This appeal summons us to examine this case a second time. Again at issue is an order of the District Court requiring production by the Department of State1 of certain documents2 requested by the Washington Post Company (the Post) pursuant to the Freedom of Information Act (FOIA).3 In that court the Department contended, as it does here, that revelation of these materials would "constitute a clearly unwarranted invasion of personal privacy," and, accordingly, that they are exempted from mandatory release by FOIA's Exemption 6.4 The District Court entered summary judgment for the Post on the ground that the strong public interest in disclosure of governmental information outweighed what the court deemed would amount to a relatively insignificant invasion of privacy.5

We hold that in the circumstances portrayed by the record, the information solicited by the Post is potentially accessible under FOIA. We find, however, that the District Court undertook to resolve a serious factual dispute concerning the extent of harm consequent upon an unveiling of that information. We therefore reverse the production order and remand the case to that court for further proceedings in harmony with this opinion.

In September, 1979, the Post asked the State Department to release, pursuant to FOIA, any documents indicating whether Dr. Ali Behzadnia and Dr. Ibrahim Yazdi were United States citizens or held valid United States passports.6 At the time, both of these individuals lived in Iran and were prominent figures in the governmental hierarchy of that country. The Department denied the Post's request, invoking FOIA Exemption 6, which authorizes withholding of "personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy."7 The Department asserted that disclosure of any record revealing that Dr. Behzadnia or Dr. Yazdi is a naturalized United States citizen would result in embarrassment and physical harm, and possibly even death.8 This disposition was affirmed by the Department's Council on Classification Policy, which felt that "the privacy interests to be protected are not incidental ones, but, rather are such that they clearly outweigh any public interests which might be served by release of the requested information."9

The Post then instituted the present suit in the District Court to enjoin the State Department from withholding the data requested.10 That court held that the records in question are not "similar files" protected by Exemption 6, and accordingly granted summary judgment for the Post.11 On appeal, this court agreed that the records are not "similar files," and therefore did not consider the possible impact of disclosure upon any privacy interest involved.12 The Supreme Court, however, finding this reading of "similar files" too narrow, reversed.13 The Court held that Congress intended that the phrase be given a broad meaning, and thus to include any file containing "[g]overnment records on an individual which can be identified as applying to that individual."14 This court in turn remanded to the District Court for determination of whether public release of such records would amount to an intrusion upon personal privacy above the statutorily tolerated level.15

Subsequently, the State Department learned that Dr. Behzadnia was no longer living in Iran. The Department then sought to comply with the Post's request as to him, but was unable to locate any material relating to issuance of a United States passport to him or any other document associated with his name.16 From that point the case proceeded with respect to Dr. Yazdi alone.

Both sides moved for summary judgment in the District Court, thus confronting it with the task of balancing the public interest in disclosure against the privacy interest of Dr. Yazdi. The State Department relied on affidavits averring that public dissemination of the information desired by the Post could be embarrassing to Dr. Yazdi and could even expose him to physical harm.17 The Post, on the other hand, pointed to several press accounts of events in Iran, and to books published since the Iranian Revolution that had referred to Dr. Yazdi's reputed ties to the United States. For example, both Inside the Iranian Revolution, by a former director of the State Department operations center, and Mission to Iran, by a former ambassador to Iran, state that Yazdi is a United States citizen.18 Because information of that character was already in the public domain and the dire consequences predicted by the State Department had not occurred, the District Court concluded that the effect on personal privacy from release of the requested documents would be insubstantial.19

Against this personal privacy interest the District Court balanced what it found to be a strong public interest in favor of disclosure. The court identified two elements of this public interest. First, the court observed that whether prominent officials of the Iranian government are American citizens is a legitimate matter of public concern, especially because such information "would shed light on the backgrounds and motivations of these individuals and on the composition of the body of officials exercising political power in Iran."20 Second, the court declared that this information would reveal what steps, if any, the United States Government had taken to revoke the order naturalizing Dr. Yazdi and to cancel his certificate of naturalization.21 The records pursued by the Post might show, the court reasoned, whether governmental officials had been derelict in their duties by failing to institute proceedings toward that end.22 Weighing this public interest against Dr. Yazdi's privacy interest, the court found that disclosure of the sought-after information would not constitute a clearly unwarranted invasion of privacy.23

Time and again, this court has emphasized that FOIA cases are not immune to summary-judgment requirements.24 Only upon a suitable showing "that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law" is summary judgment in order.25 By the same token, an agency may not defeat its opponent's right to an evidentiary hearing on such an issue merely by filing an affidavit purporting to support a motion for such a judgment.26 Rather, "the requester may ... produce countervailing evidence,"27 and if any genuine issue of material fact remains, summary judgment is improper.28

In light of these uncontroversial precepts, the District Court erred in awarding summary judgment to the Post. The record makes abundantly clear a factual dispute going to the very heart of the case: the extent of potential harm to Dr. Yazdi should the Department release information on his citizenship.29 That issue could properly be resolved only by trial, not by factfinding on the basis of materials tendered in support of and in opposition to summary judgment.30

When we review refused FOIA requests, we are often called upon to assess the probable consequences of releasing particular information. And when the litigants quarrel over key factual premises for a determination on that score, we have unhesitatingly ruled that summary judgment is inappropriate. For instance, as our Exemption 431 cases well illustrate, we have frequently found that disputes over the likelihood or extent of harm from disclosure preclude summary judgment:

Where there is a conflict in the affidavits as to what adverse consequences will flow from the revelation of the facts contained in the documents sought to be disclosed, then it appears that there is indeed a conflict regarding very material facts which calls for some type of adversary procedure. The District Court thus attempted to resolve the conflict in the ultimate facts without having the evidence before it.... Summary judgment was not appropriate.32

Similarly, in applying the privacy-balancing test of Exemption 7(C),33 we have held that "differing assessments of the actual harm which disclosure would inflict" generate an issue of fact unsuitable for resolution on summary judgment.34 The same conclusion follows inexorably here.

Facing controverted issues of fact, the District Court proceeded to resolve a pivotal conflict in the affidavits respectively tendered by the Post and the State Department on the cross-motions for summary judgment.35 Courts are forbidden, however, to conduct trial by affidavit and thus deprive litigants of their right to an evidentiary hearing on issues of fact.36 As we have said in the past, "[w]e think there is a right of confrontation ... and so the parties should have the right to examine the affiants either by depositions or in open court.... [T]he case should be tried like any other adversary proceeding."37

This limitation on the use of summary judgment is not a mere technicality. The integrity of a court's de novo judgment rests upon an adversarial system of testing for truth when critical adjudicative facts are subjects of a contest.38 Obviously, the more difficult the issues and equal the weight of expert opinion in a case, the greater the role that process plays in honing the court's judgment. It is no wonder, then, that we have recognized "the advantages of adversary procedures in testing the strength of the government's position in FOIA cases,"39 and have declared that "[t]he importance of maximizing adversary procedures in suits such as this cannot be gain-said."40 On remand, the District Court must heed these teachings, and resort to some salutary alternative to summary judgment.

III. THE DEGREE OF DEFERENCE DUE THE STATE DEPARTMENT

Our dissenting colleague, however, would direct an award of summary judgment, not to the Post, but instead to the State Department. The rationale for that course is that an evidentary hearing of this question would be "wholly unsuited to the capacities of the judicial process."41 Put another way, the dissent contends that we must defer to the State Department's assessment of the seriousness of the threat of harm to Dr. Yazdi because the courts lack the ability to deal with it. This thesis, advanced relentlessly and with singular energy, disregards firm precedent, ignores the lengthening record of satisfactory judicial performance in adjudication of questions of predictive fact under FOIA, and counsels an abdication of judicial responsibility to unbridled Executive Branch discretion in defiance of an explicit congressional directive that courts review agency withholding claims de novo.

When FOIA was originally enacted in 1967, Congress foresaw the need for de novo judicial review42 "in order that the ultimate decision as to the propriety of the agency's action is made by the court and [to] prevent [the proceeding] from becoming meaningless judicial sanctioning of agency discretion."43 In all cases, it was the reviewing court, not the agency subjected to review, that was ultimately to determine the propriety of the agency's action in withholding the requested information.44 In 1974, Congress reaffirmed and strengthened this position in the face of a strenuous effort to legislate a narrower standard for review of Exemption 1 claims premised on exigencies of the national security.45 As we have characterized the result,

[t]he [1974] legislative history underscores that the intent of Congress regarding de novo review stood in contrast to, and was a rejection of, the alternative suggestion ... that in the national security context the court should be limited to determining whether there was a reasonable basis for the decision by the appropriate official to withhold the document.46

In light of this development, how can it be maintained that Exemption 6 privacy claims call for anything less than full de novo review? These claims are very different from Exemption 1 national security claims, and accordingly the arguments for judicial deference in the Exemption 1 context have no force when a court weighs a personal privacy interest. More startlingly, the approach advocated by the dissent would involve deference to a degree even beyond that envisioned by congressional proponents of a narrow scope of judicial review in the national security context. Although Congress expressly provided for de novo review of Exemption 6 claims and did not prescribe any special deference to agency judgments underlying them, the dissent urges us to defer to the State Department, on the grounds that judges know too little about conditions in Iran to second-guess the agency47 and that "any reader of the Washington Post"--including, presumably, judges--can see that the State Department prediction "is very plausible."48

Such an exercise in institutional self-denegration would flatly ignore the oversight function that Congress imposed on the courts. In addition to the express statutory command to review agency withholdings de novo, the legislative record is replete with indicia of congressional confidence in the capacity of the judiciary to evaluate FOIA questions bearing on national security and foreign policy. Before examining the reviewing stance appropriate to disposition of the instant appeal, we revisit the record to demonstrate the unequivocal conviction of Congress that courts can and should review FOIA requests implicating factors of the sort operative here.

B. De Novo Review in the Exemption 1 Context

When Congress amended FOIA in 1974, it did so in significant part to reverse the Supreme Court's decision in EPA v. Mink,49 which had foreclosed in camera inspection of classified documents.50 Representative Mink, the plaintiff in that case, explained the purpose of Congress:

Our intention in making this change is to place a judicial check on arbitrary actions by the Executive to withhold information that might be embarrassing, politically sensitive, or otherwise concealed for improper reasons rather than truly vital to national defense or foreign policy. We are not saying that any material must be released, only that it must be submitted to an impartial judge to determine whether its withholding meets the provisions and purposes of the act.51

Adoption of the in camera inspection provision of the Act sparked extensive debate on the standard appropriate for judicial review of Exemption 1 claims. The bill, as reported out by the Senate Judiciary Committee, stipulated that an exemption claim involving classified material was to be sustained unless it was determined that the decision to classify lacked a reasonable basis under criteria set forth by the governing executive order.52 Senator Muskie, however, introduced an amendment to remove this language, warning that the statutory "presumption"53 contained in the committee bill would "make the independent judicial evaluation meaningless ... go[ing] far to reduce the judicial role to that of a mere concurrence in Executive decisionmaking."54 Review of classification decisions, the Senator urged, ought to turn on the merits of the exemption claim rather than the force of a statutory presumption.55 And, emphasizing the importance of review de novo,56 the Senator had no doubt as to the ability of judges to make the necessary determinations:

By telling judges so specifically how to manage their inquiry into the propriety of a classification marking, we show a strange contempt for their ability to devise procedures on their own to help them reach a just decision. Moreover, by giving classified material a status unlike that of any other claimed Government secret, we foster the outworn myth that only those in possession of military and diplomatic confidences can have the expertise to decide with whom and when to share their knowledge. ... I object to the idea that anything but full de novo review will give us the assurance that classification ... has been brought under check.... I cannot understand why we should trust a Federal judge to be able to sort out valid from invalid claims of Executive privilege in the Watergate affair but not trust him or his colleagues to make the same unfettered judgments in matters allegedly connected to the conduct of defense or foreign policy.57

Opponents of the Muskie amendment raised the spectre of judicial shortcomings that the dissent resurrects today,58 but the arguments of the proponents prevailed. The Senate passed the amendment by a vote of 56 to 29.59 Thereafter, the Conference Committee adopted the in camera inspection provision of the Senate bill as amended, adding no more than a directive that "courts, in making de novo determinations in [Exemption 1] cases, ... will accord substantial weight to an agency's affidavit concerning the details of the classified status of the disputed record."60 The House then accepted, by the overwhelming margin of 349 to 2, the Conference bill authorizing in camera inspection on review of Exemption 1 claims.61

President Ford, however, contending that "the courts should not be forced to make ... [decisions] in sensitive and complex areas where they have no particular expertise,"62 vetoed the legislation and the debate resumed. Members of both Houses again voiced their confidence in the competence of the courts to review Exemption 1 claims de novo,63 and voted resoundingly to override the veto.64 As we have observed en banc, "this vote of confidence in the competence of the judiciary affirms our own belief that judges do, in fact, have the capabilities needed to consider and weigh data pertaining to the foreign affairs and national defense of this nation."65

In balancing the disclosure and the privacy interests staked out in this case, the District Court will have to engage in a searching de novo examination of the evidence to be introduced before it in order to gauge the likely consequences of revealing the citizenship information requested by the Post. We have already demonstrated that unquestioning deference to the State Department's affidavits would be unseemly in an Exemption 1 context, and the more so where, as here, no claim of national security or foreign policy concern has been advanced.66

Nor, in our view, is a holding in favor of the Department required by what the dissent characterizes as "common sense."67 Much of the content of the Department's affidavits is contradicted by concrete evidence identified by the Post.68 And whatever weight the opinion of the Department, as a presumed expert in the foreign relations field, is able to garner, deference cannot extend to blatant disregard of countervailing evidence.69 Indeed, it is ironic that the dissent should propose to resolve through "common sense" an issue it claims lies beyond the competence of the judiciary.

The central question in this litigation is one of predictive fact--the expectable consequences of a public release of particular information. Though Congress has conferred "substantial weight" upon agency affidavits only in Exemption 1 cases,70 the State Department's affidavits in this case are still entitled to careful consideration. But, when, as here, conclusions drawn in agency affidavits are hotly contested by other evidence, both the need for and the propriety of a judicial resolution emerge.

The State Department's affidavits predict that Iranians who have connections with the United States face possible harm. One of two affidavits upon which the Department relies heavily states that "[a]n official of the Government of Iran who is reputed to be an American citizen would, in my opinion, be in physical danger from some of the revolutionary groups that are prone to violence."71 Another affidavit avers that "there continues to be intense anti-American sentiment in Iran and several revolutionary leaders have been strongly criticized in the press for their alleged ties to the United States. Any person suspected of having such a connection would likely be subject to severe scrutiny, suspicion and possible harm."72 Yet, as we have noted, there already has been widespread public speculation concerning a relationship between Dr. Yazdi and the United States. There have been published accounts, drawing on sources both within and without the United States Government,73 that would appear more than sufficient to provoke the reprisals the Department dreads. But, as the District Court observed and neither party has disputed, "the danger anticipated by the affidavits submitted by [the Department] has not materialized. On the contrary, in months following that report, Dr. Yazdi's governmental responsibilities, according to [the Department's] affidavits, were increased."74 The publicity given Dr. Yazdi's alleged United States citizenship, the fact that he has not been the subject of retaliation, and the further fact that he seemingly enjoys heightened status in the Iranian Government, combine to pointedly controvert the Department's prediction of harm from disclosure of the documents the Post wishes to inspect.

It was not until the Department's motion for summary judgment was filed--four years after the Post's initial FOIA request--that the Department advanced the contention that the threat to Dr. Yazdi arises not so much from unverified information about a United States citizenship as from "official confirmation" thereof.75 Certainly under some circumstances official corroboration may have effects qualitatively different from those emanating from unofficial information already publicly known.76 But whether a particular confirmation likely will produce a more profound effect, or increase an effect beyond a legally acceptable limit, is clearly a question of fact, and as such an appropriate subject of inquiry at an evidentiary hearing.77

Moreover, outright deference to the State Department's risk appraisal would ignore the Supreme Court's admonition, in earlier remanding this very case, that "[t]he public nature of information may be a reason to conclude, under all the circumstances of a given case, that the release of such information would not constitute a 'clearly unwarranted invasion of personal privacy.' "78 Although this "impl[ies] the existence of a low-level privacy interest in ... records despite their public availability somewhere in the nation,"79 it also suggests that the interest fades commensurately with public availability of those records. We recently recognized that in some cases the dissemination of already-public information may result in "particular harm" to an individual and counsels against disclosure.80 The existence of "particular harm" in this case, however, depends precisely upon the resolution of an issue of predictive fact, which renders an evidentiary exploration indispensable.

For the reasons discussed, we remand this case to the District Court for further proceedings leading eventually to appropriate factfinding should any issue of material fact persist. This remand will serve several highly useful purposes. First, we note that the District Court denied the Post any opportunity for discovery.81 In FOIA cases, as in other litigation, discovery is an important tool for truth-testing,82 and each mode of discovery requested by the Post83 was appropriate in a FOIA action.84 The Post made known that its discovery efforts were directed toward ascertaining any factual premises for the Department's estimate of resulting harm to Dr. Yazdi, the basis for the Department's assertion that already-public information on Dr. Yazdi's citizenship would not dissipate any such harm, and the Department's role in prior disclosures by certain of its former employees and the Immigration and Naturalization Service.85 The discovery proposed by the Post thus would have borne directly on the all-important question of potential harm to Dr. Yazdi from release of the documents requested. With these discovery objectives, there is hardly any risk of accidental disclosure of putatively exempted material, a factor that has been crucial to limitations on discovery in other FOIA cases.86

After completion of discovery, a hearing might be necessary to evaluate conflicting evidence.87 The Post should be afforded the opportunity to cross-examine the Department's witnesses and, of course, the Department should enjoy the same option respecting witnesses selected by the Post.88 Similarly, both sides should be allowed to present the testimony of expert witnesses,89 and non-experts with first-hand knowledge on the state of affairs in Iran, particularly with respect to attitudes toward Iranians having political connections with the United States.90

We reverse the District Court's summary judgment in favor of the Post. We also reject the contention that predictive judgments by the State Department on the nature and impact of future behavior of foreign nations and their citizens are inviolate in Exemption 6 proceedings. We direct the District Court to accord the parties full de novo review, and we hold that unless the Department discharges its burden of sustaining its action,91 the mandatory disclosure directive of FOIA will require release of documents requested. We remand this case to the District Court for further proceedings consistent with this opinion.

So ordered.

The majority's remand for an evidentiary hearing to determine the "fact" of Dr. Yazdi's degree of peril will require a hearing on a question wholly unsuited to the capacities of the judicial process. The facts judges, and juries, find are ordinarily about past events or future harms from past events predicted according to disciplines such as medicine or economics or common sense views of likely outcomes. The "fact" the majority here requires the district court to "find" is not like a determination whether an automobile involved in a collision ran a red light, whether a defendant distributed a controlled substance, or whether business competitors agreed to fix prices. The opportunity for cross-examination, which the majority claims must be afforded appellee, is required for sorting out facts of that sort, not for making political predictions about what might happen to Dr. Yazdi under wholly unknowable circumstances in the future, which is rather like using a trial to predict the outcome of the next Presidential election.

If an evidentiary hearing is the best method of deciding questions like these, then the State Department and the Central Intelligence Agency ought to use the federal courts and cross-examination to arrive at the "truth" on such questions as what the Ayatollah Khomeini will do next with respect to Iraq or the prospects of dissident factions within Libya. It is improbable that anybody would suppose that the result, whatever it might be, was a finding of "fact." It would be speculation only. The question here is similar: what actions might be taken in the future with respect to a particular individual by the government of Iran, or by any one of a number of political or religious factions, some of them presently unknown, within that turbulent and volatile nation.1

The majority requires the Department of State and the Washington Post to have a hearing on the latter "issue," that they cross-examine each other's affiants, and that both sides "present the testimony of expert witnesses, and nonexperts with first-hand knowledge, on the state of affairs in Iran, particularly with respect to attitudes toward Iranians with American ties." Maj. op. at 39.2 It is difficult to know how the district court will try "the state of affairs in Iran" other than to ask whether there is any danger to an Iranian politician who has denied that he is an American citizen and then is shown to be one by the American government. If more than that is proposed, it is not clear what it is. Given the events of the past several years, moreover, it is hard to believe an evidentiary hearing is required to learn Iranian attitudes toward America.

Following the hearing, the majority presumably would have the district court make findings as to what the evidence adduced established. This court will then review, on a clearly erroneous standard, the "findings" about the "state of affairs in Iran," "attitudes toward those with American ties," and what all of this might portend for Dr. Yazdi's continued well being.

The majority seems to completely miss my point that courts do not deal in predictive political facts. I agree with the majority that courts can, and should, determine facts involving foreign events and actors. The problem with the "facts" in this case is not that they involve foreigners, acts taking place on foreign soil, or even events taking place in Iran. See, e.g., Islamic Republic of Iran Broadcasting v. Sotheby Parke-Bernet, Inc., 839 F.2d 780 (D.C.Cir., 1988) (court upholds a jury's factual determination involving the confiscation of a violin in Iran).

What places this case so totally outside the province of judicial capacity is that the district court is asked to predict political events and violence in Iran. While courts are competent to determine issues of predictive fact, I know of no legal principle which allows a court to speculate about future political events and then call its guess a "fact."

One can dress up any sort of political speculation to look like a judicial determination of "facts," but the dress does not alter the reality. Since the world's best intelligence services cannot infallibly predict the occasions and the targets of terrorism, it would seem unduly optimistic for the D.C. Circuit to suppose that its trial processes can do so. In truth, what will come out of such a process in the present case will be an uninformed guess by unqualified judges about what might happen to Dr. Yazdi in a nation torn by factional violence, in a culture wholly different from our own, in a political context we do not understand, in a city we have never seen, and that is over 6,000 miles from where we sit.

It is appalling that anyone's life, liberty, or property is to be made to depend on so preposterous a judicial undertaking. The majority fails to acknowledge, even in passing, that the privacy interest asserted here is quite different from the typical Exemption 6 privacy interest. We are not being asked to determine whether disclosure of Yazdi's citizenship status will unduly embarrass him or perhaps cut short a promising career in Iran by damaging his political reputation. Instead the privacy interest asserted is the possibility that physical danger to life and limb will flow from disclosure. Surely there can be no asserted privacy interest graver than the risk of physical harm or loss of life.

No court could conceivably find that Dr. Yazdi will be subject to no danger whatever or that he will certainly be injured. This means that the best a factfinder can honestly attempt is to estimate the odds on injury to Dr. Yazdi and then decide what degree of risk crosses the threshold of an unwarranted "invasion of privacy." That is a most peculiar judicial function. I would have thought any increase in danger to a man's life would cross the threshold. And, as I have suggested, no judge can honestly say that increased danger will not result. It should be enough to support the summary judgment here that neither we nor the trial court can possibly say with any assuredness, with or without a hearing, that releasing the information sought would pose no danger to Dr. Yazdi.

Although FOIA provides that reviewing courts must examine the record de novo, 5 U.S.C. Sec. 522(a)(4)(B) (1982), in the particular facts of this case the factors to be considered are so far beyond the competence of the judiciary to assess that according substantial weight to the factual allegations and opinions of the agencies is appropriate. See, e.g., CIA v. Sims, 471 U.S. 159, 176, 105 S. Ct. 1881, 1891, 85 L. Ed. 2d 173 (1985); Weissman v. CIA, 184 U.S.App.D.C. 117, 565 F.2d 692, 697 (1977). Congress specifically recognized this principle with respect to Exemption 1 of the FOIA:3

Executive departments responsible for national defense and foreign policy matters have unique insights into what adverse affects [sic] might occur as a result of public disclosure of a particular classified record. Accordingly, [Congress] expect[s] that [f]ederal courts, in making de novo determinations in section 552(b)(1) cases under the Freedom of Information law, will accord substantial weight to an agency's affidavit concerning the details of the classified status of the disputed record.

S.Rep. No. 1200, 93d Cong., 2d Sess. 12 (1974), U.S.Code Cong. & Admin.News 1974, p. 6290, see also 120 Cong.Rec. 17,024 (1974) (Sen. Muskie) ("[By suggesting de novo review in Exemption 1 cases,] I am not asking the courts to disregard the expertise of the Pentagon, the CIA, or the State Department. Rather, I am saying that I would assume and wish that the judges give such expert testimony considerable weight.").

In Halperin v. CIA, 203 U.S.App.D.C. 110, 629 F.2d 144 (1980), this court applied the "logic" of Congress' statement with respect to Exemption 1 in its review of a CIA decision not to release the names of attorneys or law firms retained by the CIA to perform legal services connected with classified activities. The CIA claimed that the records were exempt from disclosure under Exemption 3 of the FOIA, which protects matters that are "specifically exempted from disclosure by statute," 5 U.S.C. Sec. 552(b)(3) (1982). The statute the CIA relied upon was 50 U.S.C. Sec. 403(d)(3) (1976), which provides that "the Director of Central Intelligence shall be responsible for protecting intelligence sources and methods from unauthorized disclosure[.]" The CIA refused to disclose the names sought because, among other reasons, the disclosure could result in harm to the individuals identified and could lead to further disclosure of intelligence sources and methods. We held that the question of "whether the predicted danger is a reasonable expectation ... is precisely ... [the] point that a court, lacking expertise in the substantive matters at hand, must give substantial weight to agency statements, so long as they are plausible and not called into question by contrary evidence or evidence of agency bad faith." Halperin, 629 F.2d at 149.

The "logic" of according deference to agency opinions regarding foreign policy and this court's holding in Halperin should provide persuasive reason to accord "substantial weight" to the State Department's assessment of the potential harm to Dr. Yazdi from disclosure of the requested information.4 First, the reasons for deference in foreign policy matters are applicable here, for an estimate of the degree of danger in which disclosure would place Dr. Yazdi depends upon expert knowledge about the internal situation and politics of Iran, matters within the competence of the State Department and far outside ours.

Second, it is anomalous to accord deference to an agency's opinion "that the disclosure of the identity of an attorney doing work for the CIA might expose him to adverse action from hostile powers," Halperin, 629 F.2d at 148, but to refuse to accord any deference to an agency's opinion that "[a]n official of the Government of Iran who is reputed to be an American citizen would ... be in physical danger from some of the revolutionary groups that are prone to violence." Affidavit of Harold H. Saunders p 2, J.A. at 19. In both cases the expertise to which we defer is the same: the ability to assess the political climate in a foreign nation and to determine the attendant risk of harm to individuals from disclosure of information regarding those individuals. And, in each case judges "lack the expertise necessary to second-guess such agency opinions." Halperin, 629 F.2d at 148.

Application of this standard in assessing Dr. Yazdi's privacy interest reveals that the privacy interest at stake is indeed quite substantial. For example, in describing the situation in Iran in 1980, Harold H. Saunders, then Assistant Secretary of State for Near Eastern and South Asian Affairs, stated that "[a]ny individual in Iran who is suspected of being an American citizen or of having American connections is looked upon with mistrust. An official of the Government of Iran who is reputed to be an American citizen would, in my opinion, be in physical danger from some of the revolutionary groups that are prone to violence." Affidavit of Harold H. Saunders p 2, J.A. at 19. Yazdi was Iran's Foreign Minister but resigned soon after the takeover of the American Embassy in Tehran. He apparently remained a member of the Revolutionary Council for some time. Id. p 3, J.A. at 19. He also sat in the Iranian parliament, see Supplemental Affidavit of Harold H. Saunders p 2, J.A. at 80, though at oral argument counsel for the State Department stated his understanding that Yazdi was no longer a member of that body.

In a supplemental affidavit, Assistant Secretary Saunders noted that "[t]here continues to be intense anti-American sentiment in Iran and several revolutionary leaders have been strongly criticized in the press for their alleged ties to the United States [and] any person suspected of having such a connection would likely be subject to severe scrutiny, suspicion and possible harm." Supplemental Affidavit of Harold H. Saunders p 3, J.A. at 80. See also Affidavit of David T. Schneider, Senior Deputy Assistant Secretary of State in the Bureau of Near Eastern and South Asian Affairs: "Individuals ... arrested due to their association with Americans remain in jail"; "There has been no moderation in the anti-American rhetoric of the regime. The U.S. is still called the 'Great Satan' "; "Cumulated evidence points to Iranian involvement in acts of international terrorism, including the attacks on the U.S. Embassy in Beirut and the U.S. Marine barracks in Lebanon"; Dr. Yazdi is "a member of the faction led by Mehdi Bazargan.... This faction is already in danger due to Mr. Bazargan's public criticisms of the regime.... A meeting attended by members of the Bazargan faction was broken up by rioters who entered the building, forcing Bazargan and others to flee through a rear entrance. Dr. Yazdi's association with this faction is well known in Iran as are his relatively moderate politics including a willingness to seek accommodation with the United States." J.A. at 126-32.

These facts justify the affidavit's conclusion that "[i]f Department of State files were to confirm that Dr. Yazdi was a U.S. citizen, which he has reportedly denied, it would have results which constitute a serious threat to his personal privacy, safety, and well being, including arrest and imprisonment." J.A. at 131. It seems clear that Yazdi, who was a prominent Iranian official and who has denied being an American citizen, would be in more than ordinary peril if the American government should declare that, in truth, he is an American citizen. This court is certainly in no position to contradict the State Department's assertions on that subject.

I do not, as the majority asserts, advocate blind deference to the State Department. There is no anomaly in asserting that we should accord substantial weight to the State Department's affidavit while also noting that the conclusions stated in it are plausible. Under the standard enunciated in Halperin, the State Department's affidavits should be accorded substantial weight because "they contain reasonable specificity of detail rather than merely conclusory statements, and ... they are not called into question by contradictory evidence in the record or by evidence of agency bad faith." Halperin, 629 F.2d at 148.5 My arguments from plausibility are an appropriate examination for contradictory evidence or bad faith.

The only evidence in the record that could be construed to contradict the Department's affidavits is the previous publication of allegations of Dr. Yazdi's U.S. citizenship and lack of any known subsequent harm to Dr. Yazdi. But, as the State Department contends, it is the official confirmation of Dr. Yazdi's citizenship that would result in a serious threat to his safety and well being. It is common knowledge that foreign governments often react much more strongly when facts they already know or suspect are given official confirmation. An authoritative statement from the United States government about a person's American citizenship is certain to be regarded as more credible than unofficial assertions published in the media. Whatever suspicions may have been occasioned by the two books previously mentioned, official confirmation would remove all doubt. And even if no doubt exists among those in Iran who might take action, a public statement by our government may still provoke a reprisal that otherwise would not take place, for an official statement of this sort, unlike media accounts, may be seen by members of a foreign government as requiring a public response to something they would have preferred to ignore.

This circuit has previously noted the distinction between speculation and official acknowledgment. Thus, we have remarked that the Soviet Union had long known of flights of U-2 planes over Soviet territory and knew specifically that a U-2 they had brought down was engaged in gathering intelligence. But it was only when President Eisenhower publicly confirmed those facts and that he had approved the mission that the Soviet Union denounced the American action and cancelled a scheduled summit conference. See Afshar v. Department of State, 226 U.S.App.D.C. 388, 702 F.2d 1125, 1130-33 & n. 7 (1983) (citing Phillippi v. CIA, 211 U.S.App.D.C. 95, 655 F.2d 1325, 1332 (1981)); Military Audit Project v. Casey, 211 U.S.App.D.C. 135, 656 F.2d 724, 743-45 (1981).

In the context of FOIA, this circuit has noted that "[u]nofficial leaks and public surmise can often be ignored by foreign governments ... but official acknowledgment may force a government to retaliate." Afshar, 702 F.2d at 1130-31. Although that statement was made about a foreign government's cooperation with U.S. intelligence services, its underlying rationale is applicable here. The distinction between a foreign government's reaction to rumor or speculation and official confirmation is not dependent on the type of information released or the FOIA exemption involved. Instead, reaction to official confirmation takes place "[i]n the world of international diplomacy, where face-saving may often be as important as substance...." Phillippi, 655 F.2d at 1333. I believe that under this persuasive precedent, the Washington Post's reliance on the impact of widespread speculation about Yazdi's citizenship on his privacy interest is misplaced.

This court should instead give substantial weight to the State Department's allegations that official confirmation of Yazdi's alleged citizenship would subject him to "physical danger" and "severe scrutiny, suspicion and possible harm." It is clear that the Department asserts "threats to privacy interests more palpable than mere possibilities." Department of the Air Force v. Rose, 425 U.S. 352, 380 n. 19, 96 S. Ct. 1592, 1608 n. 19, 48 L. Ed. 2d 11 (1976).

In the case before us, we have not merely the State Department's informed estimate of the dangers, backed by a factual recounting of events in Iran that is not disputed, but also an estimate that, as any reader of the Washington Post knows, is very plausible. Very little deference to the State Department's knowledge is required. This court is not asked, on the basis of undisclosed expertise, to believe anything more than common sense alone would suggest. Given the nature of the threat in this case, and the utter inadequacy of the trial process to find the degree of Yazdi's danger, let alone to find that the danger is nonexistent, such an expert assessment, backed by recited facts, should be sufficient. I would grant summary judgment to the Department of State.