by Alastair Denniston- Code-breaking From Room 40 to Berkeley Street
An important source of material is intercept station at Mauritius, which sends it only code or cipher material. Plain language stuff taken down at Mauritius is apparently translated there and sent directly to Economic Warfare Ministry. What Berkeley gets from Mauritius is mainly in commercial codes, and 80 per cent of stuff of both MITS is picked up there. Much of it involves Indo Chinese peninsula but it covers all areas of interest to us, perhaps twenty-five per cent of material is worth circulating. The two Navy Captains were very vague of figures, but they finally produced traffic register covering the four syllable MIT form, for periods from last August to this May 15, and I counted up four hundred fifty five recorded messages. Could not count number circulated, but my guess would be around one fourth.
Third source of JIG commercial is what is picked up in England, mainly stuff from Continental points. Output of JIG Commercial Section goes in longhand to Alford House, where ‘Commercial Section’ of Berkeley Street is housed.
The material is typed up and routed to consumers. Taylor will give you in later message important facts re Alford House. Note that principal consumer, Economic Warfare Ministry, gets not only what is produced in White section but also what comes from New Delhi, where as you know considerable work is done on commercial stuff, and also whatever Mauritius turns out from its plain text interceptions. End of story re JIG at Berkeley Street.
Finally got two letters
Note to code clerk: after third, repeat, in part seven text should read ‘output’ of JIG Commercial Section goes in longhand to Aldford House, where ‘Commercial Section’ of Berkeley Street is housed etc.
***
Continuing with Denniston’s distribution and reference section: this section keeps indexes of all circulated material.
This work is done in longhand, mainly by the two girls, but it may interest Huddleston to know that Foreign Office officials and Cambridge professors, unlike Harvard Law School graduates, are not above putting pen to paper in order to produce good indexes. I have examined a number of index cards and hereby pronounce that a good workable index can be maintained in longhand and that it would not be any great intellectual feat, though it would take time, to devise system under which what our girls do not now index could be indexed in our own geographical sections by men who follow the material and without occupying many of their minutes during any day.
***
One of the indexes is a name file of all name references in special material, in which important events in lives of senders of messages are noted and also all references in messages to individuals, on cards under their names. In cases like those of our favorite oriental author* [FN Hirohashi, Japanese ambassador in Berlin] his trips and important interviews are noted. Collateral information is also noted, in much the way Durant’s section … job, though job is keyed on lower scale and they have advantage of working with file that goes back before last war. They now have 19,000 names in file, of course after eliminating those who are permanently dead. Subject indexes are kept, based not only on Berkeley Street material but also on cables of various government officers.
By far the largest subdivision in index is Treaties and under it, by countries, are set forth all details about treaties and negotiations for treaties among various governments.
By all details I mean dates and important subjects dealt with, but of course cards do not attempt to cover any whole subject further that is necessary to refer index user to underlying material. Index cards give references by dates and serial numbers to source material. Next to treaties the largest headings are ‘Conventions – Economic’ and conferences. While it may seem like going into too great detail I believe that it would be useful for our people to have exact idea of subjects considered important index headings, and I therefore set them forth:
Air Associations and Federations; Banks; Budgets; Churches; Commissions and Committees; Communists; Anarchists; Espionage Plots Outrages etc (we were told confidentially that this elaborate heading is really designed to cover subject of enemy spy organizations in various countries); Concessions and Contracts; Conventions; Coups D’Etat; Ciphers; Debts; Delegations and Missions; Disarmament (including sub-headings for disarmament conferences); Economic (under which subject headings are almost all names of commodities, with oil comprising by far the largest amount of material, but with some non commodity subjects such as ‘mergers’); Economic Number Two under which information is indexed by countries; Exhibitions; Expeditions; Finance; Frontiers and Frontier Incidents; Jews; League of Nations; Loans; Mandates; Military and Military Missions; Naval; Opium; Press; Propaganda; Railways; Refugees; Revolutions, Plots etc; Shipping; Slavery; Supply of Munitions; Tariffs and Customs; Telegraph, Telephone Lines and Wireless; Traffic in Arms and Trials.
Please have Huddleston meditate in more prayerful attitude than usual on lessons to be learned from these headings.
Note that under Treaties headings they put not only treaties and negotiations as such but also all relations between governments not otherwise classifiable. Whole impression is that, like everything else that these very practical people do, index is designed to serve its purpose as neatly as can be achieved with minimum waste of effort. Note also that, while they try to keep track of all material for sake of whatever information readily available in indexes, their primary purpose is to supply crypt people with preference file in which they may locate quickly whatever information may be helpful to them in solution work.
Another important reference item is what they call the Cabinet Book. This is [a] record of important officials in various governments throughout [the] world, but it is very different thing mechanically from what Durant keeps. It is big ledger with subdivisions by countries in alphabetical order. When you turn to pages devoted to any particular country you find clippings and longhand notes giving names of cabinet officials and elected officers in order of dates, and you may trace through pages for any country and see very quickly what governments it has had for years covered by book. It is all somewhat disorderly in appearance, but when you examine it you find that, what with clippings from Times and Telegraph and longhand notations, there is a surprising amount of information in a form where it is quickly usable.
They also maintain a file of ships referred to in material, but here you may doff my hat to brother Snow, because this is the only thing I have seen in England in this general field where we have done a better job. One of the most interesting records that they keep, and from what they say one of [the] most useful, is what they call the Diary. This is a very large bound book in which each day is assigned two opposite pages, on which notations are made in longhand of all the events of that day that might be interesting to one who is working with material that might involve such events. The important subjects are mainly country classifications, and for every day the information about any country or other subject classification is set down at the same place on either the left hand or right hand page. For instance everything about Turkey is always to be found, say, in column five on the left hand page, three quarters down the column, so that a crypt man who wanted to know what was going on in Turkey over a period of two weeks last year can see at a glance what was happening in Turkey everyday, as gleaned from intercept material, government communications, or news reports in [the] London Times or Telegraph, which are the two newspaper sources that are relied on. Note that these two newspapers are clipped daily for all news of possible diplomatic importance for each country and the clippings put in country file with intercept material and government cables, all in order of dates.
Nothing is filed twice. If an item affects Russia and Japan it is filed under either Russia or Japan but not both. According to rules of thumb evolved over the years they decide which country; but if the same subject involves communications between the opposite Ambassador and his country, they file it under that country.
You can see how little they permit duplication of effort and building up of cumbersome files.
Incidentally, coming back to the Diary, le
t it also be noted that Diary contains after each item entered the reference to where full information may be found in the various sources on which it is based. Another file maintained is the Who’s Who file, which appears to consist almost entirely of Foreign Office printed bulletins which identify diplomats in various parts of the world and other important personages. It seems that British diplomats report periodically to [the] Foreign Office on all important diplomatic and local personalities in their respective countries, giving not what might be called official information but also whatever else may contribute to Foreign Officers store of useful knowledge. For example, of one diplomat it was stated that he had a youngish wife whose myopia required her to wear very thick glasses and of another it was intimated that his wife was on the frivolous side and ‘is not interested in the community problems on which her husband works so indefatigably’. I mention all this as further evidence to the thoroughness with which these Britishers do their jobs.
[The] same section also keeps reference library for use of Berkeley Street personnel. Among books that they have published weekly which deserve note is a manual of current events, known as Kensing’s Contemporary Archives, subtitle Weekly Diary of Current Events. This is published in England and covers all events of note in the world. I understand that it is difficult to get a subscription to it now, but please ascertain whether Arlington and Gee Two get it, because I think they should if they do not now get it. There is a library of Who’s Who and various reference books of the sort that you would expect, and where they are out of date and more recent editions are now available the staff annotates the books as new information comes in. For instance the German encyclopedia of newspapers of the world Handbuch der Weltpress, compiled in 1934, has been kept up to date by staff following all information about newspapers in various countries. Staff seemed to feel that one of their most useful books was Annual Pontifico, the Vatican yearbook, though it seemed to me to be no more than a catalog of Vatican representatives throughout the world.
One very interesting set of books is a series that is being published by the ‘Naval Intelligence Section’ of [the] Admiralty about various countries of the world. This was started in nineteen forty two. I had brief look at books on France and [the] Iberian Peninsula and they appear to cover every conceivable subject that would be interesting to one interested in the particular countries. They are called ‘Geographical Handbook Series’ and if Gee Two does not have them immediate steps should be taken to get on [the] circulation list. They are not to be sold publicly but are for use within British Government. However Earnshaw-Smith said he was sure that any United States agency could get on [the] list if it wanted to. Another excellent reference book that Gee Two should take steps to get if they are not getting it is an annual publication of the Empire Parliamentary Association called ‘Report on Foreign Affairs’. This gives for each country of [the] world a chronology of all the important events during the year. It has a good index and would make [an] excellent reference book, especially for Arlington. Another publication that [I] noticed, and I think Foreign Office gets it out every so often, is Q Book of Geographical Names. They say that it is much used in identifying places referred to in messages.
Smith’s section also receives foreign language press of various countries and (this is interesting) sends it to interested sections where it is read by crypt people or translators in order that they may keep up with developments in the respective countries. They do not cover German press, which is very thoroughly covered by other Government agencies, but they read at least one newspaper from principal cities in each other country in which they are interested. Note: I now have figure for present message numbers in Denniston’s outfit to wit 117,000 odd, indicating that in last fifteen months some 17,000 messages have been circulated.
Continuing: in addition to files and indexes kept up by distribution and reference section, each geographical section maintains whatever files it may think helpful. We did not get a chance to see what files are kept, but have impression that long organized sections have built up [a] substantial amount of reference material.
All foregoing may seem to you too detailed to justify use of cables, but we are anxious to give you as full a picture as possible in order that you and Corderman may ask any questions that occur to you or suggest what further lines of investigation we should pursue.
[The] whole key to this British operation lies in the infinite pains which they take with the files while never losing sight of their very practical objectives.
***
Herein report on Near Eastern Section of Denniston’s show. Understand you have had through Johnson full information on cryptographic method which I am including cryptographic information given me for what it is worth. Section comprises eighteen people of whom seven are linguists and do the book breaking and translating, ten are decoders and clerical, and one is keybreaker for Turkish traffic. Section handles Turkey, Saudi Arabia, Iraq, Egypt, Iran, Afghanistan, [and] as minor sidelines Syria, Transjordan, Nepal and Ethiopia. Head of Section Dr Thacker is Professor of Semitic languages at Durham University and can do all three languages involved to wit Turkish, Persian and Arabic but specializes in Arabic PYIOT KISH. Most important branch comprises one keybreaker, five decoders and clerks and three translators and bookbreakers; Thornhill, a clergyman and former pupil of Thacker’s at Durham has learned Turkish in last two years, Simpson, geographer at University of London has learned Turkish since war began, Gungry learned Turkish recently at School of Oriental Studies formerly in London now in Cambridge.
Persian section three decoders and two translators; Humphreys, elderly and only permanent employee, was Archivist [at the] British Embassy Persia for many years and knows other Near Eastern languages as well as Hindustani, Russian and many others, Boyle learned Persian at [the] School of Oriental Studies.
Arabic section two decoders and two translators, Thacker and Dr Lewis lectured in Islamic history School for Oriental Studies.
Turkey uses Turkish except for one code book in French for treaties etc, Saudi Arabia and Iraq use Arabic, Egypt uses French, Iran and Afghanistan use Persian, Syria has not started code yet because has no diplomatic posts but will be in Arabic, Transjordan has no diplomats but Emi Abdullah has private cipher in Arabic which British read but Emir has not yet said anything worth circulating.
Nepal uses simple substitution cipher in English and three messages have been circulated in last eighteen months, Ethiopia has five figure book for traffic between London and Adis Ababa, British have twelve messages but cannot read them.
Turkish section receives fifteen to thirty messages per day, slightly more to Ankara than from Ankara, and circulates from two to twelve per day – perhaps six would be a fair average; majority of circulated messages are to not from Ankara. Principal circuits covered are London, Kuibyshev, Teheran, Budapest, Lisbon, Rome, Madrid, Stockholm also some Berlin, Washington, Vichy, very little South America, Helsinki, Bucharest, Sofia, good coverage Tokyo but very little traffic. Two principal secret diplomatic code books introduced [in the] middle of nineteen forty one called CANKARA and ISMET INONU, both ten thousand group two part codes.
***
In order to comply with instructions in your 2734, I will have to discontinue all discussions with Denniston, since there are no intelligence problems to discuss except those which have to do with crypt and traffic exchanges. Since Denniston is willing to let somebody here read, and if interesting to forward to Washington, all decodes not in special reserved series, the only questions that have been involved have related to whether the Government of the United States was willing to take particular classes of intelligence that way, or whether as to some or all of them it wanted to process the material itself. I will tell Denniston that these problems will be worked out directly by Corderman or perhaps through Taylor. It is disheartening, however, to work hard on getting the facts for Arlington and to get this kind of response.
While you prepare Washington Monument for appropriate part of
my anatomy, I will say flatly that Arlington does not (repeat does not) understand the problems involved in crypt and traffic exchanges. For instance, it has just thrown Denniston into a state of bafflement by asking for Iraqui [sic] keys. This request would be roughly equivalent – if we still had the Philippines – to the British asking Arlington for Philippine keys, since Iraq is not only in reality part of the British Empire but is the cornerstone, because of its oil, on which the whole Middle East and Eastern Mediterranean situation rests. Denniston, however, wanted to put Arlington in a position to read whatever Iraq stuff it might intercept, insofar as he could do it without going to the Foreign Office for specific authority. Hence he authorized his people to explain the Iraq system to Taylor, and Taylor duly passed the information along, and Denniston felt that he had met Arlington’s demands without having to create an issue here. If questioned by the Foreign Office, he could always say that Iraq uses simple substitution which any cryptanalyst who had an Arabic linguist at hand could solve in half an hour, and so he had really only told Arlington what it could have found out for itself in a very short time.
Now however, Arlington turns this matter into a major issue by a formal request for keys, forcing Denniston either to put himself on record as refusing something that Arlington specifically asked for or to create an issue by asking the Foreign Office for permission. If Denniston is right about the difficulty of solving the Iraq system, what sense does all this make?
Some of these Britishers, I think perhaps including Denniston, think we in Washington are a little bit at loose ends, and what do you suppose I would think, if my thoughts made any difference, when I get three cables in short succession, one of which says that cables were omitted because nobody asked for them, the second says they were omitted because the parties were exchanging them through censorship, and the third says they were omitted by common consent. Jesus. Taylor has been talking to Denniston about the Iraq matter and he is holding it in abeyance until Taylor gets back and has further talks, after which Taylor will communicate with Corderman.
