Kim Philby (11 page)

Over the next few weeks, I began to learn a little more about the service I had joined. The headquarters sections of SIS, apart from Section V, were mostly located in Broadway Buildings; we used the term ‘Broadway’ to mean ‘SIS apart from Section V’. In pre-war days, Section V, as its symbol indicates, had been more or less on the same footing as the Broadway Sections I, II, III, IV and VI, whose job was to circulate intelligence obtained from SIS stations abroad to various Whitehall departments, and to relay the requirements of those departments to the stations; Section V’s particular customer was MI5. The circulation sections were not purely postboxes, since they were supposed to know what was worth circulating to Whitehall, how to grade the source, and how to interpret Whitehall’s demands in terms of what it was practicable to put to the station. But few staff were necessary, and even to the end of the war the other five circulating sections never consisted of more than a handful of officers each; whereas Section V, as I have said, already numbered about twenty officers, and by 1945 must have had well over a hundred at home and abroad.

The reason for the difference lay in the nature of counter-espionage intelligence. In the study of enemy intelligence and sabotage activities, there was no clear division between foreign and British territory. Most counter-espionage targets lay abroad: the headquarters organisations of the enemy organisations, their stations in occupied and neutral countries, their recruitment of agents and dispatch to British or other territory. A great deal of enemy intelligence and subversive activity might never involve
British territory at all: for example, the watch from neutral countries on British ships and convoys; sabotage of British ships in neutral ports; the dispatch of enemy agents to countries that were allied or friendly to us; counter-espionage activities by
Abwehr
III against SIS or SOE or our allies; German intelligence relations with neutral governments. To this must be added what became much more important later, the kind of shorter-range secret intelligence operations that take place when two armies are opposed in the field: agents left behind as the enemy advances, line-crossing operations and so on.

The territorial division of the counter-espionage function between MI5 and Section V was therefore somewhat unreal, and a demarcation dispute was inevitable. There were three sensible solutions: first, that MI5 should take over the whole of counter-espionage abroad as well as at home and where necessary set up its own stations and communications, with all that that involved in expert knowledge of other countries and their languages; second, that Section V should expand within the existing demarcations of function and expertise; third, that there should be some kind of amalgamation of MI5 and Section V, or a pooling of resources and tasks. By the time I arrived, I think the principle had really been decided in favour of the second solution; at any rate, Section V was expanding fast. This was due partly to the logic of the thing, though the issues were fairly evenly balanced, but perhaps more to the determination of one man: Felix Cowgill, the head of Section V. It was Felix who fought off the MI5 challenge and, equally important, set about getting the necessary staff. Plenty of battles still lay ahead. But in my own view MI5 were not great empire builders at heart, as empire
builders go; they wanted to do their own job, but realised that increasingly it depended on what was happening abroad. If we were able to give them the necessary service, they seemed willing in the last resort to see the division of functions left unchanged.

In late 1941 the difference in resources between MI5 and Section V was still very great. MI5 had been built up rapidly in 1939–40 when people believed in fifth columns, and when thousands of refugees were being interned. It was a time when stories of German parachutists dressed as nuns, Charlie Kunz the bandleader passing information to the enemy through broadcast foxtrots, and Lord Haw-Haw’s infallible knowledge of when the town hall clock had stopped found easy credence. Although the larger lunacies had been discounted by the end of 1941, MI5 still had a swollen staff and were able to write four-page letters to us about mildly suspicious sentences appearing in a letter examined by censorship, whereas we were desperately pressed to deal even briefly with all the hard overseas intelligence that was beginning to be thrown up.

The best of our sources was already the ISOS messages, even though at this time it covered only a small proportion of German intelligence telegrams. At the end of the year there came a radical change. I was staying down in Dorset for Christmas when Kim rang me in the guarded language used on the open telephone: ‘You remember the possibility we discussed that you might have a tremendous increase of work? Well, you have.’ I returned to St Albans to find the first decoded messages from the machine cypher links on my desk. Within a few weeks I was having to study anything up to 120 a day: nearly all were
Abwehr
, a few SD. (I should mention that the term ‘ISOS’ strictly referred only to
the hand cypher material, while the machine cypher material was given the code-name ISK;
^†
but unofficially ISOS continued to be used as a generic term for the lot, and I shall so use it here.)

Although ISOS was distributed also to MI5, to specialised sections in the three forces ministries and to a few other people, it had been agreed by the chief that the prime responsibility for action on it lay with Section V. This responsibility was in turn delegated to the appropriate geographical subsection. In practice most of the opportunities for doing anything with the information, as opposed to merely studying it, lay with Vd. This is because Spain and Portugal were neutral European countries in which both the Germans and SIS maintained stations, and which gave the Germans their best outlet to Allied territory. In addition the harbour and Strait of Gibraltar were themselves important intelligence and sabotage targets. Since Spain and to a lesser extent Portugal were friendly to Germany, conditions were highly favourable for the
Abwehr
. But the situation also gave us opportunities for counter-action. The other neutral European countries were on a different footing: Sweden and Switzerland were largely cut off from the west, and Turkey was somewhat remote.

While ISOS was potentially of enormous value to us, the handling of it was hemmed about with difficulties. There was nothing like a 100 per cent ‘take’ of cyphered messages.
Abwehr
radio transmissions might be missed by the British interceptors, or received in a garbled state owing to bad atmospheric conditions.
Of those intercepted, the cyphers might be uncrackable for short or long periods. In addition, a great deal of vital information would be sent through the diplomatic bags between the German consulates and the embassies at Madrid and Lisbon, or between the embassies and Berlin; the bags were of course inaccessible to us. Frequent visits by Berlin officers to the peninsula, or by local
Abwehr
officers to Berlin, and the use of the telephone, all helped to reduce the
Abwehr
’s dependence on cypher telegrams. The messages themselves were often couched in deliberately obscure language, and both officers and agents were normally referred to by cover-names (not to be confused with aliases, i.e. the false names they might be using in public).

More important, our first consideration had to be the security of the source. It was, of course, absolutely vital not to let the Germans even suspect we were reading their cyphers. More than ever this became necessary when the machine cyphers were broken, since these, we were given to understand, operated on the same principle as those used by the German armed forces for their high-grade cypher messages; much has now been published about the Ultra source and its incomparable value to the British and American governments and armed forces. It was impressed upon us that, whatever else we might do or fail to do, the one crime we must never commit was to endanger the ISOS source. I should make it clear that although the risk was much more serious in the case of machine material than with the lower grades of cypher, we treated all kinds with equal care. We did not want to give the Germans the slightest reason to begin examining their cypher security.

As the ISOS officer in Vd it was for me to study all the intercepts affecting our area; piece together and try to make sense of
the dozens of different German operations and activities that were going on simultaneously; identify the large cast of characters, mostly referred to only by cover-names; pursue enquiries abroad with a stream of telegrams and letters to our stations in the peninsula, in Tangier and sometimes elsewhere; keep in close touch with other departments concerned, particularly MI5 over anything that involved or might involve British territory; and gradually build up an order of battle of the many German intelligence stations and sections in the area. The work involved study not only of the ISOS but also of voluminous reports from the stations, interrogation reports from MI5, back files in the SIS registry and much besides. For the first fifteen months or so I had no assistant. Kim followed the material as well as he could, but one could not learn much from merely skimming through it. The most important message of the day might be something like ‘Your 129. Yes.’ Who sent a message and to whom was often more important than the message itself, and could be a valuable indication of the nature of a German intelligence operation, or even of vital changes in
Abwehr
or SD structure and hierarchy.

Most of the messages did not make much sense by themselves. This had its advantages. The chief had a tiresome assistant who was supposed to run through the material for anything that ought to be brought to the attention of his master. Before long he was reduced to ringing me only when something like a titled name caught his eye. Otherwise he left me in peace.

The very heart of the whole problem was to marry up ISOS with what was called ground information, i.e. reports from our stations abroad and other non-cryptographic information. ISOS, purely by itself, was of little practical use. It could tell you a good deal about headquarters organisation, chains of command and
communication, levels of activity, and, within limits, intelligence operations; and it was very useful towards the end of the war for the picture it gave of the collapse of the politically unreliable
Abwehr
command and its takeover by the SD. But by itself it did not usually tell you the real names of
Abwehr
officers and agents, and was liable to give a very incomplete or misleading impression of what was going on. It was also easy, particularly in the early stages, to misinterpret the material. Kim relates in his book the story of the ‘ORKI companions’, a fabulous nonsense which was in progress when I arrived. The story has, I think, more significance than has previously been realised, but it would interrupt the narrative unduly to go into all the details, which are given in a footnote to this chapter.

An important part of the business was identification of those appearing in ISOS under cover-names. Often it was simple. A message would say, perhaps, that
HERMANO
was flying from Madrid to Barcelona or Berlin on such-and-such a date. Our stations could normally supply passenger manifests for Lufthansa or Iberia airlines as a matter of course. When the appropriate list came in, usually with surnames only and often misspelt, you scanned it for likely names. Perhaps the list for that day would be missing, or incomplete, or
HERMANO
might have decided instead to go by train or car. But with luck there would be one or two likely candidates on the list. Unless there was urgent need, you then waited for another announced journey, or it might be a hotel booking, for with the larger hotels we could usually get hold of the nightly guest lists. If one of the likely names appeared again, and if (as usually happened) you had other ground information on him, the identification was probably safe.

Sometimes the situation was much more urgent, and the
station had to be brought in to make investigations within the limits imposed by the overriding need not to compromise the ISOS source. One of the first such cases, in early 1942, concerned an agent with the cover-name pascal, of whom we knew little except that he had stayed at a small hotel in the Barcelona area on or around a particular date, and was being sent, at any rate in the first instance, to South America. The hotel was not one we had regular tabs on. Then came an ISOS message saying that the
Abwehr
had bought a ticket for
PASCAL
on the SS
Marqués de Comillas,
at a cost of (let us say) 14,875 pesetas. He was about to sail. Our Madrid station managed, first, to get guest lists for the obscure hotel, and secondly to pin the ticket cost down to one or other of six people. This was more difficult that it might sound, since there were apparently several options and extras contributing to the total ticket cost of each of the hundreds of passengers. One name appeared both on the hotel list and among the six passengers. Because of the need to exercise extreme discretion the enquiries had taken several days, and the boat was on the point of docking in Trinidad when the identification was made. We and MI5 sent Trinidad a Most Immediate signal, and poor pascal, who probably never even knew what his ticket had cost, was taken off the boat for interrogation. He was carrying compromising equipment and soon confessed. For the rest of the war he was interned in Britain. Several of his companions in internment were caught in the same kind of way.

Many identifications were difficult. Some of the characters never travelled, never moved outside the metaphysical ISOS world. It was a long time before we were able to put a name to felipe, of the
Abwehr
headquarters in Madrid, who was of particular interest because he specialised in sending agents to
England. Cover-names of staff were often changed, so that you had to identify a cover-name with its successor before you could proceed. In addition, most of the more important
Abwehr
station officers were living under aliases in Spain or Portugal, and could not readily be identified with previous records from other countries. Some of our identifications would remain on the ‘tentative’ list for months before the final clincher arrived. Occasionally it was child’s play: for instance, a message from Seville to Madrid containing a cover-name might be passed on to Berlin with the real name substituted. In the course of two years or so several hundred German intelligence staff, agents and contacts in Spain, Portugal and north-west Africa, appearing under cover-names in ISOS, were finally identified with ‘real’ people.