Chapter 5: Some Observations of Western Communications Intelligence and the Holocaust
In the preface to this Guide, we asked two major questions: What did the Allied communications intelligence agencies learn about the Holocaust and when was this intelligence known? In the preceding chapters, we have described briefly the operations of the communications intelligence system of the GC&CS and the SIS. We have described the available archival collections at NARA, the PRO, and other places. Finally, we surveyed some of the hundreds of reports produced by the cryptologic agencies that bear on the history of the Holocaust. Using this information, we should be able to answer both questions. At the same time, we can add some observations about other aspects of the archival records, specifically their limitations, how they came to be organized as they did, and their potential uses for today's scholars and researchers.
There are two points that need to be understood that relate to these two questions before we consider them. First, because the United States came into the European war only after December 1941, much of the following discussion applies to the activities and personnel of the GC&CS. The SIS did not establish a COMINT mission for the European Theater that was fully separate from the British. The SIS collaborated with the British on many European Axis military cryptologic efforts, but in generally subsidiary or complementary roles. After the agreements of 1943, the Americans relied on the British for much COMINT about Europe.1 The SIS did not receive any German Police decrypts or SS messages until after the BRUSA exchange. At that point in the war, the intelligence about the Holocaust gleaned from those sources was meager because of specific German security measures implemented gradually (but never completely effective) from late 1941 into 1943. The SIS developed some information about the Holocaust from its own intercept, mostly derived from the messages of the diplomatic missions of several countries. But this intelligence generally consisted of eyewitness reporting of events by diplomats on the scene or information they gathered from interviews or other sources.
Secondly, it is important also to remember that the GC&CS developed only one source of intelligence — communications intelligence — during the war. What the GC&CS gathered from its intercepts and the subsequent decrypts was forwarded up and across the echelons of the British intelligence, defense, and security establishments. The British (and the Americans) utilized many other sources of intelligence besides what they pulled from Axis signals. These included classic espionage agent sources, aerial photography, collection of intelligence by technical means such as radar, captured equipments and papers, reports from diplomatic missions, debriefs of prisoners, defectors, and refugees, and monitoring of foreign radio broadcasts. During the war, the Western Allies used intelligence from all these sources to formulate campaign and battle plans, political and diplomatic strategies, and to follow the course of the war in other theaters such as the Russian front. In some military campaigns, such as in North Africa, Ultra was the more decisive source of intelligence. Yet in others it was less critical to outcomes of the Battle for Britain in 1940 and the U-boat campaign. We need to keep in mind this mix of intelligence sources and the contribution of each to the Western Allied knowledge of the Holocaust. There were other sources of intelligence about what the Nazis were doing and some of these proved to be superior to COMINT.
What Was Known from Western COMINT
The archived wartime records of the British and American cryptologic agencies hold a number of translations and decryptions that bear on the history of the Holocaust. Due to limited research, only an approximate estimate can be made, and that ranges from 700 to 900 individual translations and decrypts. The information in them covers the entire period and scope of the Nazi effort to eliminate Europe of its Jews and other "undesirable" groups. Yet, while numerous and diverse, when taken together these records do not constitute a complete narrative about the event. For many of the individual dramas of the Holocaust, such as the roundup of Jews in the Netherlands, the escape of the Danish Jews, and the deportation of foreign Jews from France, the information from COMINT records is fragmentary or episodic at best.
A large number of the intercepts were originated by the Nazi organizations that were directly responsible for carrying out the Final Solution. These organizations included the various formations and commands of the German Police, the elements of the SS charged with running the various camps and supervising the roundup of Jews in occupied countries, and the SD, which actively worked along with the police and SS. The messages sent by these organizations form a contemporary chronicle of their activities. Other intercepts such as the diplomatic messages of Vichy France and the Sztojay and Szalasi regimes in Hungary contain the grotesque official rationales for the depredations inflicted on their victims. These particular intercepted messages stand as first-person indictments of those who committed the war crimes and those who aided the criminals.
Still, despite the incriminating information from the SIGINT, the information in the police and SS decrypts does not provide historians with a complete picture of the extent and nature of the Nazi program to eliminate Jews from Europe. For example, the arithmetic sum of information from the decrypts, measured in terms of reported deaths and general actions such as "liquidated inhabitants," although a grim accounting, did not give anything near the total deaths that the police units and einsatzgruppen caused. This was partly due to the Western inability to intercept and decrypt every message sent by the police or SS units. As such, much of the intercept came from the police and SS units operating with the German Army Groups Center and South. Very little was intercepted and decrypted from the police units operating in the Baltic area of operations.2 Also, the security mea-sure slowly implemented by the Police command eventually closed this pipeline of information.
The decrypts from the concentration and death camps that the British analysts reviewed have similar problems. These particular reports were periodic accounts, usually in the form of a monthly summary, of the day-to-day available slave labor population at a particular camp. Often, a camp's population report was broken down further by nationality or ethnic group. In some cases, the report was further refined with a statement of the totals from the beginning and end of each day. Absent from these reports was information about those groups who arrived at the camps and immediately were sent to the gas chambers. Why this latter information about the gassings was withheld from the camp reports is not clear. Quite possibly, the SS hierarchy wanted to maintain as much secrecy as possible about the extent of the Final Solution. Later German security precautions only added to the silence. The result was that analysts at GC&CS, using only SS and police decrypts, could not provide explicit intelligence about the mass extermination activities at Auschwitz/Birkenau and other such camps.3
Another large collection of intercepts consists of multinational diplomatic messages. Most of the intercepts were of messages from European diplomats stationed in various capitals who reported occasionally about the plight of the Jews. As mentioned earlier, the priority to collect these types of messages was never high, except for those from Japanese diplomats in Berlin and Rome. Reports by diplomats occasionally carried information about the roundup of Jews in a particular country. Only during the roundups and deportations in Vichy France in the summer and fall of 1942 and in Hungary during the late spring and fall of 1944 did the diplomatic traffic report in any depth about what was happening. In the latter case of Hungary, Allied and neutral interest in events there initially may have been related to the secret maneuvers by Budapest to abandon the Axis and sue for peace.
Still, the records of the diplomatic eyewitnesses or those who reported what had been discovered from their own inquiries, such as many of the intercepted diplomatic reports in essence were, provided a dramatic element to the narrative of the grim events that happened throughout Europe. Diplomats from neutral countries filed most of these reports. In many cables, their objectivity as observers often was challenged by the tragic and horrible nature of the events they reported. This is poignantly illustrated in the traffic sent by the Portuguese ambassador in Budapest. Other messages are a chronology of failed or futile efforts by neutral diplomats to organize or effect the rescue of some of the victims.
When the COMINT Agencies Knew about the Holocaust
For decades, an important issue for historians of the Holocaust has been to establish the point in time during the war when sufficient information from all sources reached the Allies that disclosed the full nature and extent of the German plans to eliminate the Jews. The critical time period encompasses the beginning of the massacres in the western Soviet Union, that is mid-1941, and runs to the middle or end of 1942 when the complex of death camps reached nearly full operation. But it is difficult to provide a simple answer to the second question of when the COMINT agencies knew about the Holocaust during this period. This is because any answer anticipates that, from such a date when signals intelligence was known about the massacres in Russia or the operations of the death camps, the Western cryptologic agencies could have discerned the full intent of the Nazi policy of extermination, the full extent of which became known as the Final Solution.
On the surface, the timeliness of communications intelligence — especially the German Police messages that were intercepted and decrypted shortly after the invasion of the Soviet Union began, and the SS concentration camp radio reports from as early as mid-1942 — suggests that this intelligence source could have warned of what would eventually befall Europe's Jews. However, there were two major obstacles that prevented communications intelligence from being the single "warning" factor.
The first obstacle was that the usefulness of the intercepted police and SS messages was undercut by two fundamental shortcomings in the nature of the intelligence itself. The most important of these was that these messages were intercepted after the programs of massacres and death camps had begun. The police actions in the western USSR had begun prior the intercept of the earliest police radio traffic about the massacres in the Soviet Union. The first intercept that mentioned any of the massacres probably was the 18 July 1941 radio message from the Police Regiment Center that reported the execution of 1,153 Jewish "plunderers" near the town of Slonim in Belorussia.4 This was nearly a month after Nazi Germany invaded the Soviet Union. Even more importantly, Nazi policy for the elimination of all undesired racial types during the invasion of the Soviet Union had been established back in March 1941. At that time, Hitler, in a meeting with military leaders, had established that the invasion would be a war of extermination. The arrangements to implement this policy were made in the next two months between the army and the SS.5 But nothing of this planning appeared in the intercept.
The SS messages about the concentration camps suffered from the same time lag. The messages sent by the concentration camp SS commanders and encrypted in the SS Enigma machine, were not intercepted and decrypted until early June 1942. This date was well after the death camp at Chelmo, Poland, had started operation in December 1941. It also was months after the beginning of operations of the Birkenau complex at Auschwitz in March 1942.6 The intercepts about the concentration and death camps also came well after the important conference at Wannsee in January 1942, which considered extending the Final Solution to all of occupied Europe.
The second shortcoming of the intercepts was that they contained fragmentary or unclear information that could be ambiguous to analysts who reviewed it. Therefore, the COMINT could be subject to differing interpretations regarding the full intent of the Nazi activities reported in the intercepted messages. The analysts that studied the communications intelligence could and did miss or misinterpret the significance of the information in them. Axis decrypts were filled with abbreviations, acronyms, and terms that the Allied intelligence analysts first had to understand. We can add to this mix of problems the Nazi policy of disguising references to various phases and activities of the Final Solution. Sometimes this program effectively obscured the information in the intercepts to the point where its importance went unrecognized. A good example of this was the German Police message of 11 January 1943 that listed the number of Jews and others gassed at four death camps in the previous year. The decrypted message referred to an "Einsatz Reinhardt" followed by a set of numbers for the last two weeks of 1942 and for the entire year. British codebreakers appear to have not recognized the reference and that the message was, in fact, a report on the number of victims gassed in the four death camps located in the General Government (southeastern Occupied Poland).
The second obstacle to effective use of Western COMINT was how the intelligence was received and assimilated by the analysts and officers at Bletchley Park and the rest of British intelligence structure who were familiar with the reports about the massacres, deportations, and concentration camps. Although historians are wary of making generalizations about the inner thoughts and attitudes of a population in any period, some observations about the historical context of European anti-Semitism could be useful if we are to understand, in any small way, how intelligence about the massacres and camps was received at the GC&CS.
The first thing to consider is that what the Nazis were doing was outside of the historical experience and imagination of most Europeans. Certainly the historical landscape of European anti-Semitism had been one marked with almost nineteen hundred years of unremitting ugliness — the pogroms, massacres, deportations, banishments, ghettoization, extortion, blood-laws, and auto-da-fes. But even this common heritage of brutality could hardly prepare people for what the Nazis planned and were working to fulfill an effort to eliminate Europe (and later, the world) all vestiges of Jewry. Not only were Jews targets, but also many other segments of the European population were to be eliminated — mentally and physically handicapped persons, Slavs, Poles, Romany, etc. This larger aim, too, was beyond the historical experience of Europe. And for the Nazis to carry out this plan meant that the old techniques of murder and starvation were not adequate. Hence the Nazis employed mass industrial techniques of slaughter to achieve what they wanted. Realizing that what the Nazis did was so horribly unique, could it be expected realistically that the analysts at Bletchley Park and elsewhere could have determined the totality of the Nazi plan, and the means to achieve it, based on the fragmentary evidence at hand from the intercepts? And could they have grasped the implications that the massacres in Russia and the camps were some of the means to achieve a Europe that was Judenfrei?
A second consideration is that, while many people may have received the news of the massacres and camps with horror, the truth is that much of the world's population, including many in the major Allied powers of the United States and Great Britain, held, to some degree, anti-Semitic sentiments. This observation may be unsavory to modern ears; it is, unfortunately, a historical fact. We have seen some allusions to it in the preceding pages of this guide. Most notably, there was the difficulty experienced by the United States when it came to saving some 5,000 Jewish orphans in France (See page 92). This was, in part, caused by the refusal by some in Congress to accept more Jews. Both President Roosevelt and Prime Minister Churchill were often hampered in their limited efforts to alleviate some of the suffering by the general anti-Semitic sentiment in both nations.7 Just how many British signals intelligence analysts, either individually or as a group, held this attitude is unknown.8 And how much it affected their reactions to the intelligence is likewise unknown. But it must be considered in any discussion about how the COMINT was received.
Whatever their beliefs or intentions, many people in the West, and some of them were in a position to know otherwise, simply could not comprehend or refused to believe the evidence of what was occurring in Nazi-occupied Europe.9 For example, the famous American historian, Arthur Schlesinger Jr., then an analyst in the Research and Analysis Branch of the OSS, remarked that, in the summer of 1944, even with the flood of reports about concentration camps, what was happening was interpreted as an "incremental increase in persecution rather than…extermination."10 And in December of 1944, John McCloy, the assistant secretary of war and a regular recipient of Ultra intelligence, asked Leon Kubotinski of the World Jewish Congress, "We are alone. Tell me the truth. Do you really believe all those horrible things happened?"11
The little evidence that historians of cryptology have to work with suggests that within the ranks and leadership of the GC&CS, there was a division of opinion about what the police decrypts augured regarding the extent and nature of what was happening to the Jews and other groups throughout Europe. Two examples illustrate this split. In September 1941, an intelligence analyst from Bletchley Park had attached this grim comment to a Police decrypt about massacres in Russia: "The tone of this message suggests that the word has gone out that a decrease in the population of Russia would be welcomed in high quarters and that the leaders of the three sectors [The commanders of the police units assigned to the three German Army Groups invading the Soviet Union] stand somewhat in competition with each other as to their ‘score'." This comment reportedly caused the War Office to respond with a sharp retort that this interpretation was based on a minimum of evidence.12
On the other hand, on 11 September 1941 Nigel de Grey, the deputy director of the GC&CS, noted on a series of early decrypts of German radio reports of massacres by many police battalions and the SS Cavalry Regiment that, "The fact that the Police are killing all Jews that fall into their hands should now be sufficiently well appreciated. It is not therefore proposed to continue reporting these butcheries unless so requested."13 De Grey's comment could represent either his inability to appreciate the implications of the massacres, or his willingness to ignore what the Nazis were doing. Whatever his reason, even someone as highly placed in British intelligence as de Grey, exposed as he was to the daily decrypts detailing the many massacres and other atrocities occurring in the western USSR, would not draw attention to the continuing slaughter in the western USSR.
In either case of the Police massacres in Russia and the operations of the death camps, COMINT could not provide a timely warning of the totality of the tragedies that eventually were to fall on Europe's Jews, Slavs, Poles, and others. This is not to say that the intercepts were useless. In the case of the police massacres, at the very least, the decrypts did alert the British to what was happening in the occupied Soviet Union. As for the concentration and death camps, information about them from intercepts was available only some time after the camps had begun operation. In this case, information from other sources, such as the revelations of Jan Karski, and that gathered by the American diplomats in Switzerland may have proved more useful in alerting the West to the extent of the horrors occurring in the camps.14
Despite what was known, Allied communications intelligence, by itself, could not have provided an early warning to Allied leaders regarding the nature and scope of the Holocaust.
Some Further Observations Regarding the Available Archival Records
There are limited COMINT agency records about the Holocaust
Early in this Guide, we referred to a popular misconception that the Allies intercepted all, or nearly all, Axis communications during the Second World War. This notion probably arose and gained popular credence because of a number of presentations by journalists, television reporters, and popular writers about the cryptanalytic successes against various Axis cipher machines, such as Purple and Enigma, and numerous manual codes and ciphers such as Japanese naval operational code JN-25. Whatever the authors of these presentations about Allied code-breaking successes intended, and whatever was inferred by their audiences, what subsequently developed was a lively image of "sweeping" or "vacuuming" the radio airwaves of all Axis messages. To continue the image, the Allies would then decrypt all of these intercepted messages.
The reality, as has been presented in this Guide, was much different. The COMINT system, from intercept to dissemination of the intelligence, more closely resembled a process that, in most cases, intercepted varying portions of Axis and neutral communications, and then slowly winnowed out the chaff of unusable or unimportant messages. Against the many thousands of Axis radio transmitters, or terminals, that operated on a daily basis, the Allies had a much smaller number of intercept positions at monitoring stations located around the world. They could copy only a limited number of messages. What terminals they covered, in turn, were determined by priorities that were meant to meet the strategic, theater, or operational requirements of the various allied commands. Of those messages sent by the Axis terminals that were copied, a number might have contained too many errors in the transmission by the Axis radioman, or in the transcription by the Allied monitor, which would have rendered them useless for decryption or translation. Of the copy that was usable, a portion may have contained messages encrypted in systems that could not be exploited by Allied cryptanalysts. Many messages that were successfully decrypted did not contain information that was immediately important or useful for Allied intelligence analysts. Only that information that met the priorities set by the command structure would be processed and disseminated to the appropriate recipients. Although no general set of statistics exists, as we have seen, some accounting of translations as a percent of the total intercept range from about five to fifteen percent, though targets of higher priority, such as U-boat communications, had a much higher success rate.
Of course, this explanation applies only to the radio and cable communications that the Allies were able to intercept. A large portion of German internal communications, and those sent to nearby neutrals and conquered nations or regions, was conducted on landline or submarine cable telephone and telegraph systems, which Allied monitors were unable to access. Collecting messages sent over cable required physical access to a terminal or cable through which the messages traveled. The actual acquisition of the messages required either the placement of a tap on the wire in a secret or secure overseas location, or the subversion, bribery, or the blackmail of employees of neutral or enemy national post, telephone and telegraph agencies. However, while Axis cables sent within occupied and neutral Europe were accessible by means of covert operations, protecting the tap from discovery and getting the information from them in a timely manner would have been extremely difficult and dangerous.
Another constraining factor was the prioritized requirements system established by intelligence commands. It hardly would be a surprise to say that COMINT priorities weighed most heavily towards purely military communications. Military communications, especially those by Axis air force and naval forces, most likely would not contain much information on the Holocaust. Wehrmacht communications from the Russian front might have contained some items on massacres, but probably not consistent. German Police and SS communications, which proved to be the major source on the worst depredations of the Holocaust, originally were collected as a supplement to intelligence on the military and civilian domestic situation in Germany.15 Yet, even these communications terminals were not the highest Allied priority.
Nonmilitary communications, principally diplomatic and commercial, which might carry some information about the roundups, concentration camps, or looting were assigned to a secondary priority. Then, not all diplomatic terminals were of equal importance. The American ability to exploit fully Japan's high-level diplomatic communications encrypted with Purple perforce made that country's messages a priority object of Allied intercept. Japan's terminals in the major Axis and neutral capitals were preeminent targets. Japanese communications to other countries rated much lower in importance. While the diplomatic traffic of neutrals generally rated lower in the priority for intercept.
Axis military and diplomatic terminals were emphasized when it came to tasking the COMINT system. This was because of the paramount importance of intelligence from these sources to the successful prosecution of the war. Information about the Holocaust came mostly from lower priority terminals: German Police and neutral diplomatic terminals. To raise the priority for these would have detracted from the primary Allied strategic effort to defeat the Axis, especially Germany. Intelligence elements, including communications intelligence, were tasked to acquire that information necessary to achieve this aim.
A final delimiter to information about the Holocaust was the policy adopted by the Nazis to control the spread of information about the plans and operations carried out to meet the goal of eliminating all Jews and other undesirables from Europe. While the SS and other Nazis would tout their successes in eliminating Jews and others, they did not always record them. This attitude can be illustrated by the remark by Reichsfuehrer SS Heinrich Himmler's comment at a speech to SS notables in which he stated that the Final Solution was "a glorious page in our history, never written (niemals geschriebenes), and perhaps never to be written."16 Aside from some intercept of police, SS, and SD radio messages, and occasional references to Allied publicity about the massacres, deportations, and the concentration camps, there is little other official German radio communications relating to killing operations or other activities relating to the Holocaust. This is partly attributable to the Nazi hierarchy's relatively effective policy limiting official references the actual operations to exterminate the Jews. This policy precluded detailed public discussion or announcements.
Nazi officials also developed a vocabulary of evasion, utilizing obscure bureaucratic language and euphemisms when referring to the Nazi objectives and actual techniques employed by the police, SS, SD, and others. In one directive from Martin Borman, even the expression "final solution" (Endloesung) was disallowed from correspondence. Of course the premier example of this policy of official obscurity was the January 1943 message about the results of Operation Reinhard. It appeared to be an innocuous message with numbers, dates and an unclear context. Analysts at the time missed its significance.
If the official policy was not always followed, and it was hardly a "masterpiece of deception" as claimed by German general Alfred Jodl, it did have the effect of restricting discussion and correspondence about the Holocaust.17 It should also be pointed out that the radio reports by the German police about the massacres in Russia and the SS status reports from the concentration, labor, and death camps eventually were discontinued because of doubts by the German Police and SS high commands about the security of these communications. In light of the above factors — technical limitations, German deception, and Allied military priorities — it is unlikely that Western communications intelligence could have produced more intelligence during the war than it ultimately did.
There are significant differences between the Archival record holdings of the cryptologic agencies of the United States and Great Britain.
Although the differences in the size and content of the records of the American and British cryptologic agencies should not be a surprise, it was only recently that historians understood why. There are two reasons for this gap. The first is the simple fact that the United States entered the war some twenty-six months after it had begun in Europe. During that period, the British had developed an extensive cryptologic effort that encompassed the European Theater. American expertise and capability still were sparse when it came to Axis communications in Europe, even after some staff technical exchanges with British cryptologists in 1940 and 1941. Instead, American cryptologists, because of then current military situation and a decade of pre-war expertise, had taken responsibility for communications intelligence in the Pacific Theater of operations. This dominance became more pronounced when the sole British code-breaking site in Singapore redeployed to Ceylon and then further west to British East Africa in 1942. Eventually, American cryptologists arrived in Britain and were integrated as individuals or as units into the GC&CS operations, but they remained subordinate to the British.
The second reason for the difference probably stems from a misunderstanding of the nature and timing of the official British-American exchange program known as the BRUSA Agreement of June/July 1943. Prior to the agreement, there were separate and limited, reciprocal interchanges between the GC&CS and the SIS and OP-20-G.18 The June 1943 agreement covered exchange of technical communications intelligence and operational intelligence information between GC&CS and the U.S. War Department, as represented by Special Branch and Signal Security Service. (The U.S. Navy would sign a separate, more limited, agreement. It was known as the Extension Agreement and applied to a 1942 exchange arrangement known as the Holden Agreement.) The BRUSA applied to the military sphere only; diplomatic affairs were not included and were arranged in a separate agreement between the Americans at Arlington Hall Station (AHS) and the GC&CS diplomatic section at Berkeley Street, London.19
Also, it needs to be recalled that the U.S. Office of Strategic Services had its own exchange agreement with British intelligence, specifically with its counterintelligence and counterespionage units, M.I. 5 and Section V. of M.I. 6. These exchanges included communications intelligence material that was not given to the American cryptologic elements. These arrangements account for the significant but limited numbers of British decrypts and translations that can be found in OSS records.
Also, and more importantly in regards to archival records holdings, the trade in intelligence that followed the agreements was not retroactive. Only current technical and operational intelligence was passed between London and Washington; though there may have been some limited transfer of dated technical cryptographic and cryptanalytic material used for training purposes. This explains, for example, why there were no copies of German Police decrypts (except for a few training samples sent to Arlington Hall later in the war) in the wartime files of the Signals Intelligence Service. Before the agreements in 1943, this source largely had dried up when the police and SS stopped using radio for transmitting reports about massacres in Russia and the situation in various concentration camps and, instead, passed this information by courier or cable.
The intelligence that was exchanged between the COMINT agencies did not include all of the actual Axis or neutral decrypts or translations. The logistics of such a transfer of information would have been too daunting. The amount of paper traffic would have swamped communications lines between the two capitals. Transmission rates, even of teletypewriters (50-100 words a minute), would have been inadequate. Physically transporting the material across the Atlantic was not a solution: the amount of intelligence generated would have militated against a reasonable or timely review. The digests, summaries, and relatively limited number of translations of the significant intelligence that were transmitted back and forth between the two capitals were considered sufficient.20 This solution probably suited the needs of the leadership in Washington, DC. The most important American customers of British Ultra information were the military commands located in England, such as the Supreme Headquarters Allied Expeditionary Forces (SHAEF) under General Eisenhower, the Allied naval and air authorities, and those in the North African and Italian fronts.
There also was a security angle that limited the American and British exchange. Ever since the British codebreakers had first contemplated meeting their American opposites in 1941, some officials in the London security establishments had fretted over the history of loose control of security information in Washington. The British concerns stemmed partly from the poor American control procedures, and a genuine fear of a press and society unhampered by any Official Secrets Act.21 At the same time, the British concern reveals just how much they appreciated the slender nature of their Ultra triumph. For the first few years of the war, the basis for their mastery over some of the Enigmas rested on cribs (plaintext assumed to be present in an encrypted message), individual German radio operator insecurities, and captured key and Enigma machines.22 Only slowly did the British cryptologic establishment let the Americans in on their cryptanalytic success against Enigma. Although this sharing began in early 1941, it was not until later agreements in 1942 and 1943 that exchanges of technical and operational intelligence was exchanged on a regular basis.
The result is that the current archival records of the British and American COMINT organizations reflect the division and emphasis of efforts between both countries that was established by the July 1943 Britain-United States Agreement. The British records are European-oriented and go back to 1939. American records reflect the emphasis on the Pacific campaign and, except for U-boat campaign and worldwide diplomatic intercept, are bereft of most German military and other service translations.
The Western communications intelligence agencies collected many more intercepts than they finally processed during the war.
It must be remembered that the primary purpose of the Allied intelligence agencies, including the COMINT organizations, was the collection, processing, and dissemination of information supporting the prosecution of the war against the Axis powers. As such, the processing and retention of information based on communications intelligence was contingent upon the strategic exigencies of the moment. As we have seen, intercepted messages that were not processed or used, either because of technical reasons, or overtaken by events as the operational situation changed, were rarely (if ever) retained for future processing. As was mentioned above, as much as 85 to 90 percent of all messages collected by the Allied COMINT agencies were not processed to the point of formal dissemination. This does not exclude the possibility that some intelligence was disseminated in an informal fashion. There likely were instances of hurriedly translated intercept during a fast-moving military situation that was passed along verbally or by phone. And some intercept was processed partly, perhaps only decrypted, but not translated or disseminated in the form of a serialized formal report or translation. In this latter category there are the police decrypts produced by GC&CS that remain in the original German language form. But the decrypts were retained and the information in them was disseminated to the British leadership in the form of briefings and reports.
After the war, the U.S. Army and Navy cryptologic agencies destroyed virtually all of the unprocessed intercept. With victory over the Axis, there was no operational rationale for retaining it. Only two types of "raw" traffic, intercepts that had not been decrypted or analyzed, were retained after the war. The first consisted of a small set of samples of various wartime targets for training purposes. The second set of retained intercept was that relevant to postwar activities. This latter intercept included the U.S. Navy's decryption and translation of Japanese naval messages related to the prewar period, especially those directly concerned with the attack on Pearl Harbor; Soviet espionage messages decrypted by the Army Security Agency, known under the project name of Venona; and German military attaché and diplomatic messages between Berlin and Tokyo. These latter messages contained information about subjects of continuing intelligence interest: wartime Axis technical material and equipment exchange and intelligence that Germany and Japan shared on the Soviet Union. The latter was important in the context of the growing post- war concerns about Soviet intentions and capabilities.
The finished reports that were retained by the COMINT agencies consisted, in the main, of the serialized intelligence product series produced during the war: the numerous series of translations, decrypts, summaries, or reports derived from targeted Axis and neutral military, naval, commercial, security, diplomatic communications. The other material retained was made up of technical reports about cryptologic techniques, equipment, and procedures. It is these collections, stored and available in the Public Record Office and the National Archives, which constitute the records of Allied communications intelligence from the war.
Concurrent with the postwar destruction of unprocessed intercept was the closure of many offices that had been responsible for producing the intelligence material. These closures were part of the drastic, general postwar reduction in the American and British military and intelligence establishments.23 The administrative records and intelligence reports of the various wartime cryptologic work centers were retained by their successor agencies. The papers would not be transferred to their respective national archives for public use until decades later.
In the postwar period, GC&CS, the Army Security Agency (the postwar successor to SIS), and OP-20-G would write in-house histories of their activities during the war. These general works varied in size and scope. The GC&CS histories were very detailed and extended over three dozen volumes. The ASA also produced a multivolume history of the SIS activities, though nowhere as extensive as the British work. Also, a number of short histories of the cryptologic tactical units were written. Some special problem target studies also were published in the years after the war. It was not until the 1980s that these numerous histories were released to the public, but only very slowly, and many of them contained redacted text.24
There are pertinent uses for the available records from the COMINT agencies related to the Holocaust.
Of the perhaps few million translations and decrypts published by the Allied cryptologic agencies during the war, those containing information pertinent to the Holocaust currently number between 700 and 900.25 Even in the collection of the thousands of German Police decrypts that resides in the PRO, only a limited number of them actually carry information about Holocaust-related atrocities. Most of the Police messages cover administrative, order of battle, logistical, or transportation subjects or are concerned with the security situation behind German lines in Russia. In the larger context of the history of the Holocaust, when one considers the enormous amount of public and private records available from all other sources, the COMINT contribution is quite small. Aside from this paucity of records, the information they contain for the most part is not new. Yet, there are some startling finds such as the German police message that contains the number of Jews exterminated in the General Government in 1942 during Operation Reinhard, and details of the operations in Auschwitz contained in the Vrba-Wetzler report in mid-1944. Still, the records have two particular uses for today's researchers and historians.
First of all, in some cases, the COMINT translations are the only extant and contemporary record of some aspects of the Holocaust. In particular, the GC&CS decrypts of the German police, SS, and SD messages are the only existing records of the daily operations of these organizations, since the Germans destroyed most of the original orders, reports and other correspondence. Even as episodic and fragmentary as the police and SS decrypts are, they constitute an important part of the record of the roundup of Jews in Rome and the massacres on the Russian front, and the operations of the concentration and death camps. In other cases, such as the roundup of Jews in Vichy France in 1942 and in Hungary in 1944, the communications intelligence information supplements well the already extensive sources. But, just as important as the translations and decrypts are as a source, like most other COMINT, these records constitute a contemporary record of events. These messages were written as an activity occurred or were composed shortly after while the information was fresh. It is the immediacy of these decrypts and translations that is particularly useful to scholars reconstructing events.
Second, many of the COMINT reports and translations, especially diplomatic messages, point to possible new sources of information or to aspects of the Holocaust that have been overlooked or minimized in conventional narratives. These translations also suggest new avenues of research that consider the attitudes, reactions, or activities of countries not normally associated directly with the Holocaust. In the first instance, in particular, during the roundup of Hungary's Jews, the reports filed by neutral diplomats stationed in Budapest usually not associated with large-scale, international relief work such as Mexico, Portugal, Spain and Eire appear to contain much information about the course of the SS dragnet and the measures taken to rescue Jews. The relative handful of intercepted messages from these countries indicates that their diplomats made significant attempts to help the endangered Jews. If nothing else, these diplomats were fairly reliable reporters of events during the several months of the Hungarian tragedy. These diplomatic intercepts suggest that the foreign ministry archives of those countries may contain more pertinent information than hitherto considered in conventional historical narratives about the Holocaust. Also, other countries with Jewish nationals caught up in Hungary made efforts to rescue their countrymen. Romania, which lost part of Transylvania to Hungary, was concerned over Jews from that region seized by Hungarian authorities.26
The COMINT records also indicate that the waves from the storm of the European Holocaust had reached the shores of Latin America. The reaction by Latin American countries to events in Europe varied. For example, a Vichy diplomatic cable of 24 September 1942 reported on a newly passed Bolivian legislation that further restricted immigration by Jews, Asians, and Africans, and limited the number of Jews allowed to own or work in commercial ventures.27 Other diplomatic reports, whether from neutral Latin American countries, or those sympathetic to the Axis like Argentina and Chile, or minor Allied combatants such as Mexico and Brazil, carry references to the Holocaust. Some countries, such as Brazil, were concerned for Jewish nationals who were caught up in the Nazi extermination machine.28
- The Achievements of the Signal Security Agency in World War II (Washington, D.C: Assistant Chief of Staff, G-2, 1946), 31.
- Phillips, 234-6; Michael Smith, (Bletchley Park and the Holocaust. Intelligence and National Security (Volume 19, No. 2, Summer 2004), 266-7.
- Breitman, 114-5.
- ZIP/G.P.D. 292R/18.7.41, NARA, RG 457, HCC, Box 1386.
- Christopher R. Browning. The Origins of The Final Solution (Lincoln: The University of Nebraska Press; Jerusalem:Yad Vashem, 2004), 213-224. Also, Weinberg, 190-2. For a somewhat detailed description of the military-SS meetings that followed, see Breitman, 36-42.
- As reported in Breitman, 114, German Police Decode, 5 June 1942, items 5 and 6, PRO HW 16/19,
- See Laqueur, 91-93, for a survey of British popular attitudes and Dallek, 166-8, for problems that President Roosevelt faced with American reactions to any effort to help Jews. For more on American general attitudes towards Jews and reactions to news of the Holocaust, see David M. Kennedy, Freedom from Fear. The American People in Depression and War, 1929-1945 (New York: Oxford University Press, 1999), 410-415, and William O'Neill, Democracy at War: America's Fight at Home and Abroad in WWII (Cambridge: Harvard University Press, 1999), 226-229.
- The question of anti-Semitic attitudes surfaced occasionally within the confines of GC&CS and SIS personnel. Three of the major leaders of SIS, William Friedman, Abraham Sinkov, and Solomon Kullback, were Jewish. This fact was noted in passing by GC&CS participants in early discussions with SIS in 1940-42, but appears not to have made any difference in the close cooperation of the two agencies. PRO, HW 1/2, 10 March 1941 "Signals Intelligence passed to Prime Minister Churchill including Enigma messages." And HW 14/62, 26 December 1942, "GC&CS — Directorate, Policy Papers."
- See Laqueur, 93-100 and 159-170, for examples of reaction in Washington and among worldwide Jewish groups.
- Arthur Schlesinger Jr., "The London Operation: Recollections of a Historian." George Chalou, Editor, The Secrets War: The Office of Strategic Services in World War II (Washington D.C.: Government Printing Office, 1992), 66-7.
- Kai Bird, The Chairman: John J. McCloy and the Making of the American Establishment (New York: Simon and Schuster, 1992), 206.
- As reported by Michael Smith, Station X: The Codebreakers of Bletchley Park (London: Channel 4 Books, 1998), 72-4.
- De G[rey] , 11 September 1941, PRO HW 1/62.
- Laqueur, 62-4, 109-11, 119-120.
- Hinsley, 670-1; Phillips, 5-12.
- NARA, RG 238, Nuremberg International Military Tribunal records. Document PS 1919, Prosecution Exhibit USA 190, p. 64-66. From NARA, Holocaust, the Documentary Evidence (Washington, DC: NARA, 1993), 35.
- John Lukacs. The Hitler of History (New York: Vintage Press, 1998), 193.
- For background to these exchanges, see Bradley F. Smith, The Ultra-Magic Deals, 40-129.
- Benson, 108-121.
- Bradley F. Smith, The Ultra-Magic Deals, 163-4; Benson, 111-114; See NARA, RG 457, HCC, Box1026, File 3284, "American Embassy, London, Messages to Military Intelligence Division," May 1943; Box 1119, Folder 2600, "Colonel McCormack's Trip to London, May-June 1943," June 1943. This restriction makes more sense when considering the context of the tide of intelligence reports from all sources. A January 1945 listing of sources of reports by the library section of Special Branch runs 42 pages and contains over 850 titles. See "Sources and Titles of Records Received in Document Section of the I&L Branch, SPSIR-7." 19 February 1945. NARA, RG 457, HCC, Box 1432, Folder "Sources of Titles of Reports."
- Bradley F. Smith, 58-63 and 176-180; F. Hinsley, British Intelligence in the Second World War, Its Influence on Strategy and Operations (London: Her Majesty's Stationery office, 1984), Vol. III, Part I, 52.
- F. H. Hinsley, et alia. British Intelligence in the Second World War, Its Influence on Strategy and Operations, Volume I (London: Her Majesty's Stationery Office, 1979), 336-7, 391; The matter of the "completeness" of sharing COMINT occurred again from late 1942 into early 1943, when the SIS and GC&CS clashed over access to Enigma traffic and Alan Turing's participation in the Bell Laboratory development of a voice scrambler device. See Robert L. Benson, A History of Communications Intelligence during World War II: Policy and Administration (Fort George G. Meade, MD: Center for Cryptologic History, National Security Agency, 1997), 97-108.
- Smith, Ultra-Magic Deals, 221.
- For example, The 37-volume Government Code & Cypher School Histories of World War II, the Military, Air, and Naval sets are located in Series HW 11 at the Public Records Office. Available at either U.S. National Archives, RG 457, HCC or the library of the National Cryptologic Museum are such examples of these histories as: SRH-001, Historical Background of the Signal Security Agency (3 Vols.), (Washington, DC: Army Security Agency, April 1946); SRH-279, OP-20-G File, Communications Intelligence Organization, 1942-1946 (Washington, DC: National Security Agency, 1984); SRH-135, History of the Second Signal Service Battalion, 1939-1945 (Fort George G. Meade: NSA/CSS, 1981); SRH-042, Signals Intelligence Service Staff, HQ, Third U.S. Army, Third Army Radio Intelligence History in Campaign of Western Europe (Washington, DC: Army Security Agency, October 1945).
- Interestingly, scholars of wartime cryptologic history can only estimate the total number of translations and decrypts produced by the United States and the United Kingdom. One of the reasons was the tendency of Allied cryptologists to combine more than one decrypt into a translation or a summary (or "compilation" as the British called it). Another problem was that some translation series were started and then stopped without publication, or were merged into another series. In some cases, decrypts were produced, but not published except for fragments of information, which were put into reports. Also, in some target problems, such as diplomatic intercept, there was substantial duplication in records because of the wartime exchange system. Finally, the job of tabulating these totals would require, in some cases, a manual count — a daunting project, to say the least.
In view of the above, the best estimate possible at this point is that the U.S. and U.K., at a minimum, produced over 4.5 million translations and decrypts. The U.S. produced about 800,000 (+/- 50,000), while the U.K. put out about 3.7 million translations and decrypts. In arriving at these estimates, I wish to thank Ralph Erskine for his advice on the British figures.
- MND Translation, Bern (Anastasiu) to Bucharest, 12 December 1944, H-157300, NARA, RG 457 NSA/CSS, HCC, Box 456.
- MND Translation, La Paz (Mory) to Vichy, 24 September 1942, SIS # 50917, NARA, RG 457 NSA/CSS, HCC, Box 328.
- MND Translation, Rio de Janeiro (Exteriores) to Lisbon, 12 March 1945, H-171950, RG 457 NSA/CSS, HCC, Box 474.