Opinion for the Court filed by ROBB, Circuit Judge.
ROBB, Circuit Judge:
These cross-appeals concern the state secrets privilege and its effect upon a lawsuit filed by the plaintiffs, 27 individuals and organizations formerly active in opposing participation by the United States in the war in Vietnam. The defendants are present and former officials of the National Security Agency (NSA), the Central Intelligence Agency (CIA), the Defense Intelligence Agency (DIA), the Federal Bureau of Investigation (FBI), and the Secret Service. Also joined as defendants are three communications corporations, Western Union International, RCA Global Communications, and ITT World Communications. The plaintiffs allege that the coordinated actions of the defendants violated their rights under the Constitution1 and statutes2 of the United States. Specifically, plaintiffs allege that the NSA conducted warrantless interceptions of their international wire, cable and telephone communications at the request of the other federal defendants and with the cooperation of the corporate defendants. Plaintiffs seek declaratory and injunctive relief as well as damages.
The issue before us is: should the NSA be ordered to disclose whether international communications of the plaintiffs have been acquired by the NSA and disseminated to other federal agencies? The Secretary of Defense avers that admitting or denying [85]*85the acquisitions would reveal important military and state secrets respecting the capabilities of the NSA for the collection and analysis of foreign intelligence.
A brief description of NSA and its functions is appropriate. NSA itself has no need for intelligence information; rather, it is a service organization which produces intelligence in response to the requirements of the Director of Central Intelligence. Intelligence Activities: Hearings Before the Select Comm. to Study Governmental Operations with Respect to Intelligence Activities of the U.S. Senate, 94th Cong., 1st Sess. Vol. V at 9 (1975) (Hearings). The mission of the NSA is to obtain intelligence from foreign electrical communications. Signals are acquired by many techniques. The process sweeps up enormous numbers of communications, not all of which can be reviewed by intelligence analysts. Using “watchlists” — lists of words and phrases designed to identify communications of intelligence interest3 — NSA computers scan the mass of acquired communications to select those which may be of specific foreign intelligence interest. Only those likely to be of interest are printed out for further analysis, the remainder being discarded without reading or review. Intelligence analysts review each of the communications selected. The foreign intelligence derived from these signals is reported to the various agencies that have requested it (Hearings at 6). Only foreign communications are acquired, that is, communications having at least one foreign terminal (Hearings at 9).
Two separate NSA operations are in issue here. From 1967 to 1973 the NSA conducted operation MINARET as a part of its regular signals intelligence activity in which foreign electronic signals were monitored. The second operation, SHAMROCK, employed different methods. It involved the processing of all telegraphic traffic leaving or entering the United States. NSA obtained these telegrams with the cooperation of the corporate defendants, and the telegrams were delivered to NSA in the form of paper tapes microfilm copies, or magnetic tapes.
All material acquired through MINARET and SHAMROCK was processed in the same manner. NSA included on the watch-lists the names of United States citizens which were supplied by the FBI, the Secret Service, the CIA, the Bureau of Narcotics and Dangerous Drugs, and the military intelligence services. These agencies sought information in connection with their responsibilities to investigate such areas as international narcotics trafficking, executive protection, terrorism, and possible foreign influence over domestic organizations. The names of approximately 1200 Americans were included on the watchlists at one time or another and NSA disseminated about 2000 reports to the requesting agencies. The reports were edited or summarized versions of the messages acquired. This procedure was followed with all acquisitions, both MINARET and SHAMROCK, to conceal their source.
The federal defendants responded to the plaintiffs’ allegations concerning both NSA programs by filing a motion to dismiss based upon a formal claim of the state secrets privilege by the Secretary of Defense. In an open affidavit asserting the claim, the Secretary stated that:
Civil discovery or a responsive pleading which would (1) confirm the identity of individuals or organizations whose foreign communications were acquired by NSA, (2) disclose the dates and contents of such communications, or (3) divulge the methods and techniques by which the [86]*86communications were acquired by NSA, would severely jeopardize the intelligence collection mission of NSA by identifying present communications collection and analysis capabilities.
(J.A. 39) Along with the open record affidavit, the Secretary submitted a classified affidavit for in camera examination by the court. After some procedural maneuvering in which the plaintiffs attempted to postpone the in camera inspection by the court and succeeded in obtaining a limited amount of discovery, the District Court upheld the claim of privilege with respect to operation MINARET. The court dismissed the claims which were predicated upon the privileged acquisitions because the ultimate issue, the fact of acquisition, could neither be admitted nor denied.
Regarding the activities pertaining to wire or telegraphic communications alleged to have been sent by certain of the plaintiffs within the United States and to have been acquired by NSA through the SHAMROCK source, the court found
in view of matters which have to date been made public about the SHAMROCK source, the claim of privilege cannot be extended to preclude the federal defendants from admitting or denying the fact vel non of acquisition of a plaintiff’s communication originated in the United States for transmission abroad, where it conclusively can be determined from records and materials now retained by NSA that such communication was obtained through the SHAMROCK source.
(J.A. 112-13) Accordingly, the court ordered the defendants to respond to the allegations in the complaint concerning SHAMROCK materials.
The District Court entered a partial final judgment with respect to the dismissal, see Fed.R.Civ.P. 54(b), and certified the question of the rejection of the state secrets privilege to this court under 28 U.S.C. § 1292(b). The plaintiffs and the federal defendants each appeal from that part of the ruling adverse to them.
The plaintiffs attack the District Court’s ruling on three fronts. They argue first that the procedure followed by the District Court to resolve the state secrets privilege question unfairly denied them an opportunity to litigate their constitutional claims. On the merits, they challenge the substantive conclusion that the mere admission or denial of acquisition is a state secret. Alternatively, plaintiffs contend that assuming the state secrets question was properly resolved, dismissal is inappropriate because they could go forward with their claims based upon confirmation of the existence of any of their names on the watchlists. The federal defendants support the court’s ruling with respect to MINARET, but contend that the court incorrectly denied the claim of privilege with respect to the SHAMROCK source.
We conclude that the decision of the District Court was procedurally sound, that the court correctly determined the state secrets question regarding MINARET and that the mere existence of any of the plaintiffs’ names on a watchlist is insufficient to maintain this action. Further, we hold that the court erred in failing to recognize the privilege with respect to the SHAMROCK source.
PROCEDURE
Plaintiffs argue that the procedures followed by the District Court yielded extraordinary control over this litigation to the Secretary of Defense and unfairly denied them their right to litigate their claims. The focus of the attack is upon the court’s consideration of three in camera affidavits and the in camera testimony of the Deputy Director of NSA.
It is settled that in camera proceedings are an appropriate means to resolve disputed issues of privilege, see Kerr v. United States District Court, 426 U.S. 394, 405-06, 96 S.Ct. 2119, 48 L.Ed.2d 725 (1976); United States v. Nixon, 418 U.S. 683, 714—15, 94 S.Ct. 3090, 41 L.Ed.2d 1039 (1974); United States v. Reynolds, 345 U.S. 1, 10, 73 S.Ct. 528, 97 L.Ed. 727 (1953), albeit one to be invoked cautiously, Vaughn v. Rosen, 157 U.S.App.D.C. 340, 345, 484 F.2d 820, 825 [87]*87(1973), cert. denied, 415 U.S. 977, 94 S.Ct. 1564, 39 L.Ed.2d 873 (1974). Plaintiffs do not quarrel with this proposition; rather, they argue that the District Court failed to follow the procedures outlined in our opinion in Phillippi v. CIA, 178 U.S.App.D.C. 243, 546 F.2d 1009 (1976), a Freedom of Information Act case in which the CIA, on grounds of national security, refused either to confirm or deny the existence of requested records. Sensitive to the difficulty of making decisions in camera “without benefit of criticism and illumination by a party with the actual interest in forcing disclosure,”4 we held that the Agency should
provide a public affidavit explaining in as much detail as is possible the basis for its claim that it can be required neither to confirm nor to deny the existence of the requested records. The Agency’s arguments should then be subject to testing by appellant, who should be allowed to seek appropriate discovery when necessary to clarify the Agency’s position or to identify the procedures by which that position was established. Only after the issues have been identified by this process should the District Court, if necessary, consider arguments or information which the Agency is unable to make public.
178 U.S.App.D.C. at 247, 546 F.2d at 1013 [footnote omitted].
Specifically, plaintiffs argue that before the court considered in camera material, they should have been afforded an opportunity to depose the Secretary of Defense, or to propound interrogatories to test his affidavit asserting the state secrets privilege. The District Court refused to permit oral examination of the Secretary, either by deposition or before the court, but the court did rule that plaintiffs could have discovery concerning SHAMROCK through interrbgatories approved in advance by the court.
Accordingly plaintiffs filed their first set of interrogatories to the Secretary. The court refused to approve these because in the court’s opinion they sought “the ultimate answers to questions which defendants have claimed could not be answered without jeopardizing national security issues”; instead the District Court formulated its own interrogatories to which the Secretary responded. Arguing that the answers to the court’s questions raised new issues which they were entitled to explore on the public record, plaintiffs then submitted a second set of interrogatories. Defendants responded to the second set of interrogatories with an in camera affidavit by the Director of NSA, and proffered the ex parte, in camera testimony of the Deputy Director of NSA. At this point the District Court reviewed the in camera affidavits and heard the ex parte, in camera testimony of the Deputy Director.
Plaintiffs contend that the District Court erred in not accepting their first set of interrogatories which asked defendants to explain how answers to certain questions formulated by plaintiffs would jeopardize the intelligence mission of NSA. We agree that their questions properly asked for an explanation of the assertion that the confirmation or denial of the allegations in the complaint with respect to SHAMROCK would compromise state secrets. But the questions propounded by the court likewise asked for an explanation; therefore, we find no prejudicial error in the restatement of the interrogatories by the court.
In the peculiar circumstances of this case, we think the limited amount of discovery permitted here satisfied the concerns expressed in the Phillippi case. This is not a Freedom of Information Act case with large amounts of material at issue, necessitating careful procedures to enable the parties to sort out material that can be disclosed and to focus the court’s attention on material about which there is a genuine issue. In such a case, counsel’s ingenuity in restating a request or in agreeing upon certain exclusions often produces all the material desired. In the case before us the acquisition of the plaintiffs’ communications is a fact vital to their claim. No amount of ingenuity of counsel in putting questions on discovery can outflank the [88]*88government’s objection that disclosure of this fact is protected by privilege. Thus, in these special circumstances, we conclude that affording additional discovery for the government to parry the plaintiffs’ request would be fruitless. In camera resolution of the state secrets question was inevitable.
Plaintiffs argue that once the District Court decided to conduct in camera review of the affidavits and to hear in camera testimony, counsel for the plaintiff should have been permitted .to participate under a protective order. Plaintiffs rely upon our recent decisions in Dellums v. Powell, 182 U.S.App.D.C. 244, 253, 561 F.2d 242, 251, cert. denied, 434 U.S. 880, 98 S.Ct. 234, 54 L.Ed.2d 160 (1977) and Black v. Sheraton, 184 U.S.App.D.C. 46, 60, 564 F.2d 531, 545 (1977) as well as the Supreme Court’s decision in United States v. Nixon, 418 U.S. 683, 715 n. 21, 94 S.Ct. 3090, 41 L.Ed.2d 1039 (1974). None of these decisions, however, involved a state secrets claim; in fact, we expressly distinguished the privilege at issue in the Dellums and Black cases from one in which a state secrets privilege is claimed. Dellums, 182 U.S.App.D.C. at 248, 561 F.2d at 246; Black, 184 U.S.App.D.C. at 57, 564 F.2d at 542. And the Court in United States v. Nixon strongly indicates that state secrets are not analogous to the Presidential communications at issue there. 418 U.S. at 710, 94 S.Ct. 3090. Moreover, the Special Prosecutor, who the Court indicated could be included in the in camera proceedings ordered in that case, was himself an employee of the government.
A ranking of the various privileges recognized in our courts would be a delicate undertaking at best, but it is quite clear that the privilege to protect state secrets must head the list. The state secrets privilege is absolute. However helpful to the court the informed advocacy of the plaintiffs’ counsel may be, we must be especially careful not to order any dissemination of information asserted to be privileged state secrets. “It is not to slight judges, lawyers, or anyone else to suggest that any such disclosure carries with it serious risk that highly sensitive information may be compromised.” Alfred A. Knopf, Inc. v. Colby, 509 F.2d 1362, 1369 (4th Cir.), cert. denied, 421 U.S. 992, 95 S.Ct. 1999, 44 L.Ed.2d 482 (1975). We agree with the court in Jabara v. Kelley, that
[i]n the case of claims of military or state secrets’ privilege, however, the superiority of well-informed advocacy becomes less justifiable in view of the substantial risk of unauthorized disclosure of privileged information. As the court stated in Heine v. Raus, 399 F.2d 785, 791 (4th Cir., 1968):
Disclosures in camera are inconsistent with the normal rights of a plaintiff of inquiry and cross-examination, of course, but if the two interests cannot be reconciled, the interest of the individual litigant must give way to the government’s privilege against disclosure of its secrets of state.
75 F.R.D. 475, 486-87 (E.D.Mieh.1977). Protective orders cannot prevent inadvertent disclosure nor reduce the damage to the security of the nation which may result. Therefore we reject the plaintiffs’ argument that counsel should have been permitted to participate in the in camera proceedings below. See Phillippi v. CIA, supra, 178 U.S.App.D.C. at 247, 546 F.2d at 1013.5
[89]*89Finally, plaintiffs contend that the District Court’s “nearly verbatim adoption of findings of facts prepared by defendants’ counsel compounded the error in denying plaintiffs any meaningful role in [the] ease.” Brief at 34 [footnote omitted]. The argument is without merit. The proposed order and memorandum submitted by defendants reflected the tentative conclusions expressed by the court during a status hearing. We have no reason to suspect this case received less than full consideration by the District Court.
Therefore we reject the procedural attack upon the decision of the District Court and turn our attention to the merits of the privilege claim.
THE STATE SECRETS PRIVILEGE
Plaintiffs argue that the state secrets privilege cannot extend to the “mere fact of interception” of their communications.6 Further, they contend that similar disclosures in another case, Jabara v. Kelley, 75 F.R.D. 475 (E.D.Mich.1977) have occurred without repercussions on the national security.
In his initial assertion of the privilege, quoted above at 5, the Secretary asserted that NS A intelligence collection and analysis capabilities would be jeopardized if he were required to identify whose foreign communications were acquired, or to disclose the dates or contents of the acquired communications. He also stated that he could not divulge the methods or techniques employed by NS A without endangering the Agency’s ability to carry out its mission. Plaintiffs interpret the Secretary’s position to mean that identification of the circuits which NS A monitors would jeopardize national security by alerting targets of foreign intelligence interest that messages sent over those circuits are acquired. Plaintiffs suggest, however, that admission or denial of the fact of acquisition of their communications without identification of acquired messages would not reveal which circuits NS A has targeted or the methods and techniques employed.
The plaintiffs’ argument is naive. A number of inferences flow from the confirmation or denial of acquisition of a particular individual’s international communications. Obviously the individual himself and any foreign organizations with which he has communicated would know what circuits were used. Further, any foreign government or organization that has dealt with a plaintiff whose communications are known to have been acquired would at the very least be alerted that its communications might have been compromised or that it might itself be a target. If a foreign government or organization has communicated with a number of the plaintiffs in this action, identification of which plaintiffs’ communications were and which were not acquired could provide valuable information as to what circuits were monitored and what methods of acquisition were employed. Disclosure of the identities of senders or recipients of acquired messages would enable foreign governments or organizations to extrapolate the focus and concerns of our nation’s intelligence agencies.
It requires little reflection to understand that the business of foreign intelligence gathering in this age of computer technology is more akin to the construction of a mosaic than it is to the management of a cloak and dagger affair. Thousands of bits and pieces of seemingly innocuous information can be analyzed and fitted into place to reveal with startling clarity how the unseen whole must operate. As the Fourth Circuit Court of Appeals has observed:
The significance of one item of information may frequently depend upon knowledge of many other items of information. [90]*90What may seem trivial to the uninformed, may appear of great moment to one who has a broad view of the scene and may put the questioned item of information in its proper context. The courts, of course, are ill-equipped to become sufficiently steeped in foreign intelligence matters to serve effectively in the review of secrecy classifications in that area.
United States v. Marchetti, 466 F.2d 1309, 1318 (4th Cir.), cert. denied, 409 U.S. 1063, 93 S.Ct. 553, 34 L.Ed.2d 516 (1972); cf. Bell v. United States, 563 F.2d 484, 487 (1st Cir. 1977).
The standard of review here is a narrow one. Courts should accord the “utmost deference” to executive assertions of privilege upon grounds of military or diplomatic secrets. United States v. Nixon, 418 U.S. 683, 710, 94 S.Ct. 3090, 41 L.Ed.2d 1039 (1974). The court need only be satisfied that “there is a reasonable danger that compulsion of the evidence will expose military matters which, in the interest of national security, should not be divulged.” United States v. Reynolds, 345 U.S. 1, 10, 73 S.Ct. 528, 533, 97 L.Ed. 727 (1953) [emphasis added]. We note that in the analogous context of the national security exemption in the Freedom of Information Act courts should accord “substantial weight” to the affidavit of the agency. S.Rep.No.1200, 93d Cong., 2d Sess. 12 (1974) U.S.Code Cong. & Admin.News 1974, p. 6267 (Conference Report); 120 Cong.Rec. 36,870 (1974) (remarks of Sen. Muskie); Goland v. CIA, No. 76-1800 Slip Op. at 20 n.64 (May 23, 1978); Weissman v. CIA, 184 U.S.App.D.C. 117, 122, 565 F.2d 692, 697 & n.10 (1977).
The Secretary has asserted in his public affidavit that confirmation of the identity of individuals or organizations whose communications were acquired by NSA “would severely jeopardize the intelligence collection mission of NSA by identifying present communications collection and analysis capabilities.” Supra, at 86, 598 F.2d at 5. In most cases this would be sufficient to sustain the claim of privilege. Here, however, plaintiffs’ suit depends upon the discovery of this information. Because it is the showing of necessity that determines how deeply the court must probe to satisfy itself of the validity of the claim, United States v. Reynolds, supra, 345 U.S. at 11, 73 S.Ct. 528, the court below examined the in camera affidavits and testimony. We think this was proper. Moreover, we have reviewed the in camera materials ourselves and they reinforce our conclusion from the open affidavits that the state secrets claim must be upheld. The identification of the individuals or organizations whose communications have or have not been acquired presents a reasonable danger that state secrets would be revealed.
Plaintiffs’ second argument is that the government has admitted NSA acquisition of communications in another case, Jabara v. Kelley, supra, which plaintiffs claim is indistinguishable from the case here. In the Jabara case the government admitted that six of the plaintiff’s communications had been acquired by the NSA. The genesis of that disclosure is unclear. The government tells us it was inadvertent that the court identified the NSA as the acquiring agency after the FBI had admitted receiving summaries from “another agency.” The plaintiffs retort, with some force, that after this singular disclosure was made by the court, the government responded to interrogatories admitting the acquisition and giving the approximate dates.
The precise circumstances of the disclosure in the Jabara case, however, need not concern us. Whether the disclosure there was inadvertent or intentional is irrelevant here. The government is not estopped from concluding in one case that disclosure is permissible while in another case it is not. As we have said, the identity of particular individuals whose communications have been acquired can be useful information to a sophisticated intelligence analyst. We see nothing inconsistent with the Secretary’s assertion of the privilege here and the disclosure that occurred in the Jabara case.
The official defendants appeal from the District Court’s decision not to dismiss the allegations of NSA activity di[91]*91rected at the plaintiffs’ outgoing international telegrams, the so-called SHAMROCK source. As we have seen, supra at 86, 598 F.2d at 5, the court found that the claim of privilege could not preclude confirmation that a plaintiff’s foreign telegrams were or were not acquired by NSA through the SHAMROCK source. The court thought congressional committees investigating intelligence matters had revealed so much information about SHAMROCK that such a disclosure would pose no threat to the NSA mission.
In response to the District Court’s interrogatories questioning the claim of privilege with respect to the SHAMROCK source, the Secretary filed ¿n open affidavit in which he averred that all acquisitions, whether MINARET or SHAMROCK, were processed identically. The reports consisted of edited and summarized versions of the messages. He further asserted that the original messages had been destroyed. The corporate defendants also discarded copies of the traffic they handle.
The District Court restricted its order respecting communications of a plaintiff to instances “where it conclusively can be determined from [NSA] records that [a] communication was obtained through the SHAMROCK source.” Supra at 86, 598 F.2d at 5. Even if some SHAMROCK items might be segregated by some means unknown to us, we think the affidavits and testimony establish the validity of the state secrets claim with respect to both SHAMROCK and MINARET acquisitions; our reasoning applies to both. There is a “reasonable danger”, United States v. Reynolds, supra, 345 U.S. at 10, 73 S.Ct. at 528, that confirmation or denial that a particular plaintiff’s communications have been acquired would disclose NSA capabilities and other valuable intelligence information to a sophisticated intelligence analyst.
PRESUMPTION OF ACQUISITION
Alternatively, plaintiffs contend that even if the state secrets privilege does extend to the fact of acquisition, the inclusion of a plaintiff’s name on any watchlist submitted to NSA by the CIA, FBI, DIA, or the Secret Service, presents a prima facie case of acquisition; therefore, dismissal was inappropriate. The argument has superficial appeal. The Secretary has not asserted that the lists submitted to NSA by these agencies are state secrets. The disclosures by the Senate Select Committee on Intelligence indicate that some 2000 reports were distributed by the NSA from 1969 to 1973. Final Report of the Senate Select Comm. to Study Governmental Operations with Respect to Intelligence Activities, S.Rep.No. 755, 94th Cong., 2d Sess. Vol. III at 743 (1976). A total of 1200 names of Americans appeared on the watchlists during this period. Hearings, Vol. V at 12. Thus if any of the 27 plaintiffs was among these 1200, it is possible that such plaintiff’s international communications were acquired. Reflection, however, reveals difficulties in proceeding with this action on the basis of such a presumption.
The underlying premise of the argument is that the defendants should not be permitted to avoid liability for unconstitutional acts by asserting a privilege which would prevent plaintiffs from proving their case. Cf. Alderman v. United States, 394 U.S. 165, 184-85, 89 S.Ct. 961, 22 L.Ed.2d 176 (1969). The premise is faulty. The defendants are not asserting the privilege to shield allegedly unlawful actions; the state secrets privilege asserted here belongs to the United States and is asserted by the United States which is not a party to the action. It would be manifestly unfair to permit a presumption of acquisition of the watchlisted plaintiffs’ international communications to run against these defendants.7
[92]*92Moreover, before we could recognize a presumption such as plaintiffs urge upon us we would need to know with some certainty that the existence of a name on a watchlist creates a reasonable likelihood that warrantless acquisitions of communications sent by or to that individual have occurred. Yet the available information points to no such correlation. The watchlists were used to select out communications of interest to the intelligence community; thus it appears that if a message only mentioned a name on the watchlist it would be selected by the computers for further processing. To the extent the reports resulted from acquisitions that only mentioned individuals on the watchlist, they are not relevant to this action. We have no way of knowing, without requiring disclosures which we have held to be privileged, how many of the 2000 reports generated by NSA were to or from individuals on the watchlist and how many were merely those mentioning them.8 With such uncertainty about the number of relevant reports, the conclusion that the mere existence of a name on the watchlist indicates that one or more of the individual’s communications has been acquired and analyzed by NSA is not reasonable. Therefore, we must reject the plaintiff’s argument that the acquisition of a plaintiff’s communications may be presumed from the existence of a name on the watchlist. Not only would such a presumption be unfair to the individual defendants who would have no way to rebut it, but it cannot be said that the conclusion reasonably follows from its premise.
CONCLUSION
We conclude that the Secretary’s claim of privilege should be upheld in its entirety. Therefore, that part of the District Court’s order rejecting the claim of privilege and requiring the defendants to respond to the allegations in the complaint referring to operation SHAMROCK is reversed. In all other respects the decision is affirmed. The case is remanded to the District Court for further proceedings consistent with this opinion.
So ordered.