Virtual Bletchley Park
by Tony Sale

| Back to History Index |
Tony Sale's
Codes and Ciphers

The Birth of SIGINT in Bletchley Park

Early Co-ordination of Radio Interception.

Pre World War II Radio Intercept.

Naval, World War I, Pembroke and Scarborough
Military, Chatham 1921, Sarafand 1923,
Air, Waddigton 1927
FO, GC&CS, Denmark Hill 1923

Earliest evidence of inter-service liaison:
The Cryptographic and Interception Committee 1924
Standing Sub-Committee, Co-ord of WIT intercept 1928
Became the "Y" Committee, 1938
Chair, head of GC&CS,
members, NID 9, MI 1(b), (later MI 8), AI 1(e)

YC sub-committee, programme of intercept and D/F

The Intelligence Units of the Armed Forces.

By the beginning of 1938 the situation was:

Naval Intelligence Directorate (NID) headed by Rear Admiral J A G Troup
SIS Naval Section headed by Fredrick Russel
GC&CS Naval Section headed by William Clarke

Military Intelligence Directorate (MID) headed by ????
SIS Military Section headed by ????
GC&CS Military Section headed by John Tiltman

Air Intelligence Directorate (AID) headed by ????
SIS Air Section headed by Fred Winterbotham
GC&CS Air Section headed by Josh Cooper

SIS headed by Admiral Sinclair, Deputy Menzies
GC&CS headed by Alistair Denniston reporting to Menzies and Sinclair

After the purchase of Bletchley Park in early 1938 by Admiral Sinclair, the Service Sections of GC&CS went to the Park in September 1938 to test its suitability for occupation if war came. The occupation was co-ordinated by Captain Ridley.

This was partially successful. It showed the need for better communications and more accommodation space than just the Mansion.

This first occupation was withdrawn in October 1938 and everyone returned to Broadway.

In early 1939 the construction started of wooden huts clustered around the Mansion. The first huts built were numbered 1,2,3,4 & 5. A Telephone Exchange was built outside the Billiard Room.

On 1st August 1939 SIS and GC&CS moved to Bletchley Park, "the war station". SIS occupied the first floor of the Mansion with GC&CS occupying the ground floor.

The Naval Section moved quickly into Hut 4 alongside the Mansion on the south side. The Army Section moved first into Sir Herbert Leon's dining room. When Hut 5 was completed the Army section moved there and the dining room became a mess room. The Air section moved into the drawing room on the right of the entrance and Denniston occupied the room on the left.

Knox, Jeffreys and Turing moved into the Cottage in the Stable Yard.

Early Enigma breaks

Dilly Knox and Denniston had brought back from the meeting with the Poles in the Pyry Forest, on the 25th July 1939, the wheel wirings of the German Enigma machine and the ideas for the Zygalski Sheets, the Netz, and of Rejewski's Bomba. John Jeffreys set up, in the Cottage, a production line for Netz, now known as Jeffreys Sheets. By December 1939 complete sets were prepared. One copy was taken to France to the Chateau Vignolle, where the Polish code breakers had arrived from Poland. Dilly Knox and Alan Turing first tried the Jeffreys Sheets on some unbroken 1938 traffic. When this worked they were ready to try current traffic. The French and Poles at Vignolles also tried on current traffic and early in January 1940 they succeeded. The first keys broken were for the Red, German Air Force Enigma messages.

The breaking of Enigma, in January 1940, immediately raised a large number of problems, the prime one being how to keep secret the very fact of breaking Enigma. The code breakers involved, and Denniston and "C", realised that it was only because of German errors in the use of Enigma, that it had been broken. By eliminating these errors and with just a few changes to the procedures for using Enigma, the Germans could render breaking impossible. Thus it was absolutely vital that the Germans never realised that Enigma was being broken. The results from breaking Enigma had to be disguised so that even Allied recipients could not discover that what they were receiving came from breaking Enigma.

The result was a complete isolation of the Enigma code breaking within Bletchley Park and the immediate setting up of a small section to handle the translation, emending and concealing of the Enigma decrypts before they left Bletchley Park. All outputs were to go straight to Winterbotham at SIS Broadway and they were concealed by being attributed to agents reports, just like the other outputs from SIS. The first series had the prefix CX for agents reports and series /FJ.

The dilemma of this was that SIS agents reports were thoroughly discredited by the service Intelligence organisations, so not much notice was taken of CX/FJ even though because it was Enigma decrypts it was impeccably accurate.

The section for translating Enigma decrypts was first known as the FJ Section. It was headed by Commander Saunders, a Naval Officer brought down from Broadway because of his knowledge of German and his Intelligence background. He was joined by a civilian, F J Lucas, an Army Captain Edgar and a Flt Lt from the Air Force. They moved into the first Hut 3 which was a hut between Hut 2, the recreation Hut and the Tennis Courts. They soon became known as "Hut 3" as a cover for their work and this stuck for the rest of the war despite various moves to bigger and bigger buildings.

Because the decrypted Enigma traffic was originating from the German Air Force (GAF), Hut 3 was producing intelligence primarily for Air Force Intelligence. (AID). A problem arose because there was already an Air Section within GC&CS headed by Josh Cooper. Because of the need for secrecy about breaking Enigma, Josh Cooper's section knew nothing about the output form Hut 3. The Air Section was breaking German, Italian and other non-machine codes and ciphers.

Interception

Another difficulty was with interception. In 1939 most Naval interception was being done at Flowerdown and Scarborough, long established Naval Intercept stations. The Military used Chatham for interception. The Foreign Office mainly used a Police listening station at Denmark Hill. The Air Force had a station at Waddington. But all these stations were taking very little foreign forces traffic, they were mainly taking diplomatic traffic which went to the Foreign Office.

The Military Intercept station at Chatham was the first one to take Enigma traffic. Before the January 1940 break it was thought that this was German Army Enigma. Because at this time there was very little Enigma traffic, Chatham continued taking it even after it was found that it was actually GAF traffic. However the Air Force had decided to build an intercept station at Chicksands. When Enigma traffic increased vastly during the German invasion of Norway, the Army pressed for Enigma intercepting to be transferred to Chicksands. GC&CS initially resisted this because the operators at Chicksands were mostly untrained and not expert at Enigma traffic.

There were just not enough radio sets to take all the possible Enigma traffic and GC&CS called on Denmark Hill to help.

After much pressure it was decided to build more intercept stations, the Army at Beaumanor and the Air Force at Forest Moor and Knockholt. Later stations were built in Scotland and in the West Country, in all about 20 stations were in operation at the end of the war.

Enigma interception was unique in requiring absolute accuracy in the first few groups of a message. Without this level of accuracy the early Netz method for breaking messages would not work and the later Bombe attacks also required a high degree of accuracy for the "cribs" to work.

Chatham had exceptionally good intercept operators who could consistently produce the high levels of accuracy required for breaking Enigma. GC&CS were thus very concerned at plans by MI8 to transfer Enigma intercept to the relatively inexperienced operators at Chicksands.

Gordon Welchman, in particular, demanded that GC&CS should have control of interception and some rather acrimonious correspondence ensued with Colonel Butler, head of MI8.

Traffic Analysis (WTI)

Another consequence from isolating Enigma was with Traffic Analysis. TA was called W/T I in some quarters and eventually became part of SIGINT.

A number of different groups sprang up, each tackling their own small part of TA for their own ends. For GC&CS, Enigma breaking depended heavily on knowing which parts of the German forces were communicating using Enigma. Right from the beginning listing messages was a vital activity with very close co-operation with the intercept stations. An Intercept Control section was established in Hut 6 very early on.

The War Office, in 1939, was virtually completely ignorant about the German communication systems, the order of battle and where forces were. Colonel Butler had long held the view that Traffic Analysis could provide valuable intelligence. A section was set up at 3 Cork Street in London to study the German communications system. They hoped that by studying the Chatham activity reports they would be able to make deductions about German intentions and order of battle.

Nearby in Broadway Buildings an independent section, organised by Cpt Bolitho with the blessing of MI8, was endeavouring to break down the German callsign system using tabulating machinery.

SIXTA

Lt Col Stratton started the "Central Intelligence Section" in the Spring of 1940, at Caxton St in London for "The collation of identities and locations from decodes with callsign research and D/F plots to produce diagrams of German wireless layout".

In Autumn 1940 this was divided into Long Term Research under Stratton and Short Term Research, known as the "Central Party", based at The Warren, Harpenden.

Special Liaison Party Cpt Lithgow, June 1940, Special Liaison Party (SLP) at GC&CS in Hut 3 to study callsigns and frequencies. Not expected by GC&CS to be of any value, but soon proved very useful indeed to Hut 6 and Hut 3.

VI Intelligence School was started in March 1941, to concentrate on research into German methods. Most of VI IS moved to Beaumanor in July 41 and SLP, now on staff of VI IS remained at GC&CS.

Eventually VI IS all moved to GC&CS in May 1942.

The CIS contribution became vital to the Hut 6 Enigma breaking efforts and SLP helped Hut 3.

Both CIS and SLP fused into one in Feb 44 in G Block in Bletchley Park and renamed SIXTA.

Intercept Co-ordination

Naval "Y"

In August 1940, Godfrey, DNI, proposed that Naval "Y" should be expanded to enable all German, Italian, French, Spanish, Russian and Japanese Naval traffic to be read. He advocated removal of departmental barriers so that all information from whatever source should be pooled.

Did not have much effect.

Cryptanalysis and W/T Intelligence.

Sub-committee on relations between cryptanalysis and W/T met July 41

Submissions sent in from Navy, Army and Air Force.

Conclusion: Cryptography and WTI are not mutually exclusive subjects but are inter-related throughout the process of extracting intelligence from interception and form one indivisible mechanism.

There is no basis for comparison, no clash of interests and there can be no apportionment of effort between the two.

In September 1942 Colonel Sayer produced a paper on "Interception, W/TI, Cryptography and Production of Intelligence", This re-iterated the findings of July 1941.

Finally Commander Travis in July 43 proposed, based on Sayer's paper,

Signals Interpretation

as an integrated whole:

"Who"? - Who sent it and to whom?
"Where"? - Where are they?
"What"? - What does it say?
The previous mistakes:- (1) "What" is held to be cryptography while "Who" and "Where" are held to be TA or W/T I

(2) "Who" and "Where" are treated as intelligence sources independent of "What"

All should be embraced under SIGINT


This page was originally created by the late Tony Sale
the original curator of the Bletchley Park Museum