Explanation and History of Venona Project Information
CLASSIFIED BY 5668 SLD/GCL
DECLASSIFY ON: 29K (1)(6)
(PER NSA LTR DTD 4/15/97)
Classified by Multiple
Declassify on: OADR
ALL INFORMATION CONTAINED
HEREIN IS UNCLASSIFIED EXCEPT
WHERE SHOWN OTHERWISE
[Federal Bureau of Investigation]
Office Memorandum · UNITED STATES GOVERNMENT
TO: L. V. Boardman
FROM: A. H. Belmont
DATE: Feb. 1, 1956
Boardman, Nichols, Belmont, Branigan, Tully, B. A. Wells
RETURN TO BELMONT FOR FILING
File in 1734
Purpose of attached summary is to consider possibilities of using [Venona project](S) information for prosecution. In order to view this matter in its proper perspective it was believed necessary to set forth exactly what [xxxx](S) information is as well as to briefly review the origin and history of how the Bureau came to receive this traffic. Consequently, the attached summary is divided into four parts as follows:
- I. EXPLANATION AND HISTORY OF [VENONA PROJECT](S) INFORMATION
- II. WHO HAS KNOWLEDGE OF [VENONA PROJECT](S) INFORMATION?
- III. NATURE OF [VENONA PROJECT](S) MESSAGES AND USE OF COVER NAMES
- IV. PROSECUTION
- A. Advantages
- B. Disadvantages
- C. Communications Intelligence Restrictions
- D. Political Implications
- E. International Implications
- F. Effects on Pending Cases
There is no question that justice would be properly served if Judith Coplon and the Silvermaster-Perlo groups could be successfully prosecuted for their crimes against the United States. The introduction into evidence of [xxxx](S) information could be the turning point in the successful prosecution of these subjects; however, a careful study of all factors involved compels the conclusion that it would not be in the best interests of the U.S. or the Bureau to attempt to use [xxxx](S) information for prosecution [underline by hand]: 1) the question of law involved -- whether or not the [xxxx](S) information would be admitted into evidence as an exception to the hearsay evidence rule; 2) the fragmentary nature of the messages and the extensive use of cover names therein make positive identification of the subjects difficult; 3) the severe restrictions surrounding the mention of communications intelligence data and the anticipated objections from the National Security Agency, the U.S. Communications Intelligence Board and perhaps the National Security Council against public disclosure of U.S. efforts and successes in the communications intelligence field; 4) the resultant damage to U.S. efforts in this field if the Soviets learn of the degree of success in breaking their codes during the 1940's; 5) the political implications in this an election year; 6) the international repercussions and resultant Soviet propaganda when it is disclosed that the U.S. intercepted and worked on breaking Soviet coded messages when the countries were allied against the Axis; and 7) the effects on pending espionage cases which are based on [xxxx](S) information. These factors weigh heavily against using [xxxx](S) information for prosecution.
Based on information developed from [xxxx](S) traffic, there has been prosecution of Judith Coplon, Valentin Gubitchev, Emil Klaus Fuchs, Harry Gold, Alfred Dean Slack, Abraham Brothman, Miriam Moskowitz, David Greenglass, Julius and Ethel Rosenberg, Morton Sobell and William Perl. All of these cases were investigations instituted by us directly or indirectly from [xxxx](S) information. These prosecutions were instituted without using [xxxx](S) information in court.
It is believed that we should continue this policy.
[By hand:] I agree. L. 2/8
TO CONSIDER POSSIBILITIES OF USING [VENONA PROJECT](S) INFORMATION FOR PROSECUTION
I. EXPLANATION AND HISTORY OF [VENONA PROJECT](S) INFORMATION
[Venona project](S) is a small group of cryptographers attached to the National Security Agency (NSA) who work on deciphering certain Soviet intelligence messages covering the period 1942-46. These Soviet messages are made up of telegrams and cables and radio messages sent between Soviet intelligence operators in the United States and Moscow. Consequently, this material falls within the category of communications intelligence information and as such is subject to the most stringent regulations governing dissemination on a "need-to-know" basis. These messages would seem to fall into the same category as teletypes from a feild office to the Bureau (in code) and detailed reports were undoubtedly being sent by the MGB (Soviet Intelligence Service) in the diplomatic pouch. The usual Russian method in concealing the text of these messages is to first translate the Russian plaintext into code by using a code book and then apply a one-time pad to the code, thus making it practically impossible to recover the message. Interceptions of the [xxxx](S) messages were made by the U.S. Army. The intercepted messages consist of a series of numbers. These numbers are placed on work sheets by [xxxx](S) and efforts are then made to arrive at the additive (the enciphering process). Once the additive is correctly determined the message can be read by using the MGB code book which has been partially reconstructed by NSA. The chief problem is to develop the additive. This requires painstaking effort by experienced cryptographers who can also translate Russian. Both [xxxx](S) and [xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx](S) work on this problem and work sheets have been made up by [xxxx](S) and sent to [xxxxxxxxxxx](S). Also, [xxxx](S) has a man in England working with [xxxxxxxxxx](S). In turn, [xxxxxxxxxxx](S) have a cryptographer working full-time at [xxxx](S).
The first report received by the Bureau on MGB deciphered traffic was received from Army Security Agency (predecessor of NSA) 4/16/48. Colonel L.E. Forney, Intelligence Division of the Army, advised at that time that the messages given to the Bureau were dated 1944 but the decode had just been accomplished. Col. Forney advised that the fact that the Army had been able to break into Russian traffic was to be held very closely by the Bureau. The initial decodes also had been distributed to the Navy which was also working on decoding Russian traffic at that time. The Navy in turn informed CIA about the initial decodes. Since that time, however (until May 1952 when CIA began getting [xxxx](S) information), this traffic has not been published, as in other communications intelligence traffic which is distributed to authorized consumers, but kept on work sheets for security purposes. It was made available only to the Bureau and [xxxxxxxxxxxx](S). The initial decodes of this Russian traffic were fragmentary and full of gaps. Col Forney felt that the Bureau by studying the messages and conducting investigations would be able to develop information which would assist the Army cryptographers in reading additional unrecovered portions of the messages. The Army stated these messages were part of an MGB system and subsequent study has confirmed that opinion.
II. WHO HAS KNOWLEDGE OF [VENONA PROJECT](S) INFORMATION?
From April 1948 until May 1952 the Bureau was the only U.S. agency, other than NSA, aware of this Russian MGB traffic. During that time [xxxxxxxxxxx](S) were working on the traffic in conjunction with [xxxx])S) and, consequently, we know [xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx](S) were aware of this traffic. In June 1948 Col Forney advised that the Navy and Air Force should not be notified of the existence of [xxxx](S) information. In May 1952, through the insistence and pressure put on NSA by General Walter Bedell Smith, then head of CIA, that agency was given access to [xxxx](S) information and has received and worked on it since that time. Although the Navy, Air Force and State are members of the U.S. Communications Intelligence Board (USCIB) and as such are entitled to receive [xxxx](S) information, they have not received it to date, except those original decodes which were given to the Navy in April 1948. Thereafter, the Navy was not given any other decodes.
In rare cases where consideration is given to dissemination of [xxxx](S) information by the Bureau, clearance is first obtained from NSA and if no objection is raised by that agency the information is paraphrased to protect the source. The information can then be given only to one cleared to receive communications intelligence material. Very few people in the Government are so cleared.
III. NATURE OF [VENONA PROJECT](S) MESSAGES AND USE OF COVER NAMES
The messages [xxxx](S) furnishes the Bureau are, for the most part, very fragmentary and full of gaps. Some parts of the messages can never be recovered again because during the actual intercept the complete message was not obtained. Other portions can be recovered only through the skill of the cryptographers and with the Bureau's assistance. Frequently, through an examination of the messages and from a review of Bureau files, the Bureau can offer suspects for individuals involved. When [xxxx](S) breaks out a part of the message and reads to the point where it is determined that reference is being made to certain information derived from U.S. Government records or documents, the Bureau conducts investigation to locate such records or documents. When located, these records are furnished to [xxxx](S) and if it turns out to be the correct document, [xxxx](S) uses it as a "crib" and thus is able to read previously unrecovered portions of the message. It must be realized that the [xxxx](S) cryptographers make certain assumptions as to meanings when deciphering these messages and thereafter the proper translation of Russian idioms can become a problem. It is for such reasons that [xxxx](S) has indicated that almost anything included in a translation of one of these deciphered messages may in the future be radically revised.
Another very important factor to be considered when discussing the accuracy of these deciphered messages is the extensive use of cover names noted in this traffic. Once an individual was considered for recruitment as an agent by the Soviets, sufficient background data on him was sent to headquarters in Moscow. Thereafter, he was given a cover name and his true name was not mentioned again. This makes positive identifications most difficult since we seldom receive the initial message which states that agent "so and so" (true name) will henceforth be known as "_______" (cover name). Also, cover names were changed rather frequently and teh cover name "Henry" might apply to two different individuals, depending upon the date it was used. Cover names were used for places and organizations as well as for person, as witnessed by the fact that New York city was "Tyre" and the FBI was "Hata." All of the above factors make difficult a correct reading of the messages and point up the tentative nature of many identifications.
For example, among the first messages we received in 1948 was one concerning an individual with the cover name "Antenna." The message was dated 5/5/44 and it set forth information indicating that "Antenna" was 25 years of age, a "fellow countryman" (member of CP,USA), lived in "Tyre" (New York), took a course at Cooper Union in 1940, worked in Signal Corps at Ft. Monmouth, and had a wife named Ethel. We made a tentative identification of "Antenna" as Joseph Weisbrod since the background of Weisbrod corresponded with the information known about "Antenna." Weisbrod was about the right age, had Communist background, lived in NYC, attended Cooper Union in 1939, worked at the Signal Corps, Ft. Monmouth, and his wife's name was Ethel. He was a good suspect for "Antenna" until sometime later when we definitely established through investigation that "Antenna" was Julius Rosenberg.
Cover names were used not only to designate Soviet agents but other people mentioned in the messages were given cover names. For example, "Kapitan" (Captain) was former President F.D. Roosevelt. A survey of the traffic as a whole suggests that a cover name like "Kapitan" serves a different purpose than cover names assigned to agents operation for the Soviets in an intelligence capacity. The latter type of cover names are presumably designed to protect the person of the agent directly. The "Kapitan" type of cover name merely obscures the sense and thereby affords indirect protection to the agent and at the same time is calculated to baffle foreign intelligence organizations as just what intelligence is being transmitted.
It is conceivable that if we could use [xxxx](S) information in court the Government might successfully prosecute Judith Coplon and a number of subjects in the Silvermaster and Perlo groups. It is also evident that a public disclosure of [xxxx](S) information would corroborate Elizabeth Bentley.
Judith Coplon was not mentioned by name in the messages but the identifying information set forth in the [xxxx](S) traffic, dated July 1944 and Jan. 1945, concerning the individual designated by the cover name "Sima" made it certain that "Sima" was Judith Coplon. Our subsequent investigation added additional evidence when it was determined that Coplon was still operating as a Soviet espionage agent in 1949 when she was observed in contact with her Soviet superior, Valentin Gubitchev.
The [xxxx](S) information tends to fall into certain divisions corresponding to a considerable extent to the divisions apparent in the Silvermaster-Perlo cases. From the data set forth in the messages and from our knowledge of the Silvermaster and Perlo groups, as furnished by Elizabeth Bentley, it appears that Silvermaster is identical with the individual in [xxxx](S) designated by the cover name "Robert." It also appears that "Donald" is William Ludwig Ulman; that Bentley herself is "Good Girl" and "Myrna"; whereas Helen Silvermaster appears to be "Dora"; Abraham George Silverman fits "Aileron" and Jake Golos appears identical with "Zouk." Others in the Silvermaster group have been tentatively identified with individuals designated by cover names in the [xxxx](S) material.
The Perlo group fits into the [xxxx](S) information when we examine the following message of 5/13/44:
"Mayor" (unidentified) in NYC personally prepared a report to MGB headquarters in Moscow advising that some unspecified action had been taken regarding "Good Girl" (Bentley) in accordance with instructions of "Helmsman" (Earl Browder). "Mayor" then made reference to winter and also to "Magdoff-'Kant' "(Probably Harry Magdoff). This latter reference was then followed by a statement that in "Good Girl's" opinion "they" are reliable. It was also mentioned that no one had interested himself in their possibilities.
The name "Storm" (unidentified) was mentioned and it was then reported that "Raider" (Victor Perlo), "Plumb" (Charles Kramer), "Ted" (Edward Fitzgerald) and "Kant" (Harry Magdoff) would take turns coming to NY every two weeks. "Mayor" said "Plumb" and "Ted" knew "Pal" (Nathan Gregory Silvermaster, whose cover name was later changed to "Robert").
With reference to the foregoing, it is to be recalled that Elizabeth Bentley advised that Jacob Golos informed her he had made contact with a group in Washington, D.C. through Earl Browder. After the death of Golos in 1943, two meetings were arranged with this group in 1944. The first meeting was arranged by Browder and is believed to have been held on 2/27/44. The meetings were held in the apartment of John Abt in NYC and Bentley was introduced to four individuals identified as Victor Perlo, Charles Kramer, Harry Magdoff and Edward Fitzgerald.
As can be seen from the above, if the [xxxx](S) messages (and there are several of them) could be introduced into evidence their contents, along with the testimony of Elizabeth Bentley, might convict a number of Silvermaster and Perlo subjects.
- A. Advantages
The advantages of using [xxxx](S) information for prosecutive benefits (assuming [underline by hand] it would be admitted into evidence) are obvious. It would corroborate Elizabeth Bentley and enable the Government to convict a number of subjects such as Judith Coplon and Silvermaster, who continued freedom from prosecution is a sin against justice. Public disclosure of these messages would vindicate the Bureau in the matter of the confidence we place in Elizabeth Bentley's testimony. At the same time, the disadvantages of using [xxxxxxxxxxxxxxxx](S) information publicly or in a prosecution appear overwhelmingly.
- B. Disadvantages
In the first place, we do not know if the deciphered messages would be admitted into evidence and if they were not, that would abruptly end any hope for prosecution. It is believed that the defense attorney would immediately move that the messages be excluded, base on the hearsay evidence rule. He would probably claim that neither the person who sent the message (Soviet official) nor the person who received it (Soviet official) was available to testify and thus the contents of the message were purely hearsay as it related to the defendants. Consequently, in order to overcome such a motion it would be necessary to rely upon their admission through the use of expert testimony of those who intercepted the messages and those cryptographers who deciphered the messages. A question of law is involved herein. It is believed that the messages probably could be introduced in evidence on the basis of an exception to the hearsay evidence rule to the effect that the expert testimony was sufficient to establish the authenticity of the documents and they were the best evidence available.
Assuming that the messages could be introduced in evidence, we then have a question of identity. The fragmentary nature of the messages themselves, the assumptions made by the cryptographers in breaking the messages, and the questionable interpretations and translations involved, plus the extensive use of cover names for persons and places, make the problem of positive identification extremely difficult. Here, again, reliance would have to be placed on the expert testimony of the cryptographers and it appears that the case would be entirely circumstantial.
Assuming further that the testimony of the Government cryptographers were accepted as part of the government's case, the defense probably would be granted authority by the court to have private cryptographers hired by the defense examine the messages as well as the work sheets of the Government cryptographers. Also, in view of the fragmentary nature of the majority of these messages, the defense would make a request to have its cryptographers examine those messages which [xxxx](S) has been unsuccessful in breaking and which are not in evidence on the premise that such messages, if decoded, could exonerate their clients. This would lead to the exposure of Government techniques and practices in the cryptography field to unauthorized persons and thus compromise the Government's efforts in the communications intelligence field. Also, this course of action would act to the Bureau's disadvantage since the additional messages would spotlight individuals on whom the Bureau has pending investigations.
In addition to the question of law involved, there are a number of other factors which weight against the use of [xxxx](S) information in court. These factors are most important from the Bureau's standpoint.
- C. Communications Intelligence Restrictions
[xxxx](S) information is communications intelligence information and, consequently, it is all classified "Top Secret" and is strictly controlled. Reference to the existence of communications intelligence either directly or indirectly must be avoided except among those to whom this information is necessary for the proper performance of their duties. Dissemination of communications intelligence material is made on a "need-to-know" basis and the [xxxx](S) material has been restricted even more so than other communications intelligence data since it has not been published or distributed to USCIB consumers, other than the FBI and, more recently, CIA. Before any decision could be made on using [xxxx](S) information in court the messages would have to be declassified. Such declassification would, of course, have to be passed on by USCIB and it appears evident that the approval would also have to come from the Special Committee of the National Security Council for Communications Intelligence and, as a matter of fact, it probably would necessitate approval of the President. At the same time, it appears that [xxxxxxxxxxx](S) would have to be notified since [xxxx](S) data has been decrypted through the joint efforts of the United States and the United Kingdom. In the Laucklin Currie case when we considered the possibilities of prosecution, General Canine of NSA indicated it would be highly inadvisable to reveal U.S. efforts to break the Russian code. It is believed that NSA would strongly object to any attempt to use [xxxx](S) information in court since to do so would reveal to the Soviets the degree of success the U.S. had in breaking the Soviet code which was used during 1943-46. In addition, this knowledge by the Soviets of the degree of success the U.S. had in breaking their code might work to the disadvantage of NSA in current efforts to break the Russian codes.
- D. Political Implications
It is believed that disclosure of existence of [xxxx](S) information at this time would probably place the Bureau right in the middle of a violent political war. This is an election year and the Republicans would undoubtedly use disclosure of the [xxxx](S) information to emphasize the degree of infiltration by Communists and Soviet agents into U.S. Government during the 1940's when the Democrats were in power. At the same time, the Democrats would probably strike back by claiming that the FBI withheld this information from the proper officials during the Democratic administration and at the same time would salvage what credit they could by claiming that the messages were intercepted and deciphered during the course of their administration and under their guidance. The Bureau would be right in the middle.
- E. International Implications
The Russians would undoubtedly scream that the U.S. had been expending money and manpower on intercepting and breaking the Russian code during the time the two countries were allied against a common enemy. Its propaganda machine would work overtime proving that this was evidence that the U.S. never acted in good faith during the war. Also, while no written record has been located in Bureau files to verify this it has been stated by NSA officials that during the war Soviet diplomats in the U.S. were granted permission to use Army radio facilities at the Pentagon to send messages to Moscow. It has been state that President Roosevelt granted this permission and accompanied it with the promise to the Soviets that their messages would not be intercepted or interfered with by U.S. authorities. Here, again, the Soviets would vilify the U.S. as an unfaithful ally and false friend.
- F. Effects on Pending Cases
The Bureau is currently investigating about 100 cases on individuals either mentioned in [xxxx](S) traffic or having some connection therewith. In addition, some of the subjects in the Mocase are mentioned in the [xxxx](S) traffic. The public revelation of our knowledge of this traffic and the individuals involved therein probably would cause some of these individuals who may still be operating for the Soviets to discontinue their activities and possibly disappear. The effects on these pending cases would be difficult to assess at this time.
[Initials by hand] WHB, ERT, [Illegible], JMV