Gwen: Gone Fishing

 

(FISH is the name given to encryption systems attached to teleprinters used by the German Armed Forces during the Second World War. Writing for an audience of fellow Siginters, Gwen takes a fair bit for granted, but gives a clear picture of what the day-to-day work of an analyst could be like between 1939 and 1989.)

When a signals officer at one end of a FISH link had a message or messages to send, he would tell the other end by using the Q signal QEP followed by a number. That number must have indicated from lists how the machines at both ends were to be set. Immediately after this came a stream of encoded text. The transmitting operator usually checked that his transmission was being clearly received by the other end by putting a few words in German chat. He might use a phrase denoting some grouse he had, like being cooped up in a metal caravan. Or he might use some Nazi phrase (Heil Hitler was common), which hinted at how deeply the dreadful madness had affected Germans for more than 20 years. But thereafter he went on to send the actual messages at an automatic speed, because it came from a pre-recorded tape.

I am no cryptanalyst and what I gleaned from liaison with the Testery came from a sort of tacit osmosis, rather than from direct question and answer, and besides, it is more than 70 years since I was involved. But it is clear that any hint of what the plain text of a message might be, or of any preliminary chat come to that, could be of help to a cryptanalyst. What Yvonne had seized upon in the supposedly useless intercept logs were the receipts sent by the other end of the link at various intervals, maybe 15 minutes, in the form RR15 0801, RR16 0810, RR25 0815, RR17 0817. One could then tell in which QEP or QEPs from the other end the messages had been sent and at what intervals the cryptanalyst could expect the word 'Spruchnummer' (message number) to appear, followed by the number of words in the message and the name of the addressee. The cryptanalysts already knew the last two figures of the message number, that is 15, 16, 25 and 17. Furthermore, the fact of the out of order 25 probably meant that it had a higher priority than the others. And the introduction, or preamble as we called it, of the message would give an indication of this, with the words 'urgent' or 'very urgent' (sehr dringend). And what is more, the correct identification of each seemingly anonymous outstation could he guaranteed by Traffic Analysis (TA) methods and allowed the addressee of the message to be identified. Or occasionally the addressee might be that of a subordinate formation. If that was so, it might have been possible to trace a re-encoding in Enigma on a subordinate morse network.

It was the extension of these very simple principles (guaranteed link continuity from day to day and hence the addressees of most QEPs) which boosted the efforts of the cryptanalysts. The biggest single advance was the recognition that about the same time every day the High Command sent out on all the links to Army Groups a longish message (perhaps extending over several QEPs), which were receipted on, let us say, BREAM, JELLYFISH and GURNARD, in the form 12/24, 19/31 or 37/49: the second dinome being exactly 12 more than the first. This turned out to be a surefire identifier of the High Command daily report (Tagesmeldung), thereby giving a near identical, but not completely identical, message in different ciphers on different links, but with the vital difference that the message numbers and the addressees were different.

This constituted a vast difference from the early days after 'The Brig' (Brigadier Tiltman) broke a very long message successfully by hand after weeks of work. And a man named Tutte worked out the cog numbers on the wheel and how you might identify their settings. It was also an advance on the early stages when the Testery were guessing at a bit of chat or at the addressee of a message and dragging the German text by hand or on the punningly named machine 'The Dragon' to try to find a statistically significant piece of cipher, which they handed over to the Newmanry. The Newmanry developed machines, of which much has been written, which then ran at huge speed to give the settings of five of the 10 encipherment wheels, after which the thing went back to the Testery to be completely decrypted and turned into German.

I would also like to introduce a few more of those who worked for David Rex and Yvonne. Probably the most intellectual, with an incisive mind, was jolie-laide Sergeant Mary, already halfway through her university course which led to a first class degree in Theology. Sergeant Lesley Stuart Taylor, ex-head girl of Cheltenham Ladies' College and renowned in post-war Cheltenham in various guises, wore a uniform of beautiful khaki cloth, because she was so tall that she had to be given the uniform of the World War I ATS Commandant. Her office party piece was reciting all the JELLYFISH frequencies without hesitation, deviation etc etc. There was Jeanne Cammaerts, daughter of a Belgian professor and a French diseuse and sister of the famous Maquis hero Francis (about whom we learned after the war), and herself a very good actress who played Eliza in the BP drama group production of Pygmalion. There was also a new man, a half Rumanian Jew, who of course had to have Saturday as his day off, and then there was Staff Sergeant Pauline, who always had Monday off. In winter she hunted, in summer she crewed for her father in dinghy races at Burnham on Crouch and their names often appeared as winners in The Times. Following a failed marriage, she brought up two sons by various entrepreneurial exploits and between the ages of 50 and 78 organised and led pony trekking holidays in the Andes.

Early in 1945 we were told to expect a captain whom we immediately and lastingly named Python, after the scheme for early return to the UK of officers who had fought abroad for many years. I was deputed to train him in the arcane ways of our section. l don't think he was very interested in that, but he was quite a cheerful companion. One day he and I took a walk through some meadows and woods. We had perched on a five-bar gate, when we heard some wonderful male-voice singing approaching us. Round a bend in the lane came a whole lot of Italian prisoners of war escorted by one British corporal. Of course the minute they saw a man and a girl in shirt-sleeve order sitting on a gate, even though we were only munching bread and cheese, they started shouting out what to me were undecipherable comments, but their meaning was clear enough. They were most taken aback when Ronald let loose a stream of very colloquial Italian, which was giving much more than he got. He had spent most of his childhood in Italy. where his parents kept an hotel.

There was another visit to Knockholt somewhere along the line when V2s rather than the earlier V1s were descending around the place, but we just had to carry on in a blasé way because a V2 either hit you out of the blue or sailed on in silence.

The end of the line

As the final battles raged, there was a great deal of activity on the FISH links to Army Group HQ and the parallel links to Luftwaffe HQ, though the latter had a far lower priority rating, because of the enormous amount of information on air-force mattes available from Enigma traffic. There were also some traces of Naval Fish links in the last days of the war. but these were so brief that I don't think they even had names given to them. VE Day of course meant the end of Fish and the dispersal of our group. Most of the ATS were posted to other units, where almost all of them became officers before their demob in mid-1946. Yvonne and I remained to tidy up.

Commander Travis issued a thank you to all BP personnel, together with an exhortation to remember our oath of secrecy, especially since "it may be that at some future time we may have to use the same methods". For Yvonne and me "some future time" turned out to be next week, because we found ourselves, and Knockholt, studying 2-channel Baudot links, technically just the same as FISH, but serving another would-be world conquering power. That, in essence, was the beginning of my post-war career in GCHQ and belongs to that story.

There was one day in late 1945 when I sat the entrance exam, specially designed for service people, for Lady Margaret Hall, Oxford, and went up in Michaelmas term 1946 to read English Language and Literature.

At last we have some illustration and a scrap of conversation as Alice in Wonderland thought all books should have. The photograph on the left shows me as a newly commissioned Subaltern and on the right I am on the steps of University Schools in Oxford, immediately after the most gruelling interview of my life — a 40-minute viva. The worst moments were as follows:

Miss Lascelles, somewhat disdainfully: "Miss Herbert. you mentioned in your written paper the political friends of Pope. Who were they?"

Me: total silence

The Chairman (the Merton Professor of English):"Give Miss Herbert a drink of water. Professor Rice-Oxley, she can use your glass."

Thankfully I was then asked some questions about Beowulf, the long Old English poem of which I am very fond. The first word of it, which will serve as the last word of this essay, is the attention-grabbing "Hwaet".