[Packet Document #6] UNITED STATES SENATE COMMITTEE ON THE JUDICIARY SUBCOMMITTEE ON THE CONSTITUTION WASHINGTON, D.C. 20510 October 15, 1979 Honorable G. R. Dickerson Director, Bureau of Alcohol, Tobacco and Firearms Department of the Treasury 15th and Pennsylvania Avenue, N.W. Washington, D. C. 20220 Dear Director Dickerson: As the former Chairman of the Subcommittee to Investigate Juvenile Delinquency, with oversight responsibilities for the Federal Gun Control Act, I held hearings to help assure that our federal gun control laws were being properly implemented and enforced. As the current Chairman of the committee on the Constitution it is my responsibility to assure that all of our citizens are given fair and impartial treatment in the enforcement of these laws. Recently, it was brought to my attention that a potentially serious problem exists with regard to the enforcement of Title II of the Gun Control Act of 1968, the National Firearms Act (26 U.S.C. 5801-5872). I have received copies of internal memoranda prepared by the Bureau regarding the lack of accuracy of the National Firearms Act registration system. These memoranda were obtained in the course of a civil forfeiture proceeding between the National Rifle Association and the government involving firearms confiscated from the NRA museum, and brought to my attention by the NRA. As you will see, the memoranda states that the registration system was so inaccurate that those directly responsible for administering it were concerned that innocent persons might be convicted as a result of the deficiencies. As you know, in every case brought under the National Firearms Act the government introduces a certificate stating that a thorough search of the registration system files has not uncovered the firearm in question being registered to the defendant. Honorable G. R. Dickerson Page two October 15, 1979 Since the credibility of the certification is only as valid as the registration system, any problems with the system would, as the memoranda recognize, compromise the certification. That would appear to place a major element of all prosecutions for possession under Title II in jeopardy, and casts doubt upon whether reasonable doubt could ever be overcome as to whether the firearm had been registered. This situation is of serious concern to me in that the memoranda appear to be materials that should have been released to any defendant being prosecuted under Title II, under the Brady v. Maryland doctrine. However, I understand you actively resisted such a release in the NRA case, to the point of entering a claim of executive privilege. I would appreciate your advising me as to the accuracy of the Title II registration files at this time. In the event that the deficiencies cited in the 1975 memoranda did exist, I would like any information you may have as to whether any person has been prosecuted, or any property forfeited, as a result of an incorrect certification for the Title II files from 1975 to the present. In addition, I would be interested in the Bureau's explanation for your attempt to claim "executive privilege" to prevent release of these memoranda. Your attention and response to this request will surely be of benefit to my colleagues and I on the Judiciary Committee. Sincerely, Birch Bayh Enclosure Chairman [new letter] NOV 16, 1979 Honorable Birch Bayh United States Senate Washington, DC 20510 Dear Senator Bayh: This refers to your letter of October 15, 1979, inquiring about the accuracy of the National Firearms Registration and Transfer Record which is maintained under the National Firearms Act. You also seek an explanation for ATF's claim of executive privilege in a pending forfeiture case involving the National Rifle Association with respect to internal memoranda concerning the NFA record system. Attorney General Civiletti has received a similar letter dated October 25, 1979, from Senator James A. McClure. The Department of Justice is now gathering information and researching the legal issues raised by Senator McClure. When the Justice Department completes its study, we will be in a position to fully respond to the questions posed in your letter. You can be assured that we are working closely with the Justice Department to prepare a responsive reply to your letter. Sincerely yours, G. R. Dickerson Director JBPatterson:ji 11/14/79 23,302-W [new letter] RESPONSE TO SENATOR McCLURE'S STATEMENT RELATIVE TO HIS CONCERN THAT PEOPLE HAD BEEN WRONGFULLY SUBJECTED TO CRIMINAL ACTIONS BASED UPON ERRONEOUS RECORDS. Title 26, United States Code (U.S.C.), Chapter 53, Section 5841(e) requires that a person possessing a firearm registered as required by this section shall retain proof of registration which shall be made available to the Secretary upon request. There is no record of any person having been charged with illegal possession of an unregistered firearm, subsequently presenting to this Bureau, or to the courts, proof of registration as required by law. In every criminal prosecution for possession of an unregistered firearm, a certification of registration status is prepared by the Chief of the National Firearms Act Branch for introduction as evidence. There is no record of any individual charged with illegal possession of an unregistered firearm alleging that the firearm was, in fact, registered, and taking issue with the Bureau certification reflecting non-registration. The integrity of the National Firearms Registration and Transfer Record (NFRTR) is being challenged through the use of a study of the NFRTR which was prepared by this Bureau in July 1975, and certain related memoranda. In the first place, these memoranda are more than four years old. Testimony has been offered at trial which is unrebutted, that immediately after, and in response to, these 1975 criticisms, various changes were made in the NFRTR and recordkeeping procedures. Those individuals in charge of the Record were given authorization to obtain overtime and equipment to change the filing system from Diebold machines, with limited access, to file cabinets with individual folders for each registered possessor of a firearm. In transferring the records from the Diebold machines to the new filing system, a great amount of effort and energy was expended to correct any misfilings, to eliminate excess and "dead" paper, and to preserve and protect existing registration forms. Thus, to the extent -2- that problems which were noted in the 1975 memoranda ever existed, it is clear from the unrebutted testimony introduced at trial that most of these problems, more properly conditions, were addressed and corrected as a result of a substantial amount of energy, money, and effort expended by ATF in late 1975 and early 1976. Secondly, the unrebutted explanatory testimony at trial from two witnesses with detailed knowledge of the NFRTR from December 1958 to the present, clearly demonstrated that the vast majority of "problems" identified in the 1975 study could not affect the ability of ATF to accurately certify that a particular firearm was or was not registered. [new letter] Philip B. Heymann Assistant Attorney General Criminal Division November 29, 1979 LL:JJD:ajw Lawrence Lippe, General Litigation & Legal Advice Section Criminal Division Response to letter from Senator McClure There is pending in the United States District Court for the District of Columbia a civil forfeiture action, United States v. Seven Miscellaneous Firearms, C.A. 78-1358, involving automatic weapons seized by the Bureau of Alcohol, Tobacco and Firearms (BATF) from a museum at the headquarters of the National Rifle Association (NRA). Trial has been completed, post-trial briefs submitted, and the case is before the court (Gasch J.) for decision. During discovery, the NRA sought documents pertaining to a 1975 BATF infernal study of the national Firearms Registration and Transfer Record. [footnote 1] Those documents cite a number of problems and inaccuracies in the record system. BATF initially resisted release of the documents claiming executive privilege and then relented, releasing all of the requested documents in their entirety. They formed the basis of an NRA argument that the firearms record is so rife with inaccuracies that the Government could not meet its burden in proving that the seized firearms were not properly registered as required by the National Firearms Act (Title II of the Gun Control Act of 1968, 26 U.S.C. 5801-5872). Senator McClure is a vocal opponent of gun control who has introduced legislation to repeal the Gun Control Act of 1968. Additionally, he is a member of the Subcommittee on Treasury, Postal Service and General Government which has oversight responsibility over BATF. In hearings on July 11 and 12 of this year into alleged law enforcement abuses by BATF he expressed his belief that the failure to reveal the inaccuracies in the firearms record system as stated in the 1975 memoranda requires the reversal of all convictions heretofore obtained under the National Firearms Act. It should also be noted that Senator Birch Bayh in his capacity as Chairman of the Subcommittee on the Constitution has written to BATF -------------------- 1. James Featherstone, present General Counsel for the NRA was formerly Deputy Assistant Secretary of the Treasury for Enforcement. Presumably, he gained knowledge of the study in his former position. General Litigation Section - 2 - expressing concern that these documents should have been disclosed in criminal cases under the doctrine of Brady v. Maryland, 373 U.S. 83 (1963). II The 1975 Study Section 5841(a) of Title 26, United States Code, establishes a control registry of firearms known as the National Firearms Registration and Transfer Record which shall include identification of the firearm, date of registration, and the identification and address of the person entitled to possess the firearm. [footnote 2] The act does not establish or prescribe the manner in which the registry is to be maintained. Since 1968 the record system has been composed of two basic elements: (1) a transaction record file containing original making, registration, and transfer forms arranged in alphabetical order by the name of the registrant and (2) a "back-up" index card file arranged by firearm serial number containing descriptive information about the firearm and the names of all persons to whom the firearm has been registered. The transaction record file is the official National Firearms and Transfer Record. The index card file was created by BATF for convenience to permit cross- referencing and to assist in the location of the official record where only the serial number and description of the firearm are available. Both files are examined for purposes of certifying the absence of a registration record. [footnote 3] We have thoroughly examined the BATF internal memoranda provided by Senator McClure [footnote 4] and thorough interviews of BATF personnel have established the following chronology and context for them. Title II of the Gun Control Act, in addition to ----------------------- 2. The predecessor statute, enacted in 1934, also required the registration of certain firearms. Both the old and the new acts apply to machine guns, destructive devices and other "gangster" weapons. 3. Evidence of a lack of a record of registration normally takes the form of a certificate by the record custodian admissible under Rule 27 Fed.R.Crim.P. incorporating Rule 44 Fed.R.Civ.P. 4. We had received these and other similar documents from BATF a few weeks before we received the Senator's letter. Prior to that we had no knowledge of the 1975 study or any of the matters relating to it. - 3 - creating the Registration and Transfer Record, established an amnesty period of 30 days from the effective date of the act during which time weapons could be registered with no questions asked. (See section 7273, P.L. 90-618). The amnesty period spawned a massive volume of registrations, transfers and correspondence which the clerical staff was ill equipped to handle. As a result, some weapons which did not require registration were registered, some were mistakenly registered by part number rather than serial number, and some documents were misfiled. The staff responsible for the system was aware of these problems. In 1971 a program to microfilm the entire record was undertaken in order to increase the efficiency of record searches and to safeguard the original documents. It should be noted that the record is primarily used for recording the making of weapons and the transfer of them. Searches for evidentiary purposes are but a part of the overall usage of the record. Thus by microfilming the record, it was believed, the security and integrity of the official file could be better maintained by limiting its usage to evidentiary and other high priority purposes. Unfortunately, the microfilming effort failed because of the limited capabilities of the equipment. In April 1975, the Chief of the NFA Branch requested funds to duplicate the entire record. (Tab 1). He noted the failure of the microfilming and the inefficiency of the existing mechanical file that could be used by only one person at a time. Further, he stated, "We continuously discover discrepancies and inaccuracies int he registration file which, if discovered during a trial, would destroy the future credibility of [certification or testimony of the nonexistence of a registration record]." In an attachment he cited an instance where they would have certified a weapon was registered when in fact it wa not, instances where forms should have been sent to a taxpayer but were not, a misfiling of a transfer form with a letter where the form should have been in the record, and instances arising from the amnesty period where weapons that were not required to be registered probably should have been and many dual registrations of the same firearm by two or more persons. Importantly, he did not suggest that any certification was erroneous. The request for funds "to duplicate a record which is admittedly inaccurate" was denied by the Director of the BATF on May 30, 1975. At the same time he commissioned a study group to review the record "and make recommendations to insure its accuracy and completeness" (Tab 2). An initial memorandum dated May 22, 1975, identifying the problems to be addressed by the study group and recommending the composition of the group and its framework for analysis is attached at Tab 3. The final report of the study group dated - 4 - July 1, 1975, is attached at tab 4. The report identified seventeen problem areas in the record system (see pp.3-4). The most significant of these in terms of its effect on the validity of a certification is where both the index card and the registration record are missing. It must be explained, however, that the only way to determine whether this situation exists is by first knowing that a specific individual or weapon is registered and the finding that both are missing. Obviously, if the individual has never registered a firearm or if the firearm has never been registered by anyone, no record whatsoever will exist. The report does not suggest that this problem actually existed and it cites no examples where both records were determined to be missing. [footnote 5] Indeed, none of the BATF personnel we interviewed were aware of any case where this happened. Most of the other problems identified by the group were raised in the April 3, 1975 memorandum. The report concludes by recommending that the entire firearms record be computerized. As best we can determine the Director of BATF took no official action on the study group's recommendations. The report remained in his safe until the discovery request was made in the civil case described supra. Strangely, the General Counsel's office at BATF was not made aware of the existence of these memoranda until the discovery request despite the fears expressed by their authors that the record was inaccurate and held the potential for agency embarrassment. In connection with the civil case on August 20, 1979, the Assistant Director for Technical and Scientific Services prepared a memorandum rebutting the findings of the study group and setting forth the changes made in the record system since 1975 (Tab 5). The changes consist primarily of a reorganization of the files and the employment of a different file storage system. As part of the file reorganization unnecessary paper was purged and the charge out system changed. Prior to the change when index cards were removed from the file for examination or other purpose, a charge out card containing some of the information from the document and the initials of the person removing the document would be inserted in the file. The study group criticized this practice as leading to the loss of index cards and increasing the potential for misfiling. Now, index cards removed from the file are photocopies and replaced. [footnote 6] ------------------ 5. If this problem actually existed, the only solution would be to declare another amnesty period. The Secretary is empowered to do this under existing legislation. 6. It should be kept in mind that the index card system is the "unofficial" back-up file created by BATF. The charge out practice did not affect the "official" registration documents. -5- III The Brady Issue Senator McClure states that the memoranda are "clearly material the Government is required to release to a defendant under Brady v. Maryland, 373 U.S. 83 (1963)," and suggests that the Department disclose the memoranda to all persons convicted under the National Firearms Act as well as defendants in all future cases. This sweeping conclusion misapprehends the scope of Brady and its progeny. Brady involved the failure by the prosecutor, in the face of a specific request, to disclose a confession by a codefendant which was exculpatory of the accused. The Court held that "the suppression by the prosecution of evidence favorable to an accused upon request violates due Process where the evidence is material either to guilt or to punishment." Id. 373 U.S. at 87. The Brady rule has been extended to cover material that directly impeaches a key witness whose reliability may be determinative of guilt or innocence. See, e.g., Giglio v. United States, 405 U.S. 150 (1972) (perjured testimony regarding promises of leniency); United States v. McCrane, 547 F.2d 204 (3d Cir. 1976) (letters to state agencies on witness' behalf), Boone v. Paderik, 541 F.2d 447 (4th Cir. 1976) (promises of leniency by police). Brady pertains to information known to the prosecutor but unknown to the defense. United States v. Agurs, 427 U.S. 97 (1976). The possessor of a firearm within the ambit of the National Firearms Act knows whether the firearm is registered to him or not. Indeed, persons lawfully possessing such weapons are required under the Act to "retain proof of registration which shall be made available to the Secretary upon request." 26 U.S.C. 5841(e). Thus the BATF memoranda are clearly not exculpatory. To the extent that the memoranda may be useful for impeachment, their disclosure is not constitutionally mandated, for "the mere possibility that an item of undisclosed information might have helped the defense, or might have affected the outcome of the trial, does not establish materiality in the constitutional sense." Agurs at 109,110. Where no Brady request or only a general request is made, undisclosed information will be held material only if "the omitted evidence creates a reasonable doubt that did not otherwise exist." Id. at 112. It is also clear that "[t]here are situations in which evidence is obviously of such substantial value to the defense that elementary fairness requires it be disclosed even without a specific request." Id. at 110. The memoranda do not rise to this level. They are most properly viewed as a critique of the record system drafted in an effort to determine how the system could be made more efficient and insulated from the possibility of error. The expressions of concern about the integrity of the system threatened because of discrepancies and inaccuracies are -6- overstatements having little foundation in the actual findings of the study. None of the memoranda suggest that any erroneous certifications have been made. No one we interviewed had ever heard of any case in which a certification had been made and the defendant produced evidence of registration. In the civil case the Government adduced unrebutted testimony from veteran BATF employees that at least since 1958 no person had produced a registration form or testified in court that he had registered a particular weapon after BATF had certified that that person was not the registered possessor of a particular weapon. [footnote 7] --------------------- 7. BATF issues an average of 2,000 certifications annually. Additionally, there are approximately 2,530-3,000 "look-ups" each year for investigative purposes. In light of this volume of record searches, the absence of any known errors underscores the integrity of the record system. [new letter] P?H:RLK:LL:JJD:ajw Retyped: 12/7/79 ??-22-0 December 12, 1979 Honorable James A. McClure United States Senate Washington, D.C. 20510 Dear Senator McClure: This is in response to your letter to the Attorney General of October 25, 1979, concerning the findings of the internal audit by the Bureau of Alcohol, Tobacco and Firearms of the National Firearms Registration and Transfer Record. You expressed concern that the registration system is inaccurate and incomplete; seriously impugning the accuracy of evidentiary certifications which form part of the proof in each prosecution under the National Firearms Act. Additionally, you suggest that the audit documents are material the Government is required to release under the doctrine of Brady v. Maryland, 373 U.S. 83 (1963). We have thoroughly reviewed not only the documents enclosed with your letter but also other documents pertaining to the record system and pleadings filed in the United States District Court for the District of Columbia in connection with United States v. Seven Miscellaneous Firearms (Civil Action 78-1358). Moreover, we have discussed at length the matters raised in the documents with representatives of the Office of Legal Counsel at BATF as well as officials at the National Firearms Act Branch who are familiar with the firearms registration system. Based upon this review, we are unable to agree with your conclusions that a serious problem with the integrity of accuracy of certifications existed in 1975 or exists at present. Therefore, we must also disagree with your opinion that the audit materials are Brady material. The criticisms of the record raised by the 1975 study relate primarily to the existence of misfilings, excess paper in the file, the inability to accumulate rapidly statistical data, inconvenient access to records due to limitations on the filing equipment, and the charge out procedure for removing index cards from the file. As you know, the filing system consists of two primary components: an alphabetical file of registrations and a corresponding Records General Litigation Dion (2) Reuch O?? - 2 - cross-reference file of index cards arranged by serial numbers of the firearms. Both components are examined in searches for evidentiary purposes. Only in the event that a firearm has been duly registered but both files are missing can an erroneous certification be made. The 1975 study describes no such situation. Our review and the sworn testimony in the aforementioned civil case of experienced BATF personnel reveal that no one has ever offered evidence in court that a certification was erroneous or that any defendant in a civil or criminal case had actually registered the weapons at issue. In this regard it should be noted that BATF makes approximately 2,000 such certifications each year. Moreover, it must be kept in mind that 26 U.S.C. 5841(e) requires persons possessing registered firearms to retain proof of such registration. Since the 1975 study, the record system has been reorganized and purified through several means. The mechanical file has been replaced. Excess paper unrelated to registration status has been eliminated. The charge-out system has been eliminated. Misfilings have been corrected. All new applications for making, registration or transfer are now recorded on a log in addition to the index cards. These actions have corrected most of the conditions addressed by the 1975 study. I trust the foregoing explains the position of the Department of Justice on this matter. Sincerely, PHILIP B. HEYMANN Assistant Attorney General Criminal Division By: Robert L. Reuch Deputy Assistant Attorney General Criminal Division [new letter] DEPARTMENT OF THE TREASURY BUREAU OF ALCOHOL, TOBACCO AND FIREARMS WASHINGTON, D.C. 20226 JAN 9 1980 Honorable Birch Bayh United States Senate Washington, DC 20510 Dear Senator Bayh: This is in further reference to your letter of October 15, 1979, inquiring about the accuracy of the National Firearms Registration and Transfer Record which is maintained under the National Firearms Act (NFA). You also seek an explanation for ATF's claim of execu- tive privilege in a pending forfeiture case involving the National Rifle Association with respect to internal memoranda concerning the NFA record system. By letter dated November 16, 1979, you were advised that Attorney General Civiletti had received a similar letter from Senator James A. McClure. The Department of Justice has completed its review of the various 1975 internal Bureau memoranda concerning the National Firearms Registration and Transfer Record and was unable to agree with Senator McClure that a serious problem exists or existed regarding the integrity or accuracy of NFA certifications. In addition, the Department determined that the documents were not material the Government is required to release under the doctrine of Brady v. Maryland, 373 U.S. 93 11963). The criticisms of the registration system raised by the 1975 internal memoranda relate primarily to the existence of misfilings, excess paper in the file, the inability to accumulate rapidly statistical data, inconvenient access to records due to limitations in the filing equipment, and the "charge out" procedure for removing index cards from the file. Immediately after, and in response to those 1975 criticisms, various changes were made in the ATF records and recordkeeping procedures. Those individuals in charge - 2 - Honorable Birch Bayh of the records were given authorization to obtain equipment to change the filing system from Diebold machines, with limited access, to file cabinets with individual folders for each registered possessor of a firearm. In addition, the record system was reorganized and purified through several means. Misfilings were corrected and excess paper unrelated to registration status was eliminated. As you may know, the National Firearms Registration and Transfer Record consists of two separate filing systems. In the first instance, each registered possessor of a firearm has a separate individual file folder within which is a copy of the original registration form for each and every firearm in their possession. These folders are maintained alphabetically by name of registered possessor. Secondly, an index card is prepared, indexing each weapon by serial number. This file is the index-card file; it provides both a back-up and a cross-check for the alphabetical file. Both components are examined in searches for evidentiary purposes and only in the event that a firearm has been registered but both files are missing can an erroneous certification be made. Sworn testimony by ATF personnel taken in connection with the forfeiture case of United States v. Seven Miscellaneous Firearms, Civil Action 78-1338 (D.D.C.) revealed that no one has ever produced a registration form or testified that he had registered a particular weapon after ATF had certified to a court that the person was not the registered possessor of a particular weapon. In this regard it should be noted that ATF presently makes 2,000 certifications a year for use in court proceedings. Significantly, 26 U.S.C. section 5841(e) requires persons possessing registered firearms to retain proof of such registration. You also questioned the Bureau's claim of "executive privilege" in the above case. During the discovery stages of the litigation we refused to produce, certain documents demanded by the National Rifle Association because we believed that the documents were not relevant and that the requested materials were protected from disclosure by the deliberative-process executive privilege which is embodied in the general concept of executive - 3 - Honorable Birch Bayh privilege. This decision was made with the full concurrence of the U.S. Attorney's Office and the Department of Justice. It is clear that one form of executive privilege protects the pre- decisional thought processes of the executive from public disclosure. Various passages of the documents at issue reflected advisory opinions, recommendations, deliberations, and options which comprised part of the process by which Bureau decisions and policies were formulated and were, therefore, within the deliberative process privilege. However, for the reasons explained below, ATF voluntarily waived its claim of executive privilege. The claim of privilege was withdrawn for the following reasons. First, all of the documents were over four years old, thus somewhat lessening the possibility that their release could cause immediate harm to the agency's deliberative processes. Secondly, many of the criticisms discussed in the documents have been corrected by action taken since 1975. Thirdly, and importantly, the complete evidence adduced in the case revealed that the opinions, speculations, and conclusions found in these documents, when put in proper context and perspective, were overstatements which neither impugn the integrity of the records system nor affect the ability of the Bureau to accurately certify to the registration status of a National Firearms Act weapon. We trust that this has been responsive to your inquiry. If we can be of further assistance in any way, please let us know. Sincerely yours, G.R. Dickerson Director