The Commonwealth Signals Intelligence Conference, 1946

R.M.S. 'Queen Mary' at Halifax, N.S., 1946.jpg

"R.M.S. Queen Mary at Halifax, N.S.," 1946. Clifford M. Johnston / Library and Archives Canada / PA-057032. The Canadian delegation of George Glazebrook, Bill Crean, Edward Drake and Mary Oliver travelled to the UK aboard the Queen Mary in February 1946.

Canada developed new secret intelligence capabilities during the Second World War, including the infrastructure and expertise to intercept and decrypt Japanese and French signals traffic. The Canadians involved were part of a wider Allied effort, and signals intelligence work (SIGINT, or Ultra intelligence in British and Commonwealth parlance), did not cease in peacetime. Instead, American, British, and Canadian signals intelligence organizations were repurposed to meet Cold War needs. The documents in this briefing book shed light on Canadians’ thinking about the purpose and structures of permanent postwar Canadian and allied intelligence communities (CDSG00006).

In late 1945, Canadians were kept apprised of negotiations towards a UKUSA SIGINT agreement, which later became the nucleus of the Five Eyes alliance (CDSG00001). As these bilateral negotiations progressed, the British proposed a Commonwealth SIGINT Conference in February 1946. The first item on the agenda was to discuss, agree, and arrange “the participation of Commonwealth authorities in a British global SIGINT network under broad general direction of the UK” (CDSG00014). Canadians with knowledge of the Ultra secret objected to the proposed division of SIGINT labour “on an empire basis” (CDSG00039). Unlike their counterparts from Australia, New Zealand, and India, who were unaware of UKUSA, Canadian officials stressed the need to have direct relationships with both London and Washington in the field of SIGINT (CDSG00007). Sir Edward Travis, Director of the British Government Code & Cypher School, acknowledged Canadians’ “special position” and opened the Commonwealth Conference with “a full statement on the agreement and on relations between Canada, [the] USA, and the UK” (CDSG00015 & CDSG00025). In late 1945 and early 1946, the Canadians were also concerned about the idleness of SIGINT staff in Ottawa. They therefore requested interim assignments from their British and American allies, pending a more permanent allocation of duties after the February conference. Ottawa accepted some tasks delegated by London (the specific nature of these has been withheld under the Access to Information Act), but turned down others. For instance, Canadian codebreakers were unable to complete a Chinese-book breaking assignment from Washington due to the inexperience of their Chinese linguists (CDSG00029).

A delegation of four veterans of wartime intelligence work represented Canada at the 1946 Commonwealth SIGINT Conference. George Glazebrook and Gordon Gale “Bill” Crean of the Department of External Affairs were key architects of Canada’s postwar intelligence system. The formal conference program emphasized the “offensive” aspects of SIGINT work, notably the interception of foreign states’ communications. Crean was tasked with informally discussing “defensive” aspects of SIGINT work with UK counterparts, such as codes and ciphers to protect the secrecy of Canadian and allied communications. Lieutenant-Colonel Edward Drake was the designated head of a new peacetime signals intelligence agency, which became operational later in 1946. The Communications Branch of the National Research Council, Canada's national cryptologic agency, is today styled the Communications Security Establishment. The incipient CBNRC was also represented by Mary Oliver, whose knowledge and experience in SIGINT far exceeded the bounds of her wartime administrative assistant title. Canadian diplomat Hume Wrong observed that Oliver’s addition to the Canadian delegation “created a demand for equality of rights [from women] in Travis’s show” (CDSG00043).

The Canadians sailed to London aboard the Queen Mary, stayed at the Dorchester Hotel, and made the High Commission (Canada House) their headquarters in February-March 1946. The conference dealt with the technical nuts and bolts of SIGINT work, such as landlines and radio links between intercept stations and codebreaking centres. Commonwealth representatives also grappled with wider problems of how to achieve rapid exchanges of intercepted traffic and avoid “wasteful duplication” of intelligence labour among allies (CDSG00047). The full “Commonwealth SIGINT Technical Conference 1946” file released by Library and Archives Canada under the Access to Information Act is available as CDSG00053.

The Commonwealth Signals Intelligence Conference, 1946