University of Botswana History Department

Colonial Administration Page 4:

Intelligence Reporting In Colonial Botswana, 1895-1965

by Neil Parsons

History Home Page  |  Site Index  |  Neil Parsons
Introduction  |  Page 1. Extracts from Documents (A-J)  |  Page 2. 'Charles Rey and previous commissioners of the Bechuanaland Protectorate'  |  Page 3. The Williams Regime in Bechuanaland

Intelligence gathering to 1950

As in other British colonies, district administration in the Bechuanaland Protectorate originated with a cadre of military police officers in their capacity as resident magistrates. There was only a small 'central' administration, based outside the country - at Vryburg between 1885 and 1895, and thereafter until 1964-65 at Mafikeng. The colonial military police also lost its autonomy between 1895-96 and 1903-04, when it simply constituted part of the (Rhodesian) British South Africa Police - because of the expectation that Bechuanaland would become the third province of Southern Rhodesia.

Intelligence gathering, of course, was a regular part of the activities of colonial magistrates and police. Thus, in the Botswana National Archives (BNA), we have File RC 3/7 'Mr. Moffat's Report on Khama's People. February 1895'. And in 1906 the Transvaal Native Affairs Department passes on an intelligence report from Pietersburg of a rumoured rising at Molepolole in the BP, which will link the Witbooi rising in German South West Africa with the Bambatha rising in Natal (BNA-S. 33/3). But these examples are not comparable with later formal intelligence procedures.

The earliest evidence of 'modern' security intelligence activity in the Botswana National Archives appears to be a letter from the Chief of the Union Defence Force of South Africa in Pretoria to the office of the British high commissioner (also governor-general of the Union) in Cape Town, dated 22 February 1923, which was passed on to the Resident Commissioner at Mafikeng. The letter, which was also being sent to the Secretary for Native Affairs in South Africa, contained a request from the Chief of Imperial General Staff (CIGS) in London for reports on the activities of Pan-Africanism (i.e. Garveyism) in southern Africa.

The Resident Commissioner replied: 'There is no evidence of the existence, let alone the progress, of Pan-Africanism in the Bechuanaland Protectorate, and I do not expect that there will be so long asthe tribal system is maintained.'

In fact there was a little Garveyite activity in the B.P., spilling over from South West Africa. The Resident Commissioner presumably gave his answer off the cuff, as there was no proper intelligence service at ground level. But the letter is interesting in laying out the imperial chain of command at the time, as well as the spirit of the times around 1923.

The top of the chain was CIGS, supreme military commander of British and Imperial/ Commonwealth forces, to whom military intelligence also reported.

Modern British military intelligence dates back to the Secret Service Bureau set up in October 1909, originally with army and navy departments almost immediately transmuted into Home and Foreign Sections - under Home Office (i.e. Police) and Admiralty (i.e. Navy) respectively. The Home Section of SSB was effectively 'Home and Colonial', as it was primarily responsible for intelligence within the Empire/ Commonwealth as well.

The spirit of the times in Britain around 1923 was of course the perceived threat of Bolshevism to state security at home and in the colonies. A wary eye was also being kept on other popular movements which might combine to make a greater threat, notably anti-colonial movements at home and in the colonies.

The imperial chain of command between Britain and Bechuanaland was revised in 1929-31, when South Africa became autonmous with its own governor-general separate from the high commissionership. Thus in 1933 when Tshekedi Khama caused a small 'emergency' in the B.P., use of the the Union Defence Force was ruled out as being 'foreign'. Imperial Marines were sent from Simonstown to Serowe instead. But there is no evidence of a security intelligence network as yet in the B.P. or in the other 'high commission territories' of Basutoland and Swaziland.

The British 'secret services' were shaken up considerably after the outbreak of the Second World War. The Home Section (now called the Security Service), which was concerned with counter-intelligence (countering alien spies) as well as with military security in Britain and the Empire, became known as MI5 (i.e. Military Intelligence Section 5). By 1941, MI5 had come under the control of Lord Swinton (ex Cunliffe-Lister, ex Lloyd-Graeme), former colonial minister and air minister and generally considered incompetent at both.

The Foreign Section of the 'secret service', which was more concerned with active spying abroad, continued to be known as SIS (Secret or Special Intelligence Service) but now became known as MI6 as well. More glamorous because of its naval rather than police connections (vide James Bond), MI6 does not however seem to have been a factor in Southern Africa until after 1961 when South Africa became a republic outside the Commonwealth.

The need for security intelligence during the Second World War can be seen in concern over potential sabotage by the Afrikaner nationalist insurgent group the Ossewa Brandwag. But the formal setting up of a state security apparatus, in the form of a Special Branch (SB) with nominated operatives in the Bechuanaland Protectorate Police seems to date from 1949-50. Two reasons for setting up the SB, reporting to the Chief of Police (codenamed COP!) at Mafikeng, may be offered - the first is as part of general intelligence reorganization in all colonies, and the second is specific to Bechuanaland - the 'emergency' following the Seretse Khama marriage crisis.

Britain had reorganized its security services once again in 1946. The wartime spy-operation Special Operations Executive (SOE) was diluted largely into SIS (MI6). The Security Service hung onto its wartime designation as MI5, because its initials SS now meant something else entirely!

The Cold War and the first wave of decolonization in South Asia, under a Labour government in Britain, gave a new and somewhat contradictory remit to MI5 and its offshoots in the remaining colonies. The remit may be summarized as being to suppress communism while canalizing nationalism to become pro-Western and hopefully even pro-British. The clever strategy which emerged being, as Goldsworthy and others have pointed out, to cultivate and 'turn' the most radical of acceptable nationalist leaders, bringing them out of jail to make them government ministers if necessary.

The long-serving head of the post-War MI5 was Sir Percy Sillitoe, often portrayed simnply as a former London police chief. More pertinently for our purposes he was a former junior officer in the (Rhodesian) British South Africa Police, a former rough frontier military force which had assiduously developed a smoother image by advertising in the Daily Telegraph for 'Sons of gentlemen who can ride and shoot', with no academic qualifications required.

Britain's main problem in Southern Africa was how to come to terms with the 1948 victory of the anti-communist, but in this case anti-British, National Party in mineral-rich South Africa. A dual strategy was pursued. To cultivate and 'turn' the Nationalists on the one hand, and on the other hand to develop Rhodesia as a strong alternative locus for British power and influence in the region. The British were helped by the fact that Dr. Malan, the first National Party prime minister, took on a well-known British 'mole' as his political secretary. Douglas Forsyth had previously served Smuts, and continued to run South Africa's external and security policies under Malan.

South Africa's naval and economic strategic importance for Britain was enhanced by the Cold War, gold and uranium. Instability in the Middle East threatened the Suez Canal, and the restrictions imposed by 1946 U.S. Atomic Energy Act obliged Britain to develop it own atomic bombs. Evidence of South Africa's security importance to Britain can be seen in the visit or visits of Sillitoe as head of MI5 to South Africa in 1949-50.

Within the B.P. the need for political and security intelligence was brought to a head by the totally unexpected rejection of the widely respected Regent Tshekedi Khama by his people in June 1949, and their support for his nephew Seretse Khama's marriage to an English woman. The fear of public disturbance was real. In the absence of a Riot Act in B.P. legislation, the military principles of King's Regulations were adopted: a public warning followed by tear gas, then a bugle call followed by rifle fire aimed 'only upon those persons who can be seen to be implicated in the disturbance' - later on referred to as 'the man in the red shirt'. Section 1254 of King's Regulations explains:' To fire over the heads of a crowd has the effect of favouring the most daring and guilty, and of sacrificing the less daring, and even the innocent.'

In Britain Seretse Khama's cause received strong support from the local Communist Party and its newspaper, the Daily Worker, in 1951-52. Such that the "Establishment" (MI5?) encouraged the veteran social democrat anti-colonial campaigner Fenner Brockway to set up a respectable lobby, full of Labour and Liberal and a few Conservative members of parliament, in support of Seretse Khama in 1952 - to replace an earlier support group with a communist secretary called Billy Strachan. But two years earlier Sir Percy Sillitoe of MI5 had been genuinely surprised to hear the suggestion of a communist conspiracy behind Seretse. Mind you he was being told that Ruth Khama had in effect been a communist agent set to entrap Seretse, 'to bring about active civil war or chaos' in the B.P.!

Intelligence Gathering 1950-1961:

1. Tergos Newsletters

The Bechuanaland Protectectorate in the 1950s and early 60s offers something like a complete picture of state intelligence operations, thanks largely to the chance survival of many samples of two record series otherwise destined for destruction. Copies of the TERGOS series of confidential political intelligence newsletters were systematically destroyed by administrators every few months, but should become fully available in the Public Record Office in London. But Intelligence Reports (IR) on political and security matters, passed up the security change of command, may otherwise only become available after a century or never at all.

TERGOS stood for 'Territories' Gossip', i.e. of the three high commission territories. It appears to have been the invention of, and was certainly first edited by, Nicholas Montserrat the High Commissioner's press officer and aspirant naval novelist.(His second novel, after The Cruel Sea, was based on the Seretse story and called The Tribe that Lost its Head.) TERGOS was inspired by the confidential newsletter already being circulated within Britain's Commonwealth Relations Office called OPDOM - originally to be called DOMOP for 'Dominions Opinion' but changed presumably in deference to Afrikaans sensibilities. But the monthly OPDOMs were essentially digests of what local newspapers were saying. The TERGOS would have to depend much more on word of mouth.

The first monthly TERGOS for all three territories was compiled by Montserrat for June 1950. 'TERGOS No.1. Confidential. News Report on High Commission Territories. Period 1st January - 30th June, 1950' consisted of 17 cyclostyled foolscap sheets, and began on page 1 with the heading 'Seretse and Ruth'. TERGOS No.2 followed in July 1950. The B.P. Government Secretary was told that the Commonwealth Relations Office had requested a regular news report, especially because the Bechuanaland Protectorate was not supplying a quarterly newsletter like its sister territories of Basutoland and Swaziland. The latter would continue to supply their news quarterly, while the politically more volatile B.P. supplied monthly reports. Thus only every third monthly HCT TERGOS would be full of news from Basutoland and Swaziland.

Each TERGOS was supplied to district commissioners and heads of central administration departments in the B.P., for personal information (and marked for quick destruction). The TERGOS might also include items of interest on the Union of South Africa, e.g. a Commonwealth Relations office confidential print (in fact a 5 page carbon copy) of a dispatch dated March 10, 1950, headed 'South Africa: African National Congress and All-African Convention', which noted 'the capture of the Congress by the radical [anti-Xuma] element.'

District commissioners were expected to compile a district TERGOS of local political gossip every month, which was sent on to be incorporated a fortnight later into a TERGOS for the whole territory. The final TERGOS for all three territories was then compiled in the high commissioner's office in Pretoria/ Cape Town a fortnight later. The combined HCT TERGOS failed to appear for three months in 1951, and we have not seen beyond issue 42 for December 1954. But we know that the BP TERGOS, with its district feeders, was still being compiled as late as June 1961.

The system of TERGOS and IR compiling was pepped up and rationalized by the administrative reform in the BP that followed the disastrous Serowe riot of June 1952. Divisional commissioners (equivalent to provincial commissioners elsewhere) were put in charge of the North (at Francistown) and South (at Lobatse). All district TERGOS and IR went through them, and the Northern or Southern Divisional TERGOS and the Northern or Southern Divisional Intelligence Report (NDIR or SDIR) combine the optimum mix of detail useful to historians withe the authority of some sifting and verification.

The writing of TERGOS and Intelligences Reports (IR) by the same person, the District Commissioner, every month, caused some confusion between what should be reported in which. Some officials even tried to get away with two paragraphs on one sheet of paper, one headed TERGOS and the other Intelligence Report.

In September 1960 the Government Secretary at Mafikeng felt obliged to circulate Circular Memorandum No. 94 of 1960 asking district commissioners to put political intelligence (rather than news) in the secret IR, rather than in the confidential TERGOS. He also reminded officers to destry all copies of TERGOS after six months, following his previous instruction of 24 February 1958. This circular also refers to the top intelligence committee at the Mafikeng headquarters of the Special Brach as the Central Intelligence Committee (CIC)


By this time we know that there were problems in keeping TERGOSes confidential and IRs secret. African junior district officers, like Richard Manothoko at Mahalapye, were showing the TERGOSes to their friends.

2. Intelligence Reports

The procedures for collection and recording of intelligence through committees was first laid out in Special Branch Circular No.112 of 1950. This helps to explain the monthly or fortnightly numbers of subsequent IR reports. For example, the Ngwato or Serowe Distict IR series for late 1954 and early 1955 refers to numbers 78, 89, 83, and 84, with a cross reference to Francistown IR No.114. The numbering of the Francistown district IR for June 1954 as No.94 suggests that the series began as a fortnightly issue for August 1950 - but district IR reports from elsewhere are only evident from May 1952.

From this we may also deduce that the Special Branch of the Bechuanaland Protectorate Police became active in the 1950-52 period. The Special Branch ran plain-clothes operatives, nearly always former regular policemen, to collect intelligence in public meeting places (increasingly drinking places as sorghum beer and later other alcohlic drinks became tolerated and semi-legal). Their reports were collected by the district or sub-district (white) police officer and presented to a two or three man district intelligence committee, under the chairmanship of the district commissioner, which met monthly to sift items from operatives and to comment on their reliability.

Who was involved in the IR committee process? The Chobe (Sub-) District IR No.8, for June 1956, for example, was marked for distribution to the Divisional Commissioner (North) Francistown, the Commissioner of B.P. Police Mafikeng, the Officer Commanding B.P. Police Maun, and the Officer in charge of Special Branch, resumably the one at Francistown. IR reports filed intelligence under 8 standard headings: Reports on Meetings; Movements of Prominent Members of Tribes; Attitude of Public towards Government; Race Relations; African National Congress; Communications - roads, radio, airfields, motor transport; General; and, seemingly an afterthought, Cattle.

Special Branch Circular No.4 of 2 October 1959 merely repeated the orders of No.112 of 1950 on the collection and recording of intelligence. What is more interesting is its distribution list which indicates the structure of intelligence knowhow. It indicates that Special Branch and CID were being clearly distinguished until c.1965. The circular was sent from the Office of the Commissioner of the Bechuanaland Protectorate Police at Mafikeng to the Police Divisional Superintendants North (Francistown) and South (Lobatse), to the Officers Commanding (OC) Districts Nos 1-8, the OC Criminal Records Bureau (Mafikeng), OC Special Branch North (Francistown), OC Depot (Gaborone), and 7 sub-district or border post Station Commanders - as well OC Special Branch at Headquarters. By 1960 there were clearly two OCs of the SB at Francistown and Lobatse, under the OC at Headquarters - collating the intelligence mostly collected by plainclothes SB operatives within each district or sub-district force.

Intelligence gathering between 1961 and 1964 was concentrated on the activities of South African and Namibian refugees and political activists - including the so called "pipeline" between Bechuanaland and Tanganyika up and down which activists such as Nelson Mandela moved.


By 1965 with the incoming of a new self-governing constitution in the BP, the Special Branch had become incorporated into the Criminal Investigation Department (CID) of the Police. (There had long been a prejudice against the iuitials "SB" which were also used by the Security Branch of the SA Police.) Intelligence was formally constituted as being channelled up from CID operatives through district committees (the two "divisions" or provinces having been abolished in 1963-64) to the CIC (Central Intelligence Committee) based in the new capital of Gaborone. CIC duties included keeping the 'Internal Security Plan' up to date.

However the head of CID immediately after independence, Bill Grant, was looked upon in some circles as being compromised towards neighbouring white-ruled territories, notably Southern Rhodesia. That is probably what led to his later dismissal, and also that of at least one other top CID white expatriate officer.

Intelligence reports from Mamuno, the main border post with South West Africa, covering the period from 1956 to 1966, are interesting in that they show how intelligence gathering was handed over from colonial expatriate to indigenous Batswana control in the Police with barely a blip. Relations with the SA Police over the border were always a problem, as SAP operatives thought they could cross the border with impunity and also undetected. This resulted in many small protests, and in August 1963 a red hot political row when the Namibian nationalist Kenneth Abrahams was kidnapped on the road between Ghanzi and Lobatse.

In other ways too, the late colonial period set the pattern for the relationship between the intelligence services of independent Botswana and those of neighbouring white-ruled territories after 1965. Even after the formation of the Bureau of State Security (BOSS) in South Africa, it appears that the Vryburg and Zeerust SA Police retained responsibility for liaison with the Botswana Police. In the mid-1980s an SAP sergeant from Vryburg, surely an SB operative, was fined in a Botswana court for having attempted to bribe a new Botswana Police inspector in charge at Lobatse. His defence before the magistrate was simple - hell, I've been doing this for twenty years.

Back to top

Copyright © 1999 Neil Parsons
Last updated 17 September 1999