elementsof
For
the
to
their
from
sake of
obstructiveness.
owing
the
operations
of
area
withdrawn
Governor
Rosa's
the
Colonel
he
civilian
of
efforts
as
well
as
praised
good relations,
Nyassaland although the Governor of Mozambique was directly accused of being
obstructive and unhelpful to the military effort.
"... I think time has come when the Portugueseauthorities must be told the
truth, namely that their troops in PortugueseEast Africa are totally unreliable and
both
European
African
The
danger
to
their
and
personnel
allies.
a sourceof grave
is of the poorest possible quality, and the natives of PortugueseEast Africa detest
FO 371/3128,Folio 5578,10 January 1918,TelegramNo. 16 (D), Sir L. Carnegie,Lisbon to Foreign Office, 9 January.

338

Portugueseto such an extent that when we act in conjunction with Portuguese

from
When
from
help
them.
alone
removed
acting
troops we can get no
Portuguesesphereof influence natives help us freely...

,, 974

The British Ambassador was instructed to deliver this highly undiplomatic

in
East
Africa
the
desire
that
the
to
it
the
campaign
conclude
message; was a measure of
Government
Portuguese
For
drastic
the
terms.
good measure,
government assentedto such
incapable
be
informed
their
were
of ruling effectively and
that
to
authorities
colonial
was
975
were opposed by their Affican subjeCtS.

By July, operationshad moved inland and away from the main concentrationsof
the Portugueseforces. As van Deventersenthis troops in hot pursuit of the Germansback
towards GermanEast Africa, the need for detailed assistancefaded, no doubt to his relief.
Interestingly, at the end of the war he felt it necessaryto draft a supplementaryand secret
despatchthat describedhis true thoughts about the Portuguesein the strongestpossible
language.For example:
".. Through the operations of the early months of 1918,1 managed to keep the
Portuguese troops out of direct contact with the enemy without, I think, hurting
their feelings

In May, Colonel Rosa pressed me to give his troops a more active

...
from
LISBON
Ile
believe,
to
have
I
to
urgent
representations
reason
role; owing,
...
076
disastrous...

resultswere

After cataloguing the inadequaciesof his ally's force, van Deventer summed up the
relationshipas follows:
"In fact, the Germans simply looked on the Portuguese forces or posts as
dumps.
The
Supply
Ordnance
effect of theseeasy conquestson the
convenient
and
' FO371/3128,Folio 112227,27June1918,Letter112227/W/IForeignOfficeto Secretary,
Army Council,27 June.
974
FO371/3128,Folio 115649.26June1918,TelegramG 962,vanDeventerto WarOffice,London,25 June.
"5 FO 371/3128,Folio 115649,29June1918,Letter0148/694(MO.2), DD Cubittto Under-Secretary
of State,ForeignOffice,29
June.
976
WO 158/475,1October1918,LettervanDeventerto Secretary
of Statefor War,p. 3.

339

moraleof the enemy'stroops was very great; and it can perhapsbe conceivedwhat
an incrediblehandicapthe Portugueseforces have beento me, ever sincethe enemy
,, 977

crossedthe ROVUMA in November 1917.

It was fortunatefor diplomacyand good relationsthat this documentremainedsecret.

977WO 158/475,1 October 1918,Letter

van Deventerto Secretaryof State for War, p. 5.

340

CHAPTER 10 - CONCLUSIONS
Four major points emergefrom the Iiistory of the EastAffican campaign. The first
difficulty
In
terrain.
the
and
many ways,
of
conditions,
most
notably
climatic
was
sheer
fighting was the leastof all problemsas any force was undermore threat from tropical
disease,starvationor dying of thirst than being killed in battle. When assessingthe outcome
of the campaigndue considerationmust be given for numerousand severeobstaclesfaced
by all combatants.It was a story of the utmost determinationand effort to fight in one of the
most difficult areason earth.
The second factor was the absolute requirement for a well organised system of
transport and supply. This as much as anything else determinedmany of the tactical and
both
Without
decisions
sides.
an understandingof this critical limitation,
operational
on
the course of the campaign is very difficult to understand. All suffered severely from
insufficient rations and equipment and the difficulty of the physical conditions
exacerbatedany shortages.
The third point was the immense damagecausedby tropical diseases. Virtually
no humans escaped the ravages of malaria, dysentery, typhoid or pneumonia while
animals were wiped out by trypanosiamis and horse sickness. Despite the advancesin
medicine, the combination of extremely difficult conditions, poor transportation and lack
of discipline meant that the forces suffered extremely heavily throughout. The high
levels of sicknesswere to be the outstandingfactor in the campaign.
The fourth is the quality of the generalshipwhich has been the subject of many
inaccurate generalisationsand assessments.On the British side, GeneralSmuts has been
presentedas the great general who achieved a huge and relatively inexpensive success,
capturing much of German East Affica. There is no doubt, that from his arrival, Smuts
supplieda style of dynamic leadershipthat was badly neededin East Africa. He instilled a
senseof drive and purposethat inspired both the staff and regimentalsoldier of his polyglot
army. He certainly achievedwhat he set out to do; the acquisitionof largetracts of territory
341

forces
had
in
While
his
battle
taken the capital, all the
the
cost
casualties.
minimum
with
farmland,
defeated
both
best
the
they
the
railways and some of
never
major ports,
Schuatruppein battle. Smutsleft the theatrewith his opponents'army possessingboth a
high degreeof combat effectivenessand good morale. More seriously,his claims that the
by
his
successors'needto conducta
over
were
subsequently
rebutted
campaignwas nearly
difficult and arduouscampaignfor the following year and a half. Furthermore,if he had
failed to destroythe enemy's forces,his techniqueshad largely incapacitatedhis own army;
P
ill-health
12,000
had
be
to repatriatedthrough
over
with the unfortunate Division lasting
only four monthsin the field.
There were two seriouscriticisms that can be levelled against Smuts' generalship;
the first was his reluctance to focus on the enemy's forces and the second was the dire
statein which he left his army. Regarding the first, he admitted that he was interestedin
manoeuvrenot fighting. His political concerns,both in avoiding casualtiesand capturing
ground, led him to set unrealistic goalsthat either exceededthe capabilities of his force or
left it seriously under-resourced. This sense of haste and desire to capture ground,
coupled with his inexperienceof staff work, led him to ignore the vital link betweenhis
Administrative and General Staffs. This failure to understandthat, without a supply and
transport system of sufficient capacity, lift and range, grand operational aims would
inevitably founder resulted in the troops suffering immense and, in many cases,
preventablehardships. It was not simply, as he put it, "efforts like these cannot be made
978
inflicting
but a question as to whether better
hardships
the greatest
without
on all",
planning could have alleviated the worst.
As to the second criticism,. it is undeniable that there were serious shortcomings in
'
the provisioning, feeding and medical care of the force during his tenure in command,
although he preferred to refer to the obstacles of nature rather than the inadequacies of
planning.

Two major enquiries found serious faults with the organisation of his forces

and the results were plain to see. It was only his political connections and influence that
prevented either report from becoming publicly available.
" Crowe, General Sinuts'Campaign, Foreword by JC Smuts,
pp. xii-xiii.

342

The final phaseof General Smuts's command endedin sombremisery. Although

he proclaimed the successof his campaign to the outside world and was initially taken at
his word, the reality was rather different. If his army had been in a bad way by
September1916, it was in a shocking state by the time of his departurein January 1917.
He had driven his force hard, but without considerationfor the essentialfactors of supply,
transport and health. Now, it was largely ineffective and it would fall to others to restore
its effectiveness.
He was succeededby General Hoskins, a regular soldier of wide East African
experience. Hoskins had no opportunity to influence the final unsuccessfuloffensive of
he
in
forced
him
halt
himself
The
found
he
to
and
command.
rains
which
unexpectedly
by
left
He
Smuts.
his
to
the
trying
tenure
shattered
army
the
rebuild
spent
remainder of
haphazard
did
inspire
his
but
requests
not
confidence
planning and
was a capableofficer,
in the War Office and ultimately resulted in his downfall. However, he immediately set
his
Norforce
his
force
as
principal offensive
while maintaining
about refurbishing
formation. The Wintgens raid did not help his reputation in London, but, given the
have
done
it.
Pursuit of a small,
little
to
there
prevent
anybody could
situation,
was
be
East
Africa
in
bush
force
to
the
was
never
going
easy and Hoskins was
of
ruthless
happened.
finally
to
the
than
round
raiders
up
sooner
not
unlucky on several occasions
Most importantly, he realised that the way to beat the Germans was to bring them to
battle and to deprive them of resources,most particularly food. He set in train the
had
for
but
to give way to another to
victory,
necessaryreorganisation and strategy
achieveit.
In turn, General van Deventer never fulfilled his instructions although he came
close on a number of occasionsand the Schutztruppewas reducedto very small numbers.
He had learned from his earlier experiencesof the folly of simply trying to capture
territory and the absolutely critical need for an efficient system of logistics.
Consequently,he was prepared to delay moves in order to build up adequatestocks of
supplies,although, when the opportunity presenteditself, he wasjust as capableas Smuts
343

in demanding tremendous efforts on reduced rations. He adopted Hoskins's dual

strategiesof attrition and food denial, and pursued it vigorously. Likewise, once he
found himself operating in areas free of the tsetse fly he re-introduced mounted forces
that operatedto great effect. On the other hand, he did have a tendency to over-control
and at times gave far too many detailed instructions to column commandersoperating in
barely mappedbush. In his keennessto end the campaign, he occasionally committed
too many troops to elaborate moves; apart from complicating matters, they also put too
great a strain on the supply system and restricted the range of key columns. Largely
unnoticedwas the vast amount of work that he directed on road making, which easedhis
supply difficulties considerably. The northern portion of PortugueseEast Africa was
virtually a trackless expanseat the end of 1916; by war's end it was possible to drive a
car from Lindi to Quelimaneand Zomba.
The greatest factor that denied him successwas the performance of the allied
Portugueseforces. Their ineptitude gave von Lettow the opportunity he neededto carry
on the campaignon numerousoccasions.

Generalvon Lettow had succeededin his self-imposedtask of drawing away

British strength,but at considerable
costto his own strength.While the numberof troops
actuallydivertedfrom the main theatresof war weremuchfewerthanhe proclaimed,his
campaigncostthe British a greatdealof moneyandtied up badly neededshippinguntil
the endof the war. In the process,he hadlost the bulk of the Schutztruppe
andthe whole
of GermanEastAfrica, but the retentionof the colonywasnevera priority.
He had marked successagainst Smuts in 1916 and showed great tactical cunning
throughout the campaign. He was undoubtedly a very able commander and achieved a
number of notable victories. On the other hand, the bulk of pitched battles were fought in
1917 and 1918, in which victory was often divided between the two
sides. While von
Lettow was always a dangerousfoe and was quick to pounce on any column that
strayed
too far, his judgement was not always as good as he portrayed. He was particularly taken
with the concept of decisive victory and he sought to win battles of annihilation rather
344

than to inflict steadyattrition. As a result, the British suffered heavily, but so too did the
Schuatruppe, and its losseswere much more difficult to replace. Of course, such sharp
blows frequently halted the advancing British columns and forced reorganisation,giving
von Lettow time to withdraw along pre-planned routes, but they cannot hide the fact that
his forces declined from over 9,000 strong at the end of 1916 to less than 1,500 by
December1917.

His, and others' claims for him, of being a guerrilla fighter are not borne out by
the facts. He was a bush fighter and tactician of the first order, but there is scantevidence
of him employing any of the preceptsof the concept in the modem, acceptedsense. His
forces followed normal German military doctrine, adapted for African conditions and
generally fought as conventional companies. Von Lettow made no real use of armed
civilian irregulars and nor did he try to raise the populace against the invading British.
There appearto have been few attempts to subvert or demoralisethe enemy's troops and
followers, while he recruited only few replacementsduring his sojourn in Portuguese
territory. On the contrary, it was his opponentswho showed a far greatermastery of such
techniquesas they employed over a thousand armed Intelligence Scouts, recruited from
the local population, as well as raising several battalions of ex-German Askaris and
inspiring uprisings against German rule, as with the Makonde. They also distributed
leaflets encouragingdesertionand did their utmost to weaken support for the Germans.
Von Lettow never showed any interest in politics, a key element of guerrilla
warfare, and treatedthe governor with disdain. He treatedthe campaign in East Africa as
a strictly military problem and consideredthat everything else should be subordinatedto
its achievement. In outlook, he sharedmuch more with the conservative officers
of the
PrussianArmy than with such noted guerrilla leadersas Mao, Giap or even his opponent
Smuts. This is not to diminish von Lettow's achievements,but an understanding true
of
guerrilla warfare was not one of them.

345

CHAPTER

11 -BIBLIOGRAPHY

UNPUBLISHED PRIMARY SOURCES

Public RecordOffice (PRO), Kew, London

CAB Series
5
8
21
22
23
24
44
45
103

CO Series
534
551
619

FO Series
371

WO Series
32
33
95
106
346

158
310
808

BodlieanLibrary, Oxford
HarcourtPapers

British Library, London

Buxton Papers

Bundesarchiv/Militararchiv,Freiburgim Breisgau
Boell Papers
von Lettow-Vorbeck Papers

Cambridge
UniversityLibrary
SmutsPapers
Imperial War Museum,London
Duff Papers,Sir HL
Northey Papers

King's CollegeLondon, Liddell Hart Centrefor Military Archives

RobertsonPapers

SouthAfrican National Archives, Pretoria

GovernorGeneralPapers
SmutsPapers

SouthAfrican National DefenceForce,DocumentationDirectorate,Pretoria

PersonnelArchives and Records- van Deventer
German South-West Africa (GSWA) - WWI Papers (confusingly the East
347

African campaignpapersare classified under this seriesof files)

AG 1914-1921
Secretaryof Defence,Group 2
Wiltshire RecordOffice
Long Papers

PUBLISHED PRIMARY SOURCES AND MEMOIRS

Amery, L S, My Political Life Volume2 War and Peace1914-1929, London: 1953.

Anon, "The Campaign in East Africa, With Special Referenceto the Operationsfrom the
N. E. Rhodesia-NyasalandBorder, Under Brig. -General Edward Northey, C.B., A. D.C.",
Journal ofthe Royal United ServicesInstitute, LXII, (1917), pp. 269-285.
Brett-Young, Francis, Marching on Tanga: with General Smuts in East Africa, London:

CollinsSons& Co, 1917.

Britten, Lt Col W E, "PortugueseEastAfrica 1918", Royal EngineerJournal, XXXIX), pp.
430-444.

Buchanan,Angus, Three Yearsof War in EastAftica, London John Murray, 1920.

Callwell, Maj Gen Sir C E, Experiencesofa Dug-Out, London: Constable,1920.
Charlewood,Cdr C J, "Naval Actions off the TanganyikaCoast 1914-1917",11,Tanganyika
Notesand Records,LV, (September1960),pp. 154-180.
Churchill, W S, The World Crisis: 1911-1914,Volume 1, London: Thomas Butterworth
Limited, 1923.

348

Crowe, Brig Gen JHV, General Smuts'Campaign in East Aftica, London: John Murray,
1918.
Deppe,Ludwig, Mit Lettow-Vorbeckdurch Afrika, Berlin: Verlag August Scherl, 1919.

Dolbey, Capt R V, Sketchesofthe East Aftican Campaign,London: JohnMurray, 1918.

Downes,W D, With the Nigerians in GermanEastAfrica, London: Methuen, 1919.

Fendall, Brig Gen C P, The East Aftican Force: 1915-1919,London: HF&G

Witherby,

1921.

G.D.H. (A South African), "With Smuts in German East Africa", Journal of the Royal
UnitedServicesInstitute, LXI, 1916,pp. 820-824.
Hancock,Sir WK and van der Poel, Jean,Selectionsfrom the SmutsPapers: TheFields of
Fire: VolumeIII June 1910 - November 1918, Cambridge: Cambridge University Press,
1966.

Hankey,Lord, TheSupremeCommand,2 Volumes, London: Allen and Unwin, 1961.

Hauer,August Dr, Kumbake:ErlebnisseeinesArtze in Deutsch-Ostafrika1914-1919,

Berlin: Hobbing, 1922.

Hoskins,Lt Gen A R, Despatchfrom the General Officer Commanding-in-ChiefEastAfrica

Relating to Operations in East Africa 20 January to 31" May 1917, dated 30 May 1917,
publishedin the London Gazette'Supplement',No. 30447,27 December1917.
Lloyd, A W, 'Vambo" or with Jannie Smuts in the Jungle: 30 EA Sketches,Cape Town:
CentralNews Agency, [n.d.].

349

Lloyd George,D, War Memoirs, 2 Volumes, London: OdhamsPress,1938.

Keen, Maj F S, "Lecture on the "Campaign in East Africa: Delivered at Simla on the 6th
October1916",Journal ofthe United ServiceInstitution ofIndia, XLVI, 1917,pp. 71-91.
King-Hall, Adrn Sir H G, Naval Traditions and Memories, London:1926.
Khl, HauptmannFranz,Der Kampf um Deutsch-Ostafrika,Berlin: Verlag Kameradschaft,

[1919].
Lettow-Vorbeck, Gen Paul Emil von, Mein Leben, published by Ursula von LettowVorbeck), Biberachan der Riss: Koehlers, 1957.
Meinertzhagen,Col R, Army Diary 1899-1926,London: Oliver and Boyd, 1960.
Methner, Wilhelm, Unter drei Gouverneuren:16 Jahre Dienst in den deutschenTropen,
Breslau:Kom, 1938.

Orr, Col G M, "The Indian Anny in East Affica 1914-1917",Journal ofthe United Service
Institution ofIndia, XLVIII, (1919), pp. 244-261.
Orr, Col G M, "Some Afterthoughts of the War in East Africa, 1914-1918",Journal the
of
Royal UnitedServicesInstitute, LXIX, (1924), pp. 692-702.
Off, Col G M, "From Rumbo to the Rovuma The Odysseyof 'One' Column in EastAfrica
in 1917",,4rmy Quarterly, VII, (1924), pp. 109-129.
Orr, Col G M, "Random Recollections of East Africa", Army Quarterly, X1, (1928),
pp.
282-293.

350

Orr, GM Col CBE DSO IA Retd. "Von LettoNVsEscape into PortugueseEast Africa,
1917",Army Quarterly, Volume XIII, (1925), pp. 50-59.
Orr, GM Col CBE DSO (IA Retd). "Smuts versusLettow A Critical Phasein EastAfrica;
August to September1916",Army Quarterly, XIII, (1925), pp. 287-300.
Orr, Col G M, "Operationson Interior Lines in Bush Warfare", Journal ofthe Royal United
ServicesInstitute, LXX, (1925), pp. 125-134.
Orr, Col G M, "1914-1915 in East Africa", Journal of the United Service Institution of
India, LVI, (1926), pp. 58-86.
Orr, Col G M, "A Remarkable Raid: East Africa 1917", Journal of the Royal United
ServicesInstitute, LXXI, (1926), pp. 73-80.
Orr, Col G M, "The Winter Campaign of 1916 in East Africa", Journal of the United

ServiceInstitutionofIndia, LX, (1930), pp. 69-79.

Overstraeten,General R van, The War Diaries of Albert I King of the Belgians, London:
William Kimber, 1954.

Oxford and Asquith, Earl of, Memories and Reflections,Volume 2, London: Cassell and
Company,1928.

Pretorius,PJ, Jungle Man: TheAutobiography ofMajPJPretorious

CMG, DSOandBar,

London: Harrap, 1947.

Reitz, Denys, Trekking On, London: Faber, 1933.

Robertson,Field Marshal Sir William, Soldiers and Statesmen1914-1918,London: Cassell

and Co, 1926.

351

Schnee,Ada, Meine Erlebnisse whrend der Kriegszeit in DeutschesOstafrika, Leipzig:

QuelleundMeyer,1918.
Schnee,Dr Heinrich, Deutsch-Ostafrikaim Weltkrieg,Leipzig: Quelle und Meyer, 1919.
Schnee, Dr Heinrich, Als Letzer Gouverneur in Deutsch-Ostaftika: Erinnerungen,
Heidelberg:Quelle und Meyer, 1964.
Schoenfeld, Fregattenkapitn Werner, Geraubtes Land.- Durchs freie Sdafrika ins
bedrohteDeutsch-Ostafrika,Dresden:DeutscherBuch- und Kunstverlag, 1942 (reprint of
1925original).
Shorthose,Capt W T, Sport andAdventure in Aftica, London: SeeleyService, 1923.
Smith-Dorrien, General Sir H C, Memoirs of 48 Years Service, London: John Murray,
1925.
Smuts, Lt Gen J C, Military

Despatches from the Commander-in-Chief,

Force, Despatch dated 360,4pril

East Africa

1916, published in the London Gazette, 'Supplement',

No. 2963 0,20 June 1916.

from the Commander-in-Chlef,

Smuts,Lt Gen JC Military Despatches
East African
Forces, Despatch dated 28h February 1917, published in the London Gazette"
Supplemenf',No. 30020of 180'April 1917
Smuts,Lt Gen J C, East Aftica: Address to the Royal Geographic Society 28 January
on
1918 by General Smuts(Reprintedftom the GeographicalJournal, March 1918), London:
Royal GeographicalSociety, 1918.

Smuts,J. C, Wartime Speeches:A Compilation of Public Utterancesin Great Britain by

Lieut.-Gen. The A Hon. JC. SmutsP.C., KC., MLA. In Connectionwith the Session
of

352

London:
Hodder
Conference
1917,
War
Imperial
Cabinet
War
Imperial
and
the
and
Stoughton,1917.
Stewart,Sir J M, Jimmie Stewart - Frontiersman: The Edited Memoirs of Major General
Sir JMStewart, (ed) Robert Maxwell, Edinburgh: PentlandPress,1992.

Thielemans,Marie-Rose,Albert ler: Carnets et Corresonpondancede Guerre 1914-1918,

Paris:Editions Duclot, 1991.

Thonihill, C J, Taking Tanganyika: Experiences of an Intelligence Officer, 1914-18,

London: StanleyPaul, 1917.

Trew,Lt Col H F, BothaTreks,London:Blackie,1936.

British
Van Deventer,Lt Gen J L, Despatchftom the GeneralCommanding-in-Chief,
Forces in East Africa, Relating to OperationsFrom 30'hMay to I" December,1917,
dated21 January1918,publishedin LondonGazette,"Supplemenf',No. 30611,5 April
1918,
Van Deventer, Lt-Gen Sir J L, Despatchftom the Lieutenant-General Commanding-in-

Chief, East African Force, 30 September 1918 published in London Gazette,

"Supplemenf', No. 31069,16 December 1918

War Office, Field ServiceRegulations,Parts I and II, London: HMSO, 1909.

Wenig, Richard,In Monsun und Pori, Berlin: Safari-Verlag, 1922.
Wienholt, Arnold, TheStory ofthe Lion Hunt, London: Melrose, 1922.
Wynn, Wynn E, Ambush,London: Hutchinsonand Co, 1937.

353

OFFICIAL PUBLICATIONS

Admiralty, Reviewof GermanCruiser Warfare 1914-18,London: IIN4SO, 1940.

Admiralty (NID),, 4 Manual ofthe Belgian Congo,London: HMSO, 1919.
Admiralty (NID), A HandbookofGerman East.4ftica, London: IMSO, 1923.
Admiralty (NID), A HandbookofPortugueseNyasaland,London: HMSO, 1920.
Admiralty (NID), A HandbookofPortugueseEastAftica, London: HMSO, 1920.
Admiralty (NID), A Handbookofthe UgandaProtectorate,London: HMSO, 1920.
Assmann,Konteradmiral Kurt, Der Krieg zur See1914 1918: Die Kaempfeder
Kaiserlichen Marine in den DeutschenKolonien, Berlin: Verlag ES Mittler & Sohn, 1935.

Perry, FW (comp), History of the Great War: Orders Of Battle ofDivisions: Part 5a The
Divisions of Australia, Canada and New Zealand and those in East Aftica, Newport,
Gwent: Ray WestlakeMilitary Books, 1992.
Blenkinsop, Gen Sir LJ
and Rainey, Lt Col J W. History of the Great War: Veterinary
Services,London: HMSO, 1925.
Collyer, Brig Gen J J, TheSouthAfricans

with GeneralSmutsin GermanEastAfrica 1916,

Pretoria:GovernmentPrinter, 1939.
Corbett,Sir Julian S, History ofthe Great War.- Naval Operations,Volumes 1-3, London:
Longmans,Greenand Co, 1920-1924.

354

Front,
Western
Transportation
War:
Great
the
History
Official
M,
on
the
A
Hemiker,
of
London: HMSO, 1937.
Hordem, Lt Col Charles, Official History of the Great War: Military Operations: East
Aftica, 1914-1916,Volume 1, London: HMSO, 1941.
Jones,HA and Raleigh, W, Official History of the Great War: The War in the Air, 6
Volumes,Oxford, 1922-1937.

Lucas, Sir C P, The Empire at War: Volume 4 Aftica, Oxford: Oxford University Press,
1925.

Macpherson,
Maj GenSir WG et al, History of the Great War:MedicalServicesGeneral
History Vol 1: Medical Servicesin the UnitedKingdom;in British GarrisonsOverseas;
South-West
Cameroons,
During
Operations
Tsingtau;
Togoland,
in
the
and
and
against
Africa. London:HMSO, 1921.
Macpherson,Maj Gen Sir WG and Mitchell, Maj T J, History of the Great War: Medical
ServicesGeneral History Vol 4: Medical ServicesDuring the Operations on the Gallipoli
Peninsula; in Macedonia; in Mesopotamiaand North-WestPersia, in East Africa; in the
Aden Protectorate, and in North Russia:Ambulance Transport During the War, London:
HMSO, 1924.
Macpherson,Maj Gen Sir W G, Herringham, Maj Gen Sir W P, Elliott, Col T R, and
Balfour, Lt Col A, History the Great War.- Medical Services:Diseasesof War, Volume
of
1, London: HMSO, [n.d.].

Mitchell, TJ and Smith, G M. History ofthe Great War: Medical services: Casualtiesand
Medical Statistics of the Great War; Nashville, Tenn: The Battery Press,[n.d.] (reprint of
1931original).

355

Murray, Sir Archibald, Sir Archibald Murray's Despatches,London: JM Dent and Son,
1920.

Newbolt, Henry, History of the Great War.- Naval Operations,Volumes 4 and 5, London:
LongmansGreenand Co, 1928- 1931
Mantey,Vizeadmiral E von and Raeder,Kapitdn zur See,E, Der Krieg zur See19141918: Der Kreuzerkrieg in den ausIdndischesGewdssern,3 Volumes, Berlin: ES Mittler
& Sohn, 1922-1937.

Reichsarchiv,Der Weltkrieg 1914 bis 1918: Die Militrischen Operationenzu Lande: 9.

Band, Berlin: ES Mittler & Sohn, 1942.

Reichsarchiv,
Der Weltkrieg1914bis 1918:Die MilitrischenOperationen
13.
Lande:
zu
Band,Berlin: ES Mittler & Sohn,1956.
Royaumede Belgique, CampagnesColoniales Belges: Tomes1-3, Bruxulles: Imprimerie
Typographiquede M. C.M., 1927-1932.
War Office, Statistics of the military Effort of the British Empire During the Great War,
1914-1920. HMSO, London, 1922.
War Office, "The Battles NomenclatureCommittee" The Official Names the Battles
of
andOtherEngagementsfought by the Military Forces of the British Empire during the Great
War, 1914-1919
War
1919,
London:
HMSO,
1922.
ThirdAfghan
the
and
SECONDARY SOURCES

Adams, RQE, Arms and the Wizard.,Lloyd George and the Ministry ofMunitions 19151916,London: Cassell& Co, 1978.

356

Palestine,
East
Africa
German
Field
Artillery
in
Aftican
South
The
F
B,
and
Adler, Brig
1915-1919,Pretona:van Schaik, 1958.
Armstrong, H C, Grey Steel: JC Smuts,A Study in Arrogance, London: Barker-Methuen,
1937.
Austin, R A, Northwest Tanzaniaunder Germanand British Rule 1889-1959,New Haven:
Yale University Press,1968.

Bald,

Detlef,

Deutsch-Ostafrika 1900-1914: Eine

Studie

ber

Verwaltung

Munich:
1970.
Erschliessung,
Interessengruppen
Wirtschftliche
und
Beadon,Col R H, Royal Army Service Corps HistorY: Volume2, Cambridge: Cambridge
University Press,1931.

Beesly,P, Room40: British Naval Intelligence 1914-1918,London: Hamilton, 1982.

Beyers,CJ and Basson,J L, Dictionary of SouthAftican Biography, Volume V, Pretoria:

Human SciencesResearchCouncil, 1987.
Bidwell, Shelford and Graham, Dominick, Fire-Power. British Army Weapons and
Theoriesof War 1904-1945,London: GeorgeAllen & Unwin, 1982.
Body, CaPt 0 G, "Bush
Weapons",
Against
Modem
Army
Quarterly,
Fighting
Forest
and
VIII, PP.314-324.
Boell, Ludwig, Die Operationenin Ostafrika, Hamburg: Walter Dachert, 1951.

Bonhain-Carter,Victor, Soldier True: the Life and Times of Field Marshal Sir William
Robertson,Bart., GCB, GCMG, KCVO, DSO, London, 1963.

357

Brelsford, W V, The Story of the Northern RhodesiaRegiment,Lusaka: The Govemment

Printer, 1954.

Brown, Ian, British Logistics on the WesternFront, Westport, CT: Praeger,1998

Brown, JamesAmbrose, They Fought for King and Kaiser: South Afticans in German
EastAftica 1916: SouthAfrica at War Volume6, Johanncsburg:Ashanti Publishing, 1991.
Buxton, Earl, General Botha, London: John Murray, 1924.

Cann,John,"Mozambique,GermanEast Africa and the Great War", Small Warsand

Insurgencies,XII, No 1, (2001), pp. I 10-141.

Chapman-Huston,
WDM, GeneralSir John Cowans:TheQuartermasterGeneralof the
Great War, 2 Volumes, London: Hutchinson, 1924.

Clayton, Anthony, The British Military Presence in East and Central Africa, Oxford:
Oxford DevelopmentReportsProject, 1982.

Clifford, Sir H C, TheGold CoastRegimentin the EastAfrica Campaign,Nashvill,Tenn:

TheBatteryPress,1995(reprintof 1920original).
Cocker, Mark, Richard Meinertzhagen: Soldier, Scientist
and Spy, London: Secker and
Warburg, 1989.
Dane, Edmund, British Campaignsin Africa and the Pacific 1914-1918,London: Hodder
and Stoughton,1919.
Davenport,TRH, South Aftica: A Modern History, London: Macmillan, 40' ed. 1991, pp
184-191.

358

Davenport,T R. H. "The South African Rebellion", English Historical Review,LXXVIII,

(January1963),pp. 73-94.
Light Railways of the First World War: A History of Tactical Rail
Communicationson the British Fronts 1914-1918,Newton Abbot: David & Charles,1962.

Davies, WJK,

Difford, I D, The Story of the Ist Cape Corps, 1915-19, Cape Town: Hortons Limited,
1921.
Eberlie, R F, "The GermanAchievement in East Africa", TanganyikaNotes and Records.
No 55, (September1960),pp. 181-213.
Eckart, Wolfgang U, Medezin und Kolonial-Imperialismus Deutschland 1884-1945,
Paderbom:Schoenigh,1997.

Engelenburg,F V, GeneralLouis Botha, Pretoria:van Schaik, 1929.

Famdale,Gen Sir M, History of the Royal Artillery., The Forgotten Fronts and the Home
Front 1914-1918,London: The Royal Artillery Institute, 1988.

Fischer,F, Germany's
Aimsin theFirst WorldWar,London:ChattoandWindus,1967.
Fisher, John, Curzon and British Imperialism in the Middle East, 1916-1919, London:
Frank Cass,1999.

Flament, F et al, La Force Publique de sa naissancei 1914: Participation des militaires

Departments
des
Brussels:
I'histoire
du
State
Public
Congo,
of
and
premMres
annees
ii
Institutions, 1952

359

Forbes,Maj Gen A, A History of the Army OrdnanceServices,Volume 3, London: Medici

Society, 1929.
Fraser,David, "British War Policy and the Crisis of Liberalism in May 1915", Journal of
Modern History, LIV, No. 1, (March 1982),pp. 1-26.
French, David, British Strategy and War Aims 1914-1916, London: Unwin and Allen,
1986.

French,David, TheStrategyofthe Lloyd George Coalition, 1916-1918,Oxford: Clarendon

Press,1995.

Freeland,JC Col, "The Indian Army Since the Great War", Journal of the Royal United
ServicesInstitute, LXXIII, (1928), pp. 757-767.
Gann, LH and Duignan, P, The Rulers of German Africa 1884-1914,Stanford: Stanford
University Press,1977.

Gann, LH and Duignan, P, The Rulers of British Africa 1878-1914. Stanford: Stanford
University Press,1978.

Gardner, Brian, German East Aftica: the Story of the First World War in East Africa.
London: Cassell,1967.

Garson,N G, "South Africa and World War F', Journal of Imperial and Commonwealth
History, VIII, No 1, (1979), pp. 68-85.
Gifford, Prosserand Lewis, William Roger (eds),Britain and Germanyin Africa: Imperial
Rivalry and Colonial Rule, New Haven: Yale University Press,1967.

360

Gillman, C, "A Short History of the Tanganyika Railways", Tanganyika Notes and
Records,XIII, (June 1942),pp. 14-56.
Gilbert,Martin, WinstonS Churchill.- Volume1111914-1916,London: Heinemann,1971.
Gilbert, Martin, Winston S Churchill.- Volume III Companion 2 1914-1916, London:
Heinemann,1971.

Goebel,F, 40,000KMZeppeIin-KrIegsfahrt, Leipzig: Koehler, 1925.

Gooch,John, The Plans of War: The General Staff and British Military Strategy c 19001916,London: Routledge& Kegan Paul, 1974.

Graharn,Brig CAL,

The History of the Indian Mountain Artillery, Aldershot: Gale and

Polden,1957.

Groener,Erich, German Warships1815-1945: VolumeI Major Surface Vessels,Annapolis

Md: Naval Institute Press,1990.
Guinn, Peter,British Strategyand Politics 1914 to 1918, Oxford: ClarendonPress,1965.
Halpern,Paul G, A Naval History World War I, London: UCL Press,1994.
of
Harrison, Mark, "Medicine
Case
Command:
Malaria
Culture
Control
in
the
the
of
of
and
the British Army during the two World Wars", Medical History, XL, (1996),
pp. 437-452.
Hatchell, G W, "The East African Campaign: 1914 to 1919", Tanganyika Notes
and
Records,XKI, (June 1946),pp. 39-45.
Hatchell, G W, "The British Occupation of the South-Western Area of Tanganyika
Territory 1914-1918",TanganyikaNotes and Records, LI, (December1958),pp. 131-155.
361

Haywood, A and Clarke, F A, The History of the Royal WestAftican Frontier Force,
Aldershot: Gale and Polden, 1964.
Haupt, Wemer, DeutschlandsSchutzgebietesin bersee 1884-1918,Friedberg: PodzunPallas-Verlag,1984.

Hazlehurst,Cameron,"Asquith as Prime Minister, 1908-1916",English Historical Review.

LXXXV, (1970), pp. 502-531.
Heichen, Walter, Helden der Kolonien: Der Weltkrieg in unserenSchutzgebieten,Berlin:
A Weichert Verlag, [n.d.].

Henderson,
W 0, Studiesin GermanColonialHistory,London:FrankCass& Co, 1962.
Hodges,GWT, "African Manpower Statisticsfor the British Forcesin East Africa, 19141918",Journal ofAfrican History, IX, No 1, (1977), pp. 101-116. '
Hodges, Geoffrey, The Carrier Corps: Military Labour in the East African CamPaign,
Westport,Conn: GreenwoodPress,1986.
Hoyt, E P, Guerrilla: Colonel
von Lettow-Vorbeck and Germany'sEast African Empire.
New York: Macmillan, 1981.
Hughes,C and Nicholson, I E, "A Provenance Proconsuls:British Colonial Governors
of
Journal of Imperial and CommonwealthHistory, IV, No 1, (1977),
1900-196011,
pp. 77106.

Hughes,Matthew, Allenby and British Strategy in the Middle East 1917-1919,London:

FrankCass,1999.

362

Hyam, Ronald, The Failure of South Aftican Expansion 1908-1948,London: Macmillan,

1972.

Iliffe, John, Tanganyika under German Rule 1905-1912, Cambridge: Cambridge

University Press,1969.

Isaacman,The Tradition ofResistancein Mozambique,Berkeley: UCLA Press,1976.

Joly de Lotbiniere, Capt H A, "Notes on the Re-organizationand Training of Indian Sapper
and Miner Field Companiesfor Modem Warfare", Royal Engineer Journal, XXIV, pp. 5564.

Kaponen, Juhani, German Colonial Policies in Mainland Tanzania 1884-1914, Studia

Historica 49, Helsinki: Finnish Historical Society, 1995
Katzenellenbogen,S E, "Southern Africa and the War of 1914-18", in MRD
War and Society,London: Paul Elek,1973,pp 107-122.
Keble Chatterton,E, Severn'sSaga,London: Hurst

Foot (ed),

and Blackett, 1938.

Keble Chatterton,E, TheKdnigsbergAffair, London: Hurst

and Blackett, 1936.
Killingray, David, "Repercussions
of World War I in the Gold Coast", Journal of. 4frican
History, XIX, No 1, (1978),
pp. 39-59.

Killingay, David, "The Idea of British Imperial African Army", Journal

of 4frican
History, XX, No 3, (1979), pp. 421-436.
Killingray, David and Matthews,J, "Beastsof Burden: British West African Carriersin the
First World War", CanadianJournal of4frican Studies,XIII, No 1, (1979), pp. 5-24.

363

in
Imperialism,
Studies
Empire,
Guardians
(ed),
David
Omissi,
Killingray, David and
of
Manchester:ManchesterUniversity Press,1999.
Killingray, David, "Labour Exploitation for Military Campaignsin British Colonial Africa,
1870-1945",Journal ofContemporary History, XKIV, No 3, (July 1989),pp. 483-502.
Kirke, E St G Maj, "Notes on Military Railways", Royal Engineer Journal, XL, No 7, pp.
207-219.

Kock, Nils, Blockadeand jungle, London: Hale, 1941.

Koponen, Juhani, Developmentfor Exploitation: German Colonial Policies in Mainland
Tanzania,1884-1914, Helsinki: Finnish Historical Society, 1994.

Koss, Stephen, "The Destruction of Britain's Last Liberal Government", Journal of

Modern History, XL, No. 1, (March 1968),pp. 257-277.
Langhorne, RTB,

"The Anglo-German Negotiations Concerning the Future of the

PortugueseColonies in Africa", Historical Journal, XVI, No 2, (1973), pp. 361-387.
Lloyd-Jones, King's African Rifles: Being
an Unofficial Account of the Origins and
Activities ofthe KAR, London: Arrowsmith, 1926.
Louis, W R, "The United States
and the African PeaceSettlementof 1919", Journal of
African History, IV, No 3, (1963), pp. 413434.
Louis, W F, Ruanda-Urundi 1884-1919,Oxford: ClarendonPress,1963.
Louis, W R, Great Britain and Germany'sLost Colonies,Oxford: ClarendonPress,1967.

GeneralLouisBotha,London:Cassell,1970.
Meintjes,Johannes,
364

McLaughlin,Peter,RagtimeSoldiers,Bulawayo: Books of Zimbabwe, 1980.

Majumdar,Lt Col Birenda, Developmentof the transport systemin the Indian Armyfrom
1760to 1914,Army Quarterly, LXXVII, (1959), pp. 250-260.
Martins, General Ferreira, Portugal's Co-operation with Great Britain in the Great War
1914-1918,Lisboa: Servicosde Informacaoe Imprensa.da EmbaixadaBritanica, 1943.
Mackay,Ruddock F, Fisher ofKilverstone, Oxford: ClarendonPress,1973.
Marder, Arthur, From the Dreadnought to the ScapaFlow: The Royal Navy in the Fisher
Era, 1904-1919,Volumes 1-5, London: Oxford University Press,1961-1970

Marder, Arthur, Fear God and Dread Nought. the CorrespondanceofAdmiral of the Fleet
F Lord Fisher ofKilverstone, 3 Volumes, London: Cape, 1952-29.
Meneses,Filipe Ribeiro,de, "Too Seriousa Matter to be Left to the Generals? Parliament
and the Army in Wartime Portugal 1914-18", Journal of ContemporaryHistory, XXXIII,
No 1, (1998), pp. 85-96.
Monson, Jamie, "Relocating Maji Maji: The Politics
in
Alliance
Authority
the
and
of
SouthernHighlands of Tanzania, 1870-1918",
Journal of, 4frican History, XXXIX, (1998),
pp. 95-120.

Mosley, Leonard,Duelfor Kilimanjaro, London: Weidenfeld & Nicolson, 1963.

Moyse-Bartlett,Lt Col H, TheKing'sAfrican Rifles, Aldershot: Gale and Polden, 1956.
Mwase, GeorgeSimeon,Strike a Blow and Die: The Story ofthe John ChilembweRising,

1970.
London:Heinernann,
365

1919.
Green,
Longmans
London:
Selous,
DSO,
Courtenay
Frederick
G,
Life
J
Millais,
of

London:FaberandFaber,1936
Millin, Sarah,GeneralSmuts,2 Volumes,
Macpherson,Fergus,Anatomy of a Conquest: the British Occupation of Zambia, 18841924,London: Longman, pp 168-175.
NaIder, Maj Gen RFH,

The Royal Corps of Signals: A History of its-Antecedentsand

Development,London: Royal SignalsInstitution, 1958.

Newitt, Malyn, A History ofMozambique, London: Hurst & Company, 1995.

Newitt, Malyn, Portugal in Africa, London: Hurst & Company,1981.

Osuntokun,Akinjide, "Disaffection and Revolts in Nigeria during the First World War,
1914-1918", CanadianJournal ofAftican Studies,V, (1974).
Osuntokun, Akinjide, Nigeria in the First World War, Ibadan History Series, London:
Longman, 1979.

Page,Melvin E (ed),Africa and the First World War, London: Macmillan, 1987.
Page,Melvin E. "The War
of THANGATA: Nyasaland and the East African Campaign,
Journal of.4frican History, XIX, No 1, (1978),
1914-191811,
pp. 87-100.
Perry, F W, The CommonwealthArmies; Manpower and Organization in the Two World
Wars,Manchester:ManchesterUniversity Press,1988.
die
deutschen
Karnpf
Kolonied', Der Erste Weltkrieg,
"Der
Wolfgang,
um
Petter,
1994,
392-411.
Piper,
Mnchen:
(ed),
pp.
Michalka
Wolfgang
366

St
1917",
in
Rising
Makombe
Africa:
East
in
the
Portuguese
of
0,
"Revolt
Terence
Ranger,
Chatto
&
London:
2,
No
Affairs
15
African
Number
Kirkwood,
K.
Antony's Papers, (ed)

Windus,1963.
Rathbone,Richard, "World War I and Africa: Introduction", Journal of .4ftican History,
XIX, No 1, (1978), pp. 1-9.
Ridgway, Brig Gen R T, "With No 2 Column", I, Army Quarterly, V, (1922), pp. 12-28.
Ridgway, Brig Gen R. T, "With No 2 Column German East Africa 1917", 11,.4rmy
Quarterly, VI, (1923), pp. 247-263.
Ritter, Gerhard, The Sword and the Sceptre: The Problem of Militarism in Germany
VolumeIII. The Tragedyof Statesmanship BethmannHollweg as War Chancellor (19141917),Translatedby Heinz Norden, London: Allen Lane, 1973.

Ritter, Gerhard, 7he Sword and the Sceptre: The Problem of Militarism in Germany
VolumeIV.- TheReign of GermanMilitarism and the Disaster of 1918, Translatedby Heinz
Norden,London: Allen Lane, 1973.
Roskill, Stephen,Hankey. Man

ofSecrets, Volmne 1, London: Collins, 1970.

Rothwell, V F, British War Aims

and Peace Diplomacy 1914-1918, Oxford: Oxford
University Press,1971

Sands,Lt Col EWC,

The Indian Sappers and Miners, Chatham: Institution of Royal

Engineers,1948.

367

Savage,Donald C and Forbes Munro, J, "Carrier Corps Recruitment in the British East
Africa Protectorate,1914-1918",Journal of Aftican History, VII, No 2, (1966), pp. 313342.

Sheppard,Brig Gen S H, "SomeNotes on Tactics in the East African Campaign",Journal

Service
Institution
United
ofIndia, XLVIII, No 215, (April 1919),pp. 138-157.
ofthe
Smith, Gaddis, "Disposition of the German Colonies" in Prosser,Gifford and Louis, W
Roger,Britain and Germanyin Aftica, New Haven: Yale University Press,1967.

Smuts,J C,Jan ChristiaanSmuts,London:Cassell,1952.

Solf, Wilhelm, "Schneeund von Lettow-Vorbecle', Die DeutscheNation, (1920), pp. 8795.

Spies, S B, "The Outbreak of the First World War and the Botha Govemment", South
African Historical Journal, 1969,pp. 47-57.
Strachan,HFA,
Press,2001

The First World War - To Arms, Volwne I, Oxford: Oxford University

Stronge,Capt HCT, "Bush Warfare Against Trained Troops",Journal

ofthe Royal United
ServicesInstitution, LXVI, (1927),
pp. 603-613.
Stoecker,Helmuth (ed), German Imperialism in Africa: From the Beginnings
until the
SecondWorld War, translatedBemd Zoellner, London: C Hurst & Co, 1986.
Taylor, Stephen, The MightY Nimro&
a Life of Frederick Courteney Selous: African
Hunter andAdventurer, London: Collins, 1989.

368

Thatcher,W S, The Fourth Battalion, Duke of Connaught'sOwn 10th Baluch Regimentin

the Great War, Cambridge:CambridgeUniversity Press,1932.
Woodward, David P, Lloyd George and the Generals, London: Associated University
Presses,1983.

Taute, Dr M, "A Gennan Account of the Medical Side of the War in East Africa, 19141918", TanganyikaNotes and Records,VIII, (December1939), pp. 1-20.
Thompson, E S, "A Machine Gunner's Odyssey through German East Africa7, Part 1,
SouthAftican Military History Journal, VII, No 3, (June 1987); Part II, VII, No 4, (June
1987);and Part III, VII, No 6, (December1988).

Turner,J, "Cabinets,CommitteesandSecretariatsTheHigherDirectionof War", in KM

Burk (ed), War and State: The TransformationofBritish Government1914-1918,London:
GeorgeAllen & Unwin, 1982,pp. 57-83.
Turner,John,British Politics
Vincent Snjith,

Press,
1992.
University
Yale
New
Haven:
Great
War,
the
and

J D, "Britain, Portugal and the First World War, 1914-1918",European

StudiesReview,IV, No 3, (1974),
pp. 207-238.
Vincent Smith, J D, "The Portuguese Republic and Britain, 1910-1914", Journal of
ContemporaryHistory, X, No 4, (1975), pp. 707-727.
Waley, Daniel, A Liberal Life: Lord Buxton, Newtimber: New Timber Publications,1999.
Warhurst, P R, "Smuts and Africa:
a Study in Sub-Imperialism", South Aftican Historical
Journal, XVI, (1984), pp. 82-100.

369

Woodhouse, Capt H L, "East Africa", Journal of the United Service Institution of India,
XLVI, (1917),pp. 329-337.
Woodhouse, Capt H L, "Railway Demolitions and Repairs", Royal Engineers Journal,
XKIV No 2, (1920),pp. 49-54.
,
Woodhouse,Capt H L, "Notes on Railway Work in EastAfrica", Royal EngineersJournal,
XXVII, No 1, (1923),pp. 37-46.
Woodhouse,Maj H L, "Railway Organisationfor the Indian Army", Journal of the Royal
United ServicesInstitute, Voltune LVI, pp. 95-102.
Woodward, D R, "Britain in a Continental War: The Civil Military Debate over the
StrategicalDirection of the War of 1914-1918",Albion, XII, (1980), pp. 37-65.
Wylly, H C, The Loyal North Lancashire Regiment, London: Royal United Service

Institution,1933.
Yearwood, Peter J. "Great Britain and the Repartition of Africa 1914-19", Journal of
Imperial and CommonwealthHistory, XVIII, No 3, (October 1990),pp. 316-341.
Zirkel, Kirsten, "Military Power in Gennan Colonial Policy: The Schutztruppeand their
Leadersin East and South-WestAfrica, 1881-1918", in Guardians ofEmpire, David
Killingray and David Omissi (eds), Manchester:ManchesterUniversity Press,1999,
pp.
91-113.

--170
d
GLAS
UNIVERSITY
LIBRARY