94 F.R.D. 600 | S.D.N.Y. | 1982
Discovery Order No. 24, issued by Magistrate Sinclair on April 6,1982 recommended that monetary sanctions be imposed against the governmental defendants for their failure to obey discovery orders and recommended that any future failures of the governmental defendants to comply with discovery orders be met with proof-preclusive or case-dispositive sanctions.
Opinion and Discovery Order No. 26, issued by Magistrate Sinclair on April 8, 1982, outlined procedures to be employed in gathering evidence from F.B.I. informant files without violating the confidential nature of these materials.
The parties did not file objections or appeals from Discovery Order No. 24 or Opinion and Discovery Order No. 26, and their time to make objections and appeals has expired. See 28 U.S.C. § 636 (allowing ten days); S.D.N.Y. Magistrate’s Rule, Rule 7.
The findings, conclusions and recommendations in Discovery Order No. 24 and Opinion and Discovery Order No. 26 are hereby adopted in their entirety and made the order of this Court.
So Ordered.
DISCOVERY ORDER NO. 24
KENT SINCLAIR, Jr., United States Magistrate:
During the Spring and Summer of 1981 plaintiffs repeatedly voiced an intention to move for discovery sanctions against the governmental defendants. A hearing was held on September 9, 1981 to allow the parties an opportunity to present relevant evidence on this and other topics. Plaintiffs followed the hearing with a formal motion under Rule 37, F.R.Civ.P., for sanctions against “the defendants Attorney
A. BACKGROUND.
Plaintiffs seek sanctions for these defendants’ “failure to comply with the August 14, 1980 and March 9, 1981 orders of this court and for the government’s unjustified litigation of the need to submit claims of executive privilege and other privileges to the court for adjudication.” Plaintiffs’ Memorandum of Law in Support of Motion for Sanctions, dated November 16, 1981 (“Pits. Mem.”) at 1. The defendants deny that their litigation of the need to submit privilege claims to the court was unjustified. However, for the most part defendants either admit or do not deny their failures to provide discovery; rather, they resist sanctions on the grounds that “when all the facts are considered ... defendants are [obviously] doing everything within reason to comply with this Court’s discovery orders.” Defendants’ Memorandum of Law In Opposition to Plaintiffs’ Motion Pursuant to Fed. R.Civ.P. 37 For Sanctions, dated December 31, 1981 (“Defts.Mem.”) at 12-13.
Both parties have insisted that a review of the full record is required to place this dispute in its framework. Moreover, “sanctions must be weighed in light of the full record in the case.” Cine Forty-Second Street Theatre Corp. v. Allied Artists Pictures Corp., 602 F.2d 1062, 1068 (2d Cir. 1979). The ensuing recapitulation of events is based on the court’s independent review of the entire record and draws on the parties’ versions thereof.
1. Judge Brieant’s August 1980 Order.
This hotly contested case has heavily taxed the court’s resources from its incipiency in 1977. Despite Judge Brieant’s frequent admonitions that good faith negotiations would obviate the need for constant judicial intervention in complex discovery matters, the parties proved unable to resolve the significant disputes bogging down the action, and, in November 1979, massive discovery motions were filed by the plaintiffs. These motions were opposed by the government. The argumentation on these motions was extraordinary in its detail and consumed many months of briefing.
On August 14,1980, Judge Brieant decided these motions. He ordered the defendants to conduct the following discovery: (a) a search of the ELSUR index
2. The Magistrate’s March 9, 1981 Order.
Upon learning that Judge Brieant had assigned the case to me for all pretrial purposes, I scheduled a conference for October 27, 1980. At that time, defendants were directed to perform, inter alia, the following tasks: (1) deliver the Doron Weinberg JUNE documents for in camera inspection by November 16, 1980; (2) produce FBI Headquarters “see-references”
It became clear even before the October 27,1980 conference that the parties did not share an understanding of how and when Judge Brieant’s August 14, 1980 Order would be implemented. Another issue which emerged in pre-conference papers and at the hearing itself was the question of what to do about the numerous documents and parts of documents withheld from production on the grounds of privilege. A briefing schedule was set up and another conference was scheduled. See Tr. of 10/27/80 Conference at 70.
On the privileges issue, the defendants took the position that, although the law clearly required submission of privilege claims for review by the court, it would be unduly burdensome for the defendants to submit such claims in this case because the number of claims was extremely large and would make presentation burdensome. See, e.g., Defendants Memorandum in Opposition to Plaintiffs’ Motion to Compel, dated October 29, 1980 (“Defts.Mem. of 10/29/80”).
As defendants outlined the burdens of privilege claim submissions at the ensuing conference, November 25,1980, a somewhat startling fact emerged: when the defendants produced the F.B.I. Headquarters (“H.Q.”) files on the NLG, many, if not most, documents were produced in redacted form but no record had been kept of why the deletions were made. Thus, documents were produced that were blank; documents were produced that had been subjected to extensive deletions; documents were produced that were partially deleted — and there was no way to tell if the deletions were made on the ground of executive privilege, informants’ privilege, attorney-client privilege or some other ground. Indeed, the F.B.I. did not know either and could not explain the H.Q. file deletions without starting from scratch and reviewing each page once again and marking each redaction with a deletion code.
It also developed that some H.Q. file documents were deleted on the grounds that they or portions of them were irrelevant (“IB” deletions); this assessment was made by F.B.I. staff at a time when the defendants were utilizing an improper, overly narrow standard of relevancy. See August 14, 1980 Opinion. Since no record of these deletions had been kept, it was, similarly, impossible to determine whether a deletion was on relevancy grounds or some other ground without starting from scratch and
In short, in late 1980 it became sadly clear that a substantial portion of previous document production had been handled in such a manner that complete re-production was required if the defendants were to comply with Rule 26, as applied to this case by Judge Brieant, and if the defendants were to be required to submit claims of privilege for court review, as required under the law and Orders in this case.-
A second topic raised at the November 25, 1980 conference was the concept of a rolling discovery schedule. Plaintiffs expressed their desire that the defendants produce the fruits of their discovery tasks on a periodic or “rolling” basis, rather than waiting until the entire task was completed before anything was produced. Defendants stated that they did not see why discovery could not be done that way and offered to produce the fruits of discovery once each month. Tr. at 59-60.
A third topic was also brewing: F.B.I. informant files. The F.B.I. estimated at the time that around 1300 of the informants it had employed over the years had reported about NLG related matters and had been involved in defendants’ NLG related activities. The F.B.I. estimated that some 4,400,-000 documents were contained in the files of these informants. Plaintiffs wanted this material, and the defendants sought to have it deemed privileged from disclosure. The treatment of these materials had not been presented to Judge Brieant for a ruling, and the document production ordered by him did not specifically include these informants’ files.
On January 5, 1981, I issued a Memorandum Opinion which held that the burdens entailed in the submission of defendants’ privilege claims for review were not so great as to exempt defendants from the obligation properly to present such claims. The law simply does not permit parties to arrogate to themselves the right to make final evidentiary privilege determinations. The January 5,1981 opinion also proposed a schedule for some discovery items, exclusive of future document production (and related privilege claims) ordered by Judge Brieant and supplementary interrogatory answers to be culled from that future production. The parties were permitted time to comment on the proposed schedule.
After the submission of extensive comments on the matters raised in the January 5, 1981 Opinion, a Memorandum and Order directing discovery from defendants was issued on March 9, 1981. The discovery obligations imposed by it were understood by defendants to involve:
(a) submission of affidavits regarding the SAC files and “Do Not File” documents, the HUNTER files, and field office responses to production instructions;
(b) submission of claims of privilege for the Weinberg JUNE documents; [(c) reprocessing of H.Q. files to affix a deletion . code and reprocessing of selected IB deletions and all 1A deletions]; (d) submission of an affidavit by one of the Felt-Miller prosecutors; and (e) submission of thirteen informant files consisting of more than 350 volumes for review by the Court.
... [(f)] supplementation of answers to interrogatories 1, 2, 11, 12 and 22 of the
Defts.Mem. at 5-6 (footnote omitted). These tasks were to be performed within 60 to 90 days. The March 9 Order did not order a schedule for the new document production necessitated by Judge Brieant’s August 1980 decision (estimated by the government to involve 550,000 pages); it did not order a log prepared now for all privilege claims that might arise out of the new production (the defendants understood this, see Memorandum In Support of Defendants’ Appeal From Magistrate’s Order . .. dated 4/13/81 (“Defts.App.”) at 12; it did not order production or deny protection of the informant files (estimated at that time by defendants to contain 4,400,000 documents); it did not schedule the provision of supplementary interrogatory answers that Judge Brieant had ordered to be given in connection with the new document production. As to these items for which no schedule was established, the parties were required to (a) submit plans for disposition of the informants’ files; and (b) meet and negotiate a schedule for document production, privilege claim review, and provision of supplementary answers to interrogatories on a monthly, rolling basis. See March 9, 1981 Memorandum and Order at 12-15.
3. The Spring and Summer of 1981.
At the end of March, 1981, defendants submitted letters and affidavits which outlined their views on the March 9 Order. They “informed” the court that the small items could be complied with,
The defendants’ position was a surprising one for a number of reasons that went beyond its unmitigated arrogance. First, they did not challenge the substance of the March 9, 1981 Order, only the time limits it imposed. This was surprising because the 60-90 day time limit imposed (a) related to some tasks, like supplementing interrogatory answers, that had been ordered performed six months before by Judge Brieant; and (b) defendants’ time to comply would be extended by any appeal to Judge Brieant. March 9, 1981 Order at 11.
Second, one reason given by them for their alleged inability to comply with the March 9 Order was that they preferred to work on discovery tasks of their own choosing. Defendants had learned, subsequent to issuance of the March 9 Order, that they had bungled virtually a11 their prior production. The March 9 Order had directed them (a) to affix a deletion code to redactions in documents produced from “Headquarters files,” see id. at ¶ 3(c); Order of January 5, 1981 at 6-7; (b) to review headquarters documents redacted on relevancy grounds and non-headquarters documents redacted on relevancy grounds identified by plaintiffs, see March 9 Order at 6, ¶ 3(c); to review all documents deleted on 1A grounds. See ¶3^). The defendants informed the court that the great bulk of this task — coding the F.B.I.H.Q. file on the NLG
Third, as became more fully documented when the defendants appealed the time limits of the March Order to Judge Brieant (a process that lasted nearly 3 months), the entire claim of inability to comply was swimming with red herrings and took the essential form of a non sequitur. To begin with, the defendants falsely stated to Judge Brieant that the March 9 discovery order required them within 90 days to perform all the new production (an estimated 550,000 pages) ordered in August 1980 by Judge Brieant:
Mr. Gimbel: I think this is set forth in our affidavit, but there are 550,000 pages and that all has to be produced within 90 days. That just can’t be done. }J« sje >}e #
Mr. Gimbel: Your Honor, we are trying. We have 14 people working full time. There are 550,000 documents. * * * * * *
One of two Transcripts dated May 20, 1981 at 9,10. The affidavit referred to is that of Agent Vornberger, dated March 31, 1981 (Exhibit A to Defendants’ Notice of Appeal), which goes on at great length about document production ordered by Judge Brieant on August 14, 1980. This production was not among the tasks directed by me to be done within 90 days; rather, the parties were directed to negotiate a monthly production schedule for this material and for the supplementary interrogatory answers to be culled from it. Yet, the defendants falsely asserted that I ordered these tasks performed within 90 days. Next, the defendants raised objections to the “rolling discovery” concept. They stated that, in their view, it would be better to do one thing at a time. Thus, defendants thought that to avoid a “frightful waste of time”, Defts.App. of 4/13/81 at 11, it would be better to do all the document production ordered by Judge Brieant and then later, sometime in 1984 or 1985 at the earliest, begin supplementing interrogatory answers and, later still, begin making a log and submitting privilege claims. Id. at 11, 12, 13. This, of course, would mean that all documents in the universe of materials ordered produced by Judge Brieant would have to be reviewed three or four times instead of just once, and this would take at least four or five more years at the defendants’ projected pace for the initial review, putting the trial back to 1990! The March 9 Order reasonably required a rolling schedule
The defendants’ final argument to Judge Brieant was that the United States lacked the resources and manpower to comply with the March 9 Order.
On June 3, 1981, Judge Brieant affirmed
Over the next couple of months it became crystal clear that the defendants were not going to comply with the March 9, 1981 Order and the August 14,1980 Opinion. On July 24, 1981, I issued Discovery Order No. 3, which found that the defendants’ careless estimates and projections of the burdens involved in complying with their discovery obligations were unworthy of reliance.
Meanwhile, the court was attempting to fashion a method for balancing the plaintiffs’ discovery rights to, the defendants’ privilege in, and the court’s duty toward the large numbers of documents in informant files. These matters had vexed other courts,
On another front, a July 15, 1981 letter from defendants brought to a close a thoroughly pathetic performance by them of the duty to negotiate in good faith with plaintiffs. In the March 9, 1981 Order, I directed the parties to meet and negotiate and then to submit within one hundred (100) days joint or several plans to the court with regard to a schedule for, inter alia, the document discovery ordered by Judge Brieant. Nothing happened within 100 days, so a thirty-day extension was granted at the parties’ request. The parties attempted to meet and negotiate, but meetings were repeatedly cancelled, and new counsel for defendants — who appeared to know very little about discovery in this case — were assigned this task, necessitating efforts by plaintiffs’ counsel to educate them. Then defendants offered a “comprehensive proposal,” which, incredibly, purported to “offer” the plaintiffs this deal: we will do some of the document production ordered by Judge Brieant in 1980 if you waive all other rights to discovery including all other discovery previously ordered by Judge Brieant. Dollinger to Krinsky, letter of July 8, 1981. The unreality of their “comprehensive proposal” was apparently lost on defendants; a week later, on July 15,1981, they sent me a letter purporting to contain the planning proposal I had given them over 130 days to prepare, and this two and a half page letter was devoid of any specifics and simply repeated the strange offer made to plaintiffs.
4. The September 9, 1981 Hearing and Afterwards.
The September 9 hearing was thereupon held, six months after the March 9 Order and nearly thirteen months after Judge Brieant’s August 1980 Opinion. As of that time, the defendants had not yet completed any of the significant tasks ordered by Judge Brieant and had not yet begun others.
It cannot be said that the defendants disobeyed the black-letter requirements of Judge Brieant’s August 1980 Order, since it did not impose time limits. Quite clearly, however, defendants’ failure to make significant progress with respect to the obligations imposed on them by that Opinion was and is utterly inconsistent with that Opinion and the frequent, unequivocal oral directives of the Judge.
Moreover, it is admitted that the defendants were in default of the March 9 Order to the extent outlined above.
Defendants defend against the imposition of sanctions on the grounds that the United States lacked the resources and manpower to comply with outstanding discovery orders and that their failure was due to inability, not willfulness, bad faith or fault.
In the course of his testimony about the impossibility of compliance with discovery orders, Agent Vornberger testified about three related subjects: the volume of materials subject to discovery, the extent of defendants’ compliance with discovery orders, and the defendants’ current allocation of manpower and resources to their discovery obligations in this case. Pre and post-hearing sworn written statements were also submitted on these three matters.
On the first two matters, several noteworthy facts emerged. Some highlights:
(a) that two packages of ELSUR indexed documents had been sent to New York for production, but after Judge Brieant ordered them produced in August 1980, the defendants’ New York counsel failed to turn them over for sixteen months. His explanation: “I had forgotten about the material ...” (Salerno Declaration, dated 12/21/81, at ¶ 3);
(b) that the informants’ files are estimated to contain less than 3,000,000 documents, as opposed to the previous estimate of 4,400,000 (Tr. at 110);
(c) that 51 informant field office files subject to outstanding discovery requests have been destroyed since January 16, 1978, most, but apparently not all, being destroyed “in compliance with normal file destruction procedures” (Vornberger Declaration, 11/10/81);
(d) that mishaps, improper and erroneous deletions and other misfeasances that had plagued prior NLG H.Q. file document production had resulted in plaintiffs originally receiving only two-thirds (15,381 pages) of the processed FBI documents that they should have received (22,279 pages) (Vornberger 12/24/81 Declaration, at ¶ 22);
(e) that defendants continue in their inability to give the court a coherent understanding of the scope of their prior document production, now representing that the figure is 94,500, see Defts. App. of 4/13/81; Defts. Mem. at 2; Dollinger
Prefatory to explaining why the United States was unequipped to comply with the discovery orders, Agent Vornberger outlined the present scope of defendants’ resource and manpower commitments to discovery obligations. He stated that a field agent at each of the 59 field offices was involved in retrieving documents and sending them to H.Q. At H.Q., approximately nine people from outside Vornberger’s unit involve themselves and look at the documents to see if there is any classified information in them. Concurrently, about 14 analysts in Vornberger’s unit and some four persons from the P.B.I.’s Office of Legal Counsel conduct a review, principally to make informants’ privilege deletions but to make other deletions as well. Persons from a Messenger Unit, a Service Unit, and a Special File Room also help out by pulling files • from bins and shelves and carrying them around H.Q. Tr. 111-118; 147; 171.
On the issue of why such limited manpower and resources were all that could be allocated by the United States, Vornberger’s testimony was brief. He stated that no more people from his unit could be put on the job because of other obligations; that there is another CDRU but that unit has jobs of its own; that more people cannot be hired because of a hiring freeze; that other F.B.I. people cannot be used because they do not have the requisite training; that people from outside the F.B.I. cannot be used because much of the documents’ contents must be kept secret. Id.
B. DISCUSSION.
Defendants assert that “when all the facts are considered .. . defendants are [obviously] doing everything within reason to comply with this Court’s discovery orders.” Defts. Mem. at 12-13. Defendants apparently mean, by this, that their failures and misfeasances must be excused because they were not intentional but merely the result of understaffing;
Simply put, defendants have done and are doing all that they reasonably can do within the confines of their resources and other discovery obligations. Defendants have not refused to conduct any court-ordered discovery — they merely have been unable to comply within the deadlines imposed.
Defendants therefore submit that they have acted reasonably and exercised utmost good faith in discharging their discovery obligations, and the failure to comply with discovery schedules is substantially justified when their limited resources are compared with the extent of plaintiffs’ demands.
Defts. Mem. at 32 (original emphasis deleted; new emphasis added). Analysis of this argument requires a discussion of (1) defendants allocation of resources and manpower towards meeting its burdens; and (2) the scope of the discovery burdens resting on the defendants.
1. Manpower and Resources.
Except for document retrieval and classification review, Vornberger’s CDRU is doing all the significant work. Vornberger says that, on the average, 14 analysts are working full time, but a review of the declaration filed by him demonstrates that the number is closer to twelve.
2. Burdens.
Defendants have never had any trouble getting Judge Brieant or me to agree that “significant discovery responsibilities of a large scope rest on the government.” Discovery Order No. 3 at p.2. Defendants have, however, been less successful at getting the court to agree with their assessment of where the burden comes from and what its approximate scope really is.
In defendants’ view, plaintiffs have a gluttonous “appetite”, Defts. App. of 4/13/81 at 14, for discovery and are wont to make “extraordinarily voluminous demands,” Defts. Mem. at 29, and this is primarily where the burden comes from. Defendants seem to think that the plaintiffs are more interested in disrupting their day-to-day activities than they are in getting relevant evidence and are making discovery requests just to be troublesome and pernicious. The fact is, however, that the defendants have accumulated an extraordinary amount of material about plaintiffs that is relevant under Rule 26. The defendants paid or used over a thousand people to report, inter alia, on plaintiffs, and these people generated an immense amount of material contained in informants’ files. Through these and other means, defendants have amassed a half-million or so more document pages in H.Q. and field office files which defendants concede are relevant to plaintiffs’ claims under Rule 26,
The sheer number of documents is not the only source of burden. As the court has always recognized, the defendants have a strong confidentiality interest in these materials, and the law places special burdens on litigants who wish to protect their confidentiality interests by asserting evidentiary privileges. Before the court can protect defendants from being overly burdened by the rules governing assertions of privilege, however, the court requires that the scope of any such claimed burden be clearly demonstrated, and defendants have resisted every attempt by the court to get them to do this. Two examples are defendants’ footdragging with respect to the production of informant files for in camera review, see supra at pp. 607-608, and defendants’ refusal to prepare a log of privilege claims for documents and portions of documents withheld from production, thus insuring that the court would be kept in the dark and prevented from fashioning an order that would relieve defendants of any burdens relating to these items. Defendants have a right to expect the court to protect them from undue burden, but defendants have been preventing the court from doing that, while at the same time
3. Burdens Flowing From the March 9 Order.
The burdens perceived by defendants to flow from the March 9, 1981 Order were largely the product of their imagination.
The parts of that order that defendants announced that they could not comply with (prior even to trying) were (1) supplementation of interrogatory answers within sixty (60) days, a time limit extended by approximately 90 more days, see Order dated March 18, 1981 (requiring commencement only of tasks not subject to reargument or appeal to the Judge); and (2) preparation of a log for documents withheld from prior production on grounds of privilege
In sum, the March 9 Order did not require the impossible. Defendants had to manufacture burdens that were not in the Order just to make out a colorable argument that compliance was impossible. In
C. SANCTIONS MUST BE IMPOSED.
I find that the March 9, 1981 Order could have been complied with given the present level of manpower and resources devoted to the case. I find also that defendants have failed to offer evidence to the contrary. I further find that, had they wanted to, the defendants could have allocated more manpower and resources to their discovery obligations in this case.
Provisions of the March 9 Order with which the defendants failed timely to comply are ¶ 2(a) (supplementary interrogatory answers); ¶ 3(a) (privilege log); ¶ 3(b) (a monthly installment of privilege claims); ¶ 3(c) (reprocessing of selected IB deletions); ¶ 3(d) (reprocessing of 1A deletions);. ¶ 4 (presentation of plan for discovery completion). ¶¶3(<:) and (d) have now been substantially satisfied, except possibly for some see-reference documents.
Upon a thorough review of the “history of this litigation.” Link v. Wabash Railroad, 370 U.S. 626, 628, 82 S.Ct. 1386, 1387, 8 L.Ed.2d 734 (1962), and the “totality of the evidence,” Ohio v. Crofters, Inc., 74 F.R.D. 12, 15 (D.Colo.), aff’d sub. nom., Ohio v. Arthur Andersen & Co., 570 F.2d 1370 (10th Cir.), cert. denied, 439 U.S. 833, 99 S.Ct. 114, 58 L.Ed.2d 129 (1978), I find that defendants’ conduct in failing to comply with discovery orders
Rule 37(b) provides a court with wide discretion in selecting an appropriate sanction. Fines and costs may be imposed, even against the United States,
I find that there is no present basis for imposing the harshest sanctions. Trial is, regrettably, still far in the future, and, at the present time, imposition of the harshest sanctions would be out of proportion to the culpability of defendants and the prejudice suffered by plaintiffs.
D. CONCLUSION.
A case such as this, filled as it is with complex substantive and discovery issues and carrying the seeds of great burdens, places special demands on the parties and attorneys involved. The parties must recognize that once they fail to resolve discovery issues whose parameters they know better than anyone, they must be prepared to adhere to the orders of the court whose assistance they seek, whether or not they agree with those orders. The alternative is utter chaos. The parties’ attorneys must recognize that they are officers of the court, as well as advocates, and as such they have an obligation to work with the court in meeting the challenges a case such as this presents to the judicial system, rather than treating the court as an adversary and hiding the ball. This is a ease that must end some day, not a holy war or an open forum for endless bickering, posturing and willful immersion in collateral issues. This case will come to an end soon, and it is up to the parties whether the issues are resolved on the merits before that happens.
. The ELSUR index is composed of three by five inch index cards containing the name of each person reasonably identified as being overheard during electronic surveillance by the FBI.
. The JUNE designation, until late 1978, was placed on documents which may have related to sensitive investigative techniques.
. The FBI’s Counterintelligence Program (COINTELPRO) generated discrete files containing an estimated 69,000 documents.
. “See-references” are designations in a particular file of a document in a second file that contains significant investigative data about the subject of the first file.
. As defendants stated, reproduction was necessary because there was no contemporaneous record of what was deleted from the NLG “Headquarters file”: “.. . for reasons I cannot explain the code was not used.” 11/25/80 Tr. at 36. Prior to this time, however, defendants had been willing to make unfounded representations about what was or was not deleted. See, e.g., Salerno letter, dated November 18, 1980 (“It is my understanding that, quite simply, the information sought was not deleted from any of the production made heretofore ...”).
. But the provision of the informants’ files for in camera review soon came to involve a series of unusual false starts described infra at 607-608.
. Although the March 9 Order directed coding of “Headquarters files,” the defendants had represented that only the F.B.I.H.Q. files on the NLG were not coded prior to production. Salerno letter of 12/15/80 at 2; 11/25/80 Tr. at 36. They further represented that only about 15,000 documents would have to be coded. 11/25/80 Tr. at 19. Defendants understood the direction to code “Headquarters’ files” as a direction to code only the “NLG Headquarters main file.” Salerno letter of 3/30/81. Besides the NLG H.Q. file, prior H.Q. file production consisted of only one file drawer. Defts. Mem. of 10/29/80 at 2.
. Judge Brieant did not, of course, know that defendants had completely mishandled prior production when he found, in August 1980, that the defendants had not acted in bad faith. Defendants had made much of this prior production to Judge Brieant and had frequently said that it exemplified good faith. Now, defendants stated, there was no way to tell what had been deleted in H.Q. files, Salerno letter of 3/31/81 or if field office file deletions were correct, Vornberger 3/31/81 Declaration, without reviewing the entire production.
. Judge Brieant noted the value of rolling discovery to expedite and narrow discovery. See One of Two Transcripts dated May 20, 1981 at 14.
. On a similar note, the defendants suggested that it would be better to make the Attorney General sit down for a month or whatever it might take to review all executive privilege claims, instead of spending ten minutes or so a week on it. In any event, nothing in the March 9 Order would have prevented them from presenting all executive privilege claims to the Attorney General in one installment.
. Defendants attempt to explain this snafu by asserting that it was then a permissible FOIA practice to release documents without keeping records. Whether or not this is true, no satisfactory explanation has been forthcoming of why FOIA practices were used to respond to a Federal Rules of Civil Procedure discovery request. See Krinsky Aff. of 3/9/82. The FOIA may or may not require submission of privilege claims for review by the court, but the F.R. Civ.P. and case law interpreting the rules do.
. It is not clear that defendants’ false statements had any influence on Judge Brieant’s decision, but these discovery disputes — so complex, so numerous — are difficult enough to come to grips with in the absence of false statements.
. In Discovery Order No. 3, the specific example of such unreliability (there was more than one to choose from) was the defendants’ approach to estimating how many documents were in the informants files. See id. at n.l. It later turned out that the defendants’ “estimate” of 4,400,000 informant file documents was off by at least (and probably much more) 1,400,-000. Yet, defendants fully expected the court to rely on the original estimate and it was submitted in sworn form.
. Back in February 1981, the government offered to produce a witness to testify that the document production ordered by Judge Brieant really amounted to 550,000 documents. Since I intended to leave scheduling of this production to the parties and did not include a schedule for it in the March Order, and since this task was no hindrance to the performance of tasks I ordered, I did not accept this offer.
. See In Re Attorney General, 596 F.2d 58 (2d Cir. 1979).
. I.e., production using representative groups, of main and field office files, COINTELPRO documents, JUNE documents, documents indexed in ELSUR index, see-reference documents had not yet begun, though quite a bit of retrieval had taken place. Defts. Mem. at 13-25 and citations to September 9 Transcript
Other tasks have not been completed. See Pits. Reply Mem. of 2/8/82 at 1-6.
. See, e.g., May 7, 1980 Tr. at 2-4 (“the totality of the picture suggests to me that we are not getting the kind of cooperation on pretrial discovery that needs to be had in a litigation of this sort”); Feb. 15, 1980 Tr. at 60 (“you [defendants] have what they want ... meet them part way and give them as much as they are reasonably entitled to”); October 25, 1979 Tr. at 27 (“The information has to be processed. You have to get the discovery done, and it has to be reached for trial.”); October 2, 1979 Tr. at 28 (“Having given the citizen the right [to information] they have to act in a reasonable time.”); Tr. of November 23, 1977 at 15 (“When the Court enters an order in a case, the Court cannot be put in a position of pointing an empty gun or bluffing, and if I really do enter an order, if I don’t get a factual showing for or a basis for noncompliance of the order I will ... do whatever I have to do.”).
. Since the hearing, defendants have reprocessed the selected IB deletions and 1A deletions, except possibly for some of the “see-reference” production. Defts. Mem. at 13-25.
. The defendants do, in addition, indulge themselves in a cynical attempt to revive the argument put to rest by Judge Brieant and me last Spring. In attempting to excuse their failure to provide any interrogatory supplements, a log and an installment of privilege claims, they argue that these tasks had to await the performance of other tasks. Thus, they state that they preferred first to do a task that the court never ordered, i.e. review and reprocessing other than that required to affix a deletion code to uncoded redactions in documents previously produced from H.Q. files and to remedy selected IB deletions and 1A deletions, and then perform all document production ordered by Judge Brieant, and then rereview all the documents to answer interrogatories and then rereview all the documents a third time to make a log and privilege claims, so as to be ready for trial around 1990. Defendants say that this approach will result in their providing plaintiffs with the “ ‘largest amount of material possible in the least amount of time.’ ” Defts. Mem. at 28, quoting Vornberger Declaration of December 24, 1981 at ¶ 24.
First of all, as discussed previously, this approach will require three or four document reviews instead of just one. Defendants can supplement the interrogatories simply by designating produced documents, so this will not add much time to production. Preparation of a log is a very quick process which would add only seconds to the time required to review each document. So all the tasks ordered can be performed with one document review. The representative group production ordered by the Judge simply has no relation to the tasks ordered by the March 9 Order to be done within 60-90 days. The only document review ordered to be done in that time by the March 9 Order related to the affixing of a deletion code to uncoded redactions in documents previously produced from H.Q. files and selected documents with 1A and IB deletions, and defendants never said that they could not complete these tasks within the allotted time. Second, doing it the court’s way is certain to result in reducing discovery burdens on the court and the parties. As Judge Brieant noted, rulings made early in the rolling schedule will reduce the number of rulings that have to be made
Thirdly, and most importantly, the court ordered defendants to do it one way and they have no right to substitute their judgment for the court’s. Judge Brieant warned the parties that when they sought the intervention of the court there would be three views on how discovery should proceed instead of just two, and the defendants promised to comply with the court’s view. 2/15/80 Tr. at 44-46. Now, after getting the court’s view, they refuse to do what the court ordered, choosing instead to do other, unordered tasks and to conceive of an illogical, wasteful discovery plan that would geometrically increase their discovery burdens and result in great delay.
. Also, Vornberger testified that at least 90% of these documents had nothing to do with this case. My in camera review of the files produced to me tends to confirm Vornberger’s testimony.
. These figures do not include the production of unprocessed documents from H.Q. NLG files, either “7,381 pages” or “more than 7,381 pages”, depending on which contradictory sworn statement is more nearly correct. Compare Vornberger Declarations of 12/24/81 at ¶ 22 and 4/1/82 at ¶ 17. As defendants themselves admit, it is undisputed that only 15,381 pages of processed documents (numbered by them) were originally produced, see Defts. Mem. of 10/29/80 at 2; 11/25/80 Tr. at 19, and that 22,279 pages of processed documents (also numbered by them) were later produced, see Vornberger 12/24/81 Dec. at ¶ 22. This alarming episode in which one third of the concededly producible materials from a batch of production were mishandled is one for which no explanation has been provided. It would never have happened if a log had been prepared in the first place or if defendants took their discovery obligations seriously.
. The seriousness of this very real and ongoing problem bears special emphasis. It is not limited to prior document production; nearly every estimate or statement of their burdens, past, present, or future, that the defendants have made in this case, whether in sworn or unsworn form, has been retracted and replaced (1) with a new one that is significantly different (in some cases 30%-50% different); and (2) without any acknowledgment that prior estimates or statements were wrong. It should be plain to the defendants by now just how difficult this persistent practice makes discharge of the court’s responsibility to guard them from undue burden, yet it does not appear that to date they are taking this concern to heart.
The impression conveyed is that the defendants simply do not care what they represent to the court, and their course of conduct does not inspire confidence in the accuracy of the few burden estimates or statements that have not been revised.
. Since June 1981, Vornberger’s personnel have been putting in an average of 2054 hours a month. Given a 40 hour work week (or 168
. See discussion, infra, at n.30.
. And when used, apparently performed at least as satisfactorily as Vomberger’s unit. F.B.I. agents processed the H.Q. main file on the NLG and did not keep a record of their deletions, assertedly an FOIA practice at the time. Vornberger’s unit did the bulk of the processing and made enough mistakes that, in the F.B.I.’s view, it was necessary to redo their work.
. Defendants acceded to plaintiffs’ interpretation of relevancy under Judge Breiant’s August 14, 1980 Order, objecting only to producing an entire document when only part of it is relevant. See, e.g., Tr. of Oct. 27, 1980 Conference at p. 48; Salerno letter of November 18, 1980.
. After defendants finally managed to produce informant files for in camera review, the court has been able to review them and is considering provisions that will give defendants relief from a potentially great burden.
. Defendants have always understood that the log was to be prepared in the first instance only for previously produced items, then supplemented to include documents withheld from production in the ensuing months. See, e.g., Defts. Appeal of 4/13/81 at 11, 12; Deft. Mem. of 12/31/81 at 6.
. Defendants implied to Judge Brieant (1) that I had ordered all the reprocessing, see Defts. Appeal of 4/13/81 at 8; and (2) that I had ordered “new discovery” amounting to “550,-000 pages of documents.” Id.
First, as mentioned before, the reprocessing I ordered included only (1) affixing of a deletion code to uncoded redacted documents from H.Q. files; and (2) remedying of selected IB deletions and all 1A deletions. Most of the reprocessing was undertaken by defendants “voluntarily],” in “good faith,” and in the absence of court order. Defts. Mem. at 14. Defendants cannot have been confused on this point. See n.7 supra. Subsequent orders made it clear, see.Order dated May 19, 1981 (“While to a very limited degree further work has been required by intervening decisions of this court, the principal reason is that the prior coding used a wrong set of deletion categories”), and they requested clarification only as to their duty with respect to IB and 1A deletions. See Salerno letter, dated Feb. 2, 1981. Defendants stated that they could comply with the reprocessing tasks that I ordered, id. at 7, after they were alerted on January 5, 1981 that they would be required to do them. See also 9/9/81 Tr. at 236-238.
Second, I did not order the new document production defendants referred to; Judge Brieant did. Nor did I order a timetable for this production. I merely asked the parties to negotiate in good faith and present joint or several timetables. (Defendants did not do this.) Performance of this document production was to be done after the tasks ordered on March' 9, and yet the defendants insisted on doing this first, and using this as an excuse for not preparing a log, submitting any privilege claims, or providing supplementary answers to interrogatories. Defendants might have plausibly argued that uncertainty over the scope of informants’ privilege was a hindrance to the preparation of a log. See March 9 Order at 6. But they have not mentioned that this was any hindrance, and have relied solely, and even prior to issuance of the March 9 Order, see Salerno letter of 2/2/81, on the argument that preparation of a log must await completion of all document review and production.
. Defendants’ New York counsel could have been utilized and should be utilized in the future, even if this requires further security clearances for them. Special Agents could have also been used, even if this required a short training period, and they may have to be used in the future.
In other cases, defendants have used special agents — sometimes in large numbers (75-175) —to comply with discovery obligations. See 9/9/81 Tr. at 176-180; American Civil Liberties Union v. City of Chicago, 75 Civ. 3295 (N.D.Ill.); Alliance To End Repression v. Rochford, 74 Civ. 3268 (N.D.Ill.); Socialist Worker’s Party v. Attorney General, et al., 73 Civ. 3160 (S.D.N.Y.). As plaintiffs in a recent anti-trust case, some of these defendants joined in discovery efforts that resulted in the production of 26 million document pages for inspection and copying. Report to the President and Attorney General of the National Commission for the Review of Anti-trust Laws and Procedures, Volume II at p. 22 (1979).
The argument could perhaps be made that separation of powers considerations might inhibit a court from ordering the Executive Branch to devote more energy to its litigation tasks, cf. In Re Attorney General, supra, 596 F.2d at 64, but that argument has not yet been raised here. It would seem, however, that when the United States waives tort immunity, it accepts the responsibilities that go along with the waiver. Congress cannot be thwarted by an executive decision to devote plainly insufficient resources to the litigation tasks that must be performed so that an immunity waiver is not illusory. In United States v. Sumitomo Marine, 617 F.2d 1365, 1370 (9th Cir. 1980), the United States attempted to deflect sanctions by arguing that it had insufficient resources. The court rejected this argument and stated “if the cause of the government’s failures to comply with court orders is understaffing, then perhaps harsh measures will encourage those charged with funding and allocating personnel among the Justice Department’s various offices to take ameliorating action.”
. Plaintiffs also seek sanctions for the government’s opposition to their motion to compel privilege claims. I find that, although not completely justifiable, the government’s opposition can be explained as an attempt to raise the possibility of alternative resolutions to the problem, other than individual submission of claims.
Plaintiffs are apparently no longer pursuing sanctions for document destruction.
. The Equal Access to Justice Act, P.L. 96-481; 94 Stat. 2325 (1980) repealed Rule 37(f), so that costs and fees sanctions may now be imposed against the United States.
The legislative history of this law is strewn with references to the United States and its agencies’ abuse of the litigation process, and Congress expressed, through this law, the judgment that the government may not be treated differently from other litigants. H.R.No.96-1418 (1980).
. Although some false and many wrong statements have been made to the court, I make no finding that defendants have intentionally lied to the court. Nor do plaintiffs allege that defendants have intentionally lied to the court.