with whom BOWNES, Circuit Judge, joins, concurring in part and dissenting in part.
The majority is quite right in finding a waiver of the government’s right to invoke Exemption 7(D) for information that “has been actually disclosed in the source’s prior public testimony.” Supra at 1457. But, having found a waiver, the court is overly timeous in sketching its dimensions. We maintain that the scope of the waiver engendered by prior public testimony should be coextensive with that of hypothetical cross-examination. See infra Part IV. In our estimation, the court’s reluctance to accept such a rule stems from a gross miscalculation of its potential effect on the government’s ability to recruit confidential sources. That miscalculation pervades the majority’s ensuing analysis, improperly confining the disclosure mandated by the FOIA. Because Congress and the Court instruct us to read the Act less grudgingly, and to apply waiver with somewhat greater breadth, we respectfully dissent.
I
Our disagreement with the majority begins with its phrasing of the question on appeal. Rather than considering whether “ ‘confidential sources,’ by testifying at public trials concerning some of their communications to the FBI, ‘waived’ the FBI’s right to invoke exemption 7(D),” ante at 1446-1447, we should begin by deciding where the FBI fits into the exemptive scheme. The assumption that Exemption 7(D) is ultimately for the FBI to waive because courts “must consider the interests of the law enforcement agency as well as those of the source; and this fact in turn suggests that ‘waiver’ by ‘sources’ will not automatically prove sufficient to release the information,” ante at 19; see also ante at 24 (Exemption 7(D) “mainly seeks to protect law enforcement agencies”), reflects a sort of tunnel vision.
To be sure, courts must take into account the government’s stake in confidentiality— but we fail to find in the majority opinion any analysis of how a limited waiver would effect the underlying agency interest: the ability “to recruit, and to maintain, confidential sources ... to protect the flow of information to the law enforcement agency.” Supra at 1449. The majority errs— profoundly, we believe — in failing to analyze waiver from either the agency’s or the source’s point of view. Its reasoning seems instead to be premised on the remarkable assertion that: “One need only, and more reasonably, believe that, as a general matter, the more confidential information that appears in public (contrary to the true desire of the source), the harder it becomes for the law enforcement agency to recruit confidential sources.” Ante at 1454. This assertion, by its terms, puts the cart well in front of the horse.
With respect, we think proper analytical sequence starts not with the breadth of the exemption, but with the compass and conception of the statute to which it pertains. A policy of forthright and expansive disclosure undergirds the FOIA. See, e.g., Irons v. F.B.I., 811 F.2d 681, 685 (1st Cir.1987) (Irons I). The wide sweep of mandated disclosure is commensurate with the law’s *1458principal purpose: to “ensure an informed citizenry, vital to the function of a democratic society, needed to check against corruption and to hold the governors accountable to the governed.” NLRB v. Robbins Tire and Rubber Co., 437 U.S. 214, 242, 98 S.Ct. 2311, 2327, 57 L.Ed.2d 159 (1978); see also US. Dep’t of Justice v. Reporters Comm. for Freedom of the Press, — U.S. -, 109 S.Ct. 1468, 1480-82, 103 L.Ed.2d 774 (1989). Of course, exceptions were written into the statute, but there is a telling legal history behind them.
The FOIA was originally enacted because the public disclosure section of the Administrative Procedure Act (then 5 U.S.C. § 1002) “was generally recognized as falling far short of its disclosure goals and came to be looked upon more as a withholding statute than a disclosure statute.” EPA v. Mink, 410 U.S. 73, 79, 93 S.Ct. 827, 832, 35 L.Ed.2d 119 (1973). This renitency was thought due in part to a “vague” exemptions clause. Id. To alleviate uncertainty nine niches were carved, with the clear understanding that “these limited exemptions do not obscure the basic policy that disclosure, not secrecy, is the dominant objective of the Act.” Dep’t of the Air Force v. Rose, 425 U.S. 352, 361, 96 S.Ct. 1592, 1599, 48 L.Ed.2d 11 (1976). The exemptions were designed to prevent “unlimited withholding of files merely by classifying them as investigatory files compiled for law enforcement purposes.” FBI v. Abramson, 456 U.S. 615, 621, 102 S.Ct. 2054, 2059, 72 L.Ed.2d 376 (1982).
In consequence of the statute’s etiology and manifest purpose, the Supreme Court, and this court, have regularly held that exemptions are to be construed narrowly and in keeping with FOIA’s overriding goal: that the light of public inquiry be facilitated, not hooded. See, e.g., U.S. Dep’t of Justice v. Julian, 486 U.S. 1, 108 S.Ct. 1606, 1611, 100 L.Ed.2d 1 (1988); Rose, 425 U.S. at 360, 366, 96 S.Ct. at 1598, 1601; Aronson v. Dep’t of Housing and Urban Development, 822 F.2d 182, 185 (1st Cir.1987). To this end, we have consistently placed on the withholder the burden of establishing an exemption’s applicability. See, e.g., Irons I, 811 F.2d at 685. Under these interpretive constraints, we find it difficult to accept the agency’s unanalyzed assertion that a limited waiver would injure the FBI’s recruitment capability. Acceptance of an unproved assertion is, in substantive consequence, little different from acceptance of an unqualified classification.1 Without more, we should not be parties to the attempt to end run Congress’s evident will.
The majority quotes passages from congressional debates proving, it says, that Exemption 7(D), contrary to the general dictates of the Court, should “apply broadly and therefore quite literally.” Ante at 1450. The effort falls short: the scraps of legislative history cited, supra at 1450-51, do not suffice to remit our burden of analyzing the impact of waiver on legitimate agency interest. The opposite is true. Every reference in the legislative history to Exemption 7 generally, and to Exemption 7(D) in particular, says that the exemptions are to be construed stingily. See, e.g., Senate Comm. on the Judiciary & House Comm. on Government Operations, 94th Cong., 1st Sess., Freedom of Information Act and Amendments of 1974 (P.L. 93-502) Source Book: Legislative History, Texts, and Other Documents 36, 65 (Joint *1459Committee Print, March 1975) (Source Book) (quoting Report of House Committee on Government Operations); id. at 117; id. at 158 (quoting Senate Report); id. at 230 (quoting Joint Explanatory Statement of the Committee of Conference); id. at 265, 286, 293, 302, 333, 336, 349, 376, 378, 385, 413, 421 (remarks of various Senators and Representatives). It is, therefore, unsurprising that the philosophy of full disclosure is carefully embodied in the considered reports issued by the committees responsible for formulating the FOIA. See Mink, 410 U.S. at 80 & n. 3, 93 S.Ct. at 832 & n. 3.2 While paying obeisance to the need to keep the FBI’s investigatory files confidential, the meaningful portions of the legislative history do so only insofar as “[s]uocess lies in providing a workable formula which encompasses, balances, protects all interests, yet places emphasis on the fullest possible disclosure.” Id. (quoting Senate Report). Unless Congress intended that reviewing tribunals analyze assertions of agency interest, a balanced formula lies beyond our ken.
To make our point more concretely, we cite a few examples. In the paragraphs following the portions of Senator Hart’s speech, quoted supra at 1450, the senator discussed the various techniques agencies had used to “thwart access to publicly valuable information in their files.” Source Book at 451. He noted that “these amendments were necessary because the agencies have not made a good faith effort to comply with the act.” Id. at 451-52. He cited “historical records” as the archetype of materials questionably withheld.3 Id. at 452. Finally, Senator Hart seemed to assume that the protection of information was necessary not for its own sake, but to safeguard a source’s identity and reduce administrative makework. Id.; see also id. at 462-63 (Sen. Ervin, introducing letter of Sens. Mathias, Case, Javits, Baker, Kennedy, Muskie, Hart, and Ervin, and discussing adequate protection of confidentiality and administrative burden); id. at 456-57 (remarks of Sen. Hruska). Put another way, an agency’s right to claim blanket protection applied only to ameliorate the administrative “overburden[]” of redacting documents. Id. at 452 (remarks of Sen. Hart). As Senator Hart concluded: “The fact that the agency can withhold information furnished by a confidential source relieves it of the burden of showing that disclosure would actually reveal the identity of the confidential source....” Id.; see also id. at 450. Only if one understands Senator Hart as addressing the combined problem of confidentiality and administrative burden can one reconcile the remarks quoted by the majority and by us. With identity no longer an issue and with at least some information already available— the source, after all, has testified in public — Senator Hart’s statements, taken fully and in context, are hardly supportive of a “no waiver” rule.
While more cryptic, the observations of Senator Kennedy, supra at 1450, should be read in the same vein. They followed closely on the heels of Senator Hart’s remarks and appeared to adopt the sponsor’s understanding. See, e.g., Source Book at 459. President Ford’s veto message also discussed the potential compromise to “confidentiality” in terms of the administrative burden imposed (an agency would have to deal with too many documents too quickly, and information which could compromise a source’s identity might slip out). Id. at 484; see also id. at 461, 476 (Sen. Scott, characterizing presidential objections). Nowhere did the President mention the value of confidential information per se. And the comment of Senator Byrd, supra at 1450-51, though unambiguous, lacks any explanation as to why release of information, *1460once a source’s identity had already been revealed, would “dry[ ]-up ... sources of information” to any appreciable extent. Source Book at 468. That unexplained assertion should not dwarf the more authoritative guides to legislative intent contained in the committee reports and elsewhere in the congressional annals.
Under these circumstances, there is no sufficient basis for applying the exemption’s language expansively and pervasively. Rather, the Congress’s clearly-articulated preference for generalized disclosure, fully adopted by the Court, should govern our interpretation. While we are constrained “to uphold a claimed 7(D) exemption so long as the statutory criteria are met,” Irons I, 811 F.2d at 685, we ought to be more circumspect in looking for support in the language of those who, presumably, would most adamantly be opposed to withholding the documents at issue. It is the Act we must construe broadly, not the exemption (at least in the absence of a compelling rationale to do otherwise). We, therefore, read the exemption narrowly. This done, we are led inexorably to the conclusion that public testimony lifts the veil of secrecy to a greater extent than our colleagues are willing to acknowledge.
II
The majority suggests that Exemption 7(D) is the FBI’s to invoke, but fails to explain how we may understand when an agency waives the right of invocation. It makes abundant sense, from the perspective of both, or either, the agency or the source, to draw the margins of waiver consonant with the hypothetical scope of cross-examination, as we recommended in the panel opinion (now withdrawn) and continue to recommend here. See infra Part IV.
The assertion that “as a general matter, the more confidential information that appears in public (contrary to the true desire of the source), the harder it becomes for the law enforcement agency to recruit confidential sources,” supra at 1454, lies at the core of the court’s reasoning. The thesis deserves consideration. But, one must bear in mind that the proper temporal perspective is that of the agency looking for a future source and/or a source deciding whether to furnish confidential information to the agency in the future. The agency, after all, must be concerned with its ability to gain (or maintain) sources in the days to come; it possesses no time machine permitting it to go back into the past to get more sources. By the same token, while a tipster may fear betrayal of information already divulged, the information is no longer the source’s to cradle. If fear can be thought to dry up sources, it can only operate as a desiccant in futuro.
All of us seem to agree that a primary concern is the preservation of anonymity. That, at least, is what the majority tells us is the government’s view of things: “a rule of law that permits reputable historians to conduct historical research will permit, to the same extent, the most disreputable criminals to search out the identity and knowledge of those who inform against them.” Supra at 1447.4 But, by definition, public testimony will result in loss of anonymity. Put colloquially, the thugs will know that the source has ratted on them. That, for our hypothetical “most disreputable criminal,” is probably enough. He can then obtain the source’s identity and the content of publicly-revealed information (matters which the majority concedes cannot be kept in confidence when public testimony has occurred). Once that has happened, it strains credulity to think that the marginal increment of potential additional information one might gain under our suggested approach, see infra Part IV, information available by definition to the cross-examiner in the first place, could affect the source’s safety or even his peace of mind. A source does not care about the aggregate of the information available in *1461the world. From his viewpoint, real betrayal of confidentiality lies in the fact of testimony itself. To use Judge Breyer’s phrase, supra at 1454, “fine spun risk-benefit calculations” are not in order.
Furthermore, Exemption 7(D) does not exist in a vacuum; it is but one plate in an array of armor. Even if 7(D) is construed more narrowly, so that some incremental information filters through, the informer (and in some instances the government) has other available bucklers. If information sought “interfere[s] with enforcement proceedings,” it is shielded. 5 U.S.C. § 552(b)(7)(A). An unwarranted invasion of the privacy of the witness or others whose identity and knowledge might be sought is protected under 5 U.S.C. § 552(b)(7)(C).5 See, e.g., McCorstin v. U.S. Dep’t of Labor, 630 F.2d 242, 245 (5th Cir.1980) (per curiam), cert. denied, 450 U.S. 999, 101 S.Ct. 1705, 68 L.Ed.2d 201 (1981). There are similar safeguards pertaining to the “life or physical safety of any individual,” 5 U.S.C. § 552(b)(7)(F), and the “disclosure of] investigative techniques and procedures.” 5 U.S.C. § 552(b)(7)(E). These, and other, protections crafted by the Congress, collectively comprise a phalanx adequate to shelter the legitimate interests of the vulnerable without hyperextending the range and reach of Exemption 7(D).
For these reasons, we suggest that from the source’s perspective the majority’s limitation on disclosure adds little or nothing. We think the same is likely true from the agency’s perspective. Granted, the majority’s disposition of this point has the advantage of simplicity: it assumes that the greater the aggregate of disclosed information, the fewer the sources. Supra at 1454. Even putting aside our belief that such an analysis runs headfirst into the motivating spirit of the FOIA, we remain unconvinced. Agencies do not recruit sources in the aggregate, but one by one. The ability to recruit a source depends on the ability to convince him to come forward. If the potential informant realizes that he might be required to testify, that fact, not the FOIA, will be the significant deterrent to his willingness to speak. As the majority itself acknowledges, the potential source must make a crude, not a fine spun, risk-benefit calculation.
There is one other integer in the equation. Our brethren admit that the act of testifying is presumptively “contrary” to a source’s “true desire.” Supra at 1454. But, think for a moment as to what is left unasked and unsaid. At whose behest does a source testify? Generally, it is the government which triggers the testimony. The recipient agency delivers the confidential information to, say, a prosecutor, precisely because it is agreeable that the identity of the source and the brunt of the information be revealed in a public forum to serve some governmental end. In that sense, the agency itself has decided that disclosure is worth the candle. At the same time, the agency is fully conscious of the scope of the information which might be revealed and can influence it to some degree. (Certainly, the government controls the scope of direct examination, and can thereby shape the scope of cross-examination somewhat.) It is fair, logical, and consistent with the method of the statute that the government’s desire to enjoy the benefits of public testimony should define and delimit its right to invoke Exemption 7(D).
Ill
Notwithstanding, the majority gives a number of reasons why waiver analysis is inapplicable. Whereas we agree that no express waiver has transpired in this situation, the claim that waiver may not legitimately be inferred from the source’s conduct, supra at 1452-1453 is more problematic. Most notable of all, the majority is silent about whether a court can infer waiver from the government’s conduct. We believe not only that a court can, but that it should. As Judge Breyer concedes, jurists *1462“use the word ‘waiver’ as a way of insisting that an individual forego protection which he says that he does not wish to give up but which he has agreed to give up in the past.” Supra at 1453 (emphasis in original). Substitute the party who, in the majority’s view, may legitimately invoke the prophylaxis of Exemption 7(D), i.e., the government, and the analysis is precisely on point.
For much the same reason, consigning the “rule of completeness” to the scrapheap of irrelevancy, supra at 1453-1454, will not wash. While there may or may not be any “reason grounded in fairness for requiring a source who disclosed information during testimony to reveal, against his will (or to have the FBI reveal for him), information that he did not disclose in public,” supra at 1454, there is a powerful reason grounded in fairness to require that a federal agency disgorge the information it was presumably willing to reveal in court (when it elected to have its source testify). Unless we have stumbled into the land of Oz, the government’s power to invoke Exemption 7(D) cannot amount to holding its sovereign breath during an informer’s cross-examination, hoping that little damaging testimony emerges; and, if hope be rewarded, then breathing a sigh of relief and scurrying back behind the shield of confidentiality. Such a cat-and-mouse view of the exemption makes mock of Congress’s desire to open governmental activity to the scrutiny of the “informed citizenry” described by the Robbins Court, 437 U.S. at 242, 98 S.Ct. at 2327. And, insofar as the related information, once opened to public purview, might serve the ends of justice and openness, a chance to obtain the whole story is the exact aim of the FOIA.
IV
We believe that proper construction of the Act, including Exemption 7(D), demands that we adhere to the same concept of waiver which infused our first opinion. See Irons I, 811 F.2d at 686 (waiver is “the purposeful relinquishment of an appreciated right” which “can fairly be deduced from conduct”). The act of testifying is always accompanied by at least the possibility of disclosing some quantum of information beyond the precise confines of the witness’s direct answers. Cross-examiners, for instance, have reasonable latitude to inquire into related areas. See, e.g., Fed.R.Evid. 611(b). Informants who take the stand must realize that, by so doing, they expose themselves to a range of questions which may go well beyond what the witness, or the examiner on direct, chooses to present during the case in chief. Follow-on questions may fairly delve into a queue of topics not originally discussed in detail, but made relevant to the inquiry by the juxtaposition of the witness’s primary testimony in the context of the hearing as a whole. See, e.g., United States v. Fortes, 619 F.2d 108, 121 (1st Cir.1980) (having raised subjects on direct, witness “opened the door to a full and not just a selective discussion of these matters” on cross; opposing party entitled “to have [witness] complete the picture she had begun on direct”). And, just as the informer is on notice of his widened exposure, the government — practiced in such matters, and presumably astute — is equally on notice.
Implying a waiver of this genre and limited scope is in our view consistent not only with the goals of FOIA, but with familiar principles embedded in our jurisprudence. In a variety of settings, the act of testifying may constitute a waiver of privilege as to information which a person would otherwise be able to keep in the bosom of the lodge. Perhaps the most commonplace example is that testimony waives a witness’s fifth amendment privilege against self-in-culpation and invites cross-examination regarding topics “made relevant” by direct testimony. United States v. Black, 767 F.2d 1334, 1341 (9th Cir.), cert. denied, 474 U.S. 1022, 106 S.Ct. 574, 88 L.Ed.2d 557 (1985). The scope of the consequent waiver is deemed to be “coextensive with the scope of relevant cross-examination.” Id. See also Brown v. United States, 356 U.S. 148, 154-56, 78 S.Ct. 622, 626-27, 2 L.Ed.2d 589 (1958) (similar). The rendition of testimony which trenches upon the attorney-client privilege is viewed in a similar fashion. In this context, as the District of *1463Columbia Circuit has declared, “[w]hen a party reveals part of a privileged communication ..., it waives the privilege as to all communications relating to the same subject matter.” In re Sealed Case, 676 F.2d 793, 817-18 (D.C.Cir.1982); see also United States v. Jones, 696 F.2d 1069, 1072 (4th Cir.1982); von Bulow v. von Bulow, 114 F.R.D. 71, 79 (S.D.N.Y.1987). There is, of course, a basic notion of equity which un-dergirds such rules: “a privileged party cannot fairly be permitted to disclose as much as he pleases and then to withhold the remainder_” Greater Newburyport Clamshell Alliance v. Public Service Co., 838 F.2d 13, 19 (1st Cir.1988).
Nor do we perceive any compelling reason for allowing a government informer who has actually testified to limit the scope of his revelations to the precise questions asked and answered, or to permit the FBI, under the aegis of Exemption 7(D), to withhold information directly relevant to the informant’s trial testimony. Congress, we think, did not intend to cast so huge and enduring a veil of impenetrability over confidences contained in law enforcement files. That is particularly true where, as here, disclosure will further the FOIA’s cardinal objectives. Scholarly research tends to take place at the margin of knowledge. The hypothetical scope of cross-examination may well reveal information of value to those citizens superintending the government, where that same marginal information, obtained at some risk, would mean little if anything to a revenge-seeker. That balanced calculation, embodied in a rule, is exactly what the Congress desired. See, e.g., Mink, 410 U.S. at 80, 93 S.Ct. at 832.
The caselaw, while admittedly tenebrous, fails to support either the majority’s reasoning or its result. Those cases in which “courts said that even though the relevant information already is public, the FBI nonetheless need not disclose it because the source has not ‘waived’ the exemption,” supra at 1454 (emphasis in original), do not adopt the majority’s outlook. If, as our brothers contend, we must look to the government, not the source, for waiver, then these cases suggest either that the majority’s perspective is incorrect or, at least, irrelevant in terms of the question which confronts us.6
Scherer v. Kelley, 584 F.2d 170, 176 n. 7 (7th Cir.1978), cert. denied, 440 U.S. 964, 99 S.Ct. 1511, 59 L.Ed.2d 778 (1979), quoted supra at 1454-55, is clearly aimed at a much wider target than is presented here. In Scherer, the Seventh Circuit rejected the view that “all information offered” regarding “every transcription made by an investigative agent” should be disclosed. Id. We agree. But, the Scherer panel nowhere addressed possible intermediate solutions. In that sense, Scherer and the “no-waiver” cases like it, e.g., Kiraly v. FBI, 728 F.2d 273, 279-80 (6th Cir.1984), are the flip side of the line of “full waiver” cases holding that an informant’s public testimony renders the totality of the information supplied by that informant disclosa-ble. See, e.g., Van Bourg v. NLRB, 751 F.2d 982, 986 (9th Cir.1985) (persons submitting affidavits “have no reasonable expectation of confidentiality and should expect their names and testimony to be revealed if the investigation results in a formal hearing”); United Technology Corp. v. NLRB, 111 F.2d 90, 94 (2d Cir.1985); Poss v. NLRB, 565 F.2d 654, 658 (10th Cir.1977). We continue to believe that there is a middle road between these two extremes — and that ample reason exists to take it.
Accordingly, we conclude that the Smith Act informants, by testifying, waived the privilege of confidentiality under Exemption 7(D) with respect to all information relevant to the testimony which they gave. As with the fifth amendment, the contours of the consequent waiver should be drawn coterminous with the hypothetical scope of relevant cross-examination. See Black, 767 F.2d at 1341. We believe that reading *1464the exemption in this commonsense fashion treats fairly with the government, the informants, and the requestors. Moreover, such a recension is comfortably compatible with the teleology of the FOIA.
We are mindful that our brethren suggest, in a hypothetical example and by argument, that our proposed rule would be unworkable. Their tone likewise implies the practical superiority of the rule which they espouse. But, neither theory nor experience lend credence to their claims. As to the hypothetical example, supra at 1451, it is perfectly clear that Congress never intended FOIA’s exemptions to function in so crabbed a manner. Rather, Congress intended other exemptions — for example, 7(C) (privacy) and 7(F) (endangerment) — to cover the fictitious problem contrived by our colleagues. See, e.g., Source Book at 184, 293-94, 298-99, 313, 333, 338, 469, 476, 519, 523. And as to practicality, it was Senator Hart who noted that these were “determination[s] courts make all the time_” Id. at 351. In this instance, experience has proven out the legislative forecast. See, e.g., White v. IRS, 707 F.2d 897, 901-02 (6th Cir.1983) (Exemption 7(C) shields identity of those willing to provide statements); Alirez v. NLRB, 676 F.2d 423, 426-28 (10th Cir.1982) (Exemption 7(C) protects co-worker statements); Bast v. U.S. Dep’t of Justice, 665 F.2d 1251, 1254-56 (D.C.Cir.1981) (extending Exemption 7(C) to protect third parties); Fund for Constitutional Government v. National Archives and Records Service, 656 F.2d 856, 863-66 (D.C.Cir.1981) (utilizing Exemption 7(C) for benefit of unindicted Watergate figures).
Then, too, there is no reason to think that the circumscribed disclosure which we would permit has the potential of interfering with the government’s (admittedly significant) interest in safeguarding informant programs. In Irons I, we made manifest our belief that Exemption 7(D) should not be construed in such a way as would “handcuff[ ] ... [the government’s] performance of important law enforcement functions” or so as to force the government “to choose between establishing a network of confidential sources or formulating a successful prosecutorial strategy.” 811 F.2d at 689. We remain faithful to those sentiments. By procuring informant testimony, however, the government makes a strategic decision: to chance forfeiting persons’ anonymity and much of the confidential information possessed by them in order to gain the benefit of their evidence. By actually testifying, the informants themselves have made a similar choice. In stepping from the shadows into the limelight, they surrender their anonymity and accept the possibility of divulgement of all relevant information they supplied to the government.
The scope of disclosure which we believe the FOIA requires neither places an untoward burden upon law enforcement — indeed, it places no greater burden than does the scheme endorsed by the majority — nor threatens the “needless drying up of sources [which would be] at odds with the purpose of the exemption.” Shaw v. FBI, 749 F.2d 58, 62 (D.C.Cir.1984). In criminal matters, after all, the government has lived with a substantially similar protocol, the Jencks Act, which provides in pertinent part:
After a witness called by the United States has testified on direct examination, the court shall, on motion of the defendant, order the United States to produce any statement... of the witness in the possession of the United States which relates to the subject matter as to which the witness has testified.
18 U.S.C. § 3500(b) (emphasis supplied).7 We have held that, in determining whether material is disclosable under the Jencks Act, “the key question is whether the information contained in the statements ‘relate generally to the events and activities testified to’ by the witness.” United States v. Ferreira, 625 F.2d 1030, 1034 (1st Cir.1980) (citation omitted). An equivalent inquiry should be made when deciding whether information has lost the protection of Exemption 7(D) following an informant’s actual *1465testimony. The relative ease with which courts have been able to apply the Jencks rule, and rules like it, belies the majority’s stilted view of cross-examination, supra at 1455-56, and its thumb-on-the-scale weighing of situational practicalities.
Lastly, we do not expect that the limited disclosure we envision would unduly dampen witnesses’ ardor for testifying. A witness would be at no greater risk of disgorging information than he was when he offered his public testimony. Thus, the “delicate balance achieved by the [FOIA] between the interest of a private citizen and that of an investigative agency,” Scherer, 584 F.2d at 176 n. 7, would be undisturbed.
V
We are satisfied that the giving of actual testimony by a confidential informant constitutes a waiver as to the subject matter of his or her testimony, and all other information related thereto, measured by the hypothetical scope of relevant cross-examination. The salutary purposes which the FOIA was devised to serve demand no less. To the extent that our brethren take a more parochial view of what Congress purposed, we respectfully dissent. Although we agree that the district court’s turnover order was significantly overbroad and must be vacated, we would direct the entry, in its place, of an order consistent with this opinion.
. Clearly, Congress sought to limit an agency’s ability to sidestep disclosure requirements. Under the aegis of Senator Hart, blanket protection for investigatory files was forgone; as the Attorney General wrote, "[t]he primary purpose of Senator Hart's amendment to revise exemption 7 was to ... require consideration of the particular document and the need to withhold it.” Senate Comm, on the Judiciary & House Comm, on Government Operations, 94th Cong., 1st Sess., Freedom of Information Act and Amendments of 1974 (P.L. 93-502) Source Book; Legislative History, Texts, and Other Documents 515 (Joint Committee Print, March 1975) (Source Book) (quoting Attorney General’s Memorandum on the 1974 Amendments to FOIA at 5). Thus, according to the Attorney General’s contemporaneous understanding, the FBI should be required to make a plausible showing that a particular category of information falls within the statute’s desired ambit of protection. See also Source Book at 158 (Congress anticipated that "intent of the exemption” would govern disclosure) (Report of Senate Committee on the Judiciary).
. The majority’s suggestion that, in enacting the 1974 amendments Congress retreated from the views it had expressed in the 1965 Senate Report, quoted in Mink, see, e.g., supra at 1450, is, with respect, plainly wrong. In point of fact, the Senate Judiciary Committee explicitly reaffirmed those views in 1974, quoting the exact language earlier quoted by the Mink Court. See Source Book at 159; see also id. at 349 (remarks of Sen. Kennedy); id. (Sen. Hart); id. at 264 (Rep. Fascell); id. at 413 (Rep. Reid).
. Senator Hart used as an example FBI files on Bertoldt Brecht. Source Book at 452. The parallel to the materials at issue here is so striking as to be almost eerie.
. At this juncture two things puzzle us. First, the FBI apparently wishes to take its stand on documents which do not raise the fearsome spectre of revenge. Second, there is something incongruous in the notion that the hypothetical "most disreputable criminal” would proceed in search of revenge by filing an FOIA request, thereby leaving a paper trail through which his own identity could be traced and exposed.
. We find it close to impossible to believe that a court would deem a quintessential invasion of privacy such as the threat of revenge able to penetrate the Exemption 7(C) plate. See, e.g., Maroscia v. Levi, 569 F.2d 1000, 1002 (7th Cir.1977) (per curiam).
. The majority says that Ingle v. Dep’t of Justice, 698 F.2d 259 (6th Cir.1983), teaches “that the FBI cannot 'waive' the protection the exemption gives the source.” Supra at 1454. On that reading, the majority’s perspective must be incorrect, for Ingle places waiver strictly in the source’s hands. 698 F.2d at 269 ("it is properly the right of those sources themselves to waive any protection afforded by cooperating with the FBI”).
. The term “statement” as used in this context means "a written statement made by said witness and signed or otherwise adopted or approved by him.” 18 U.S.C. § 3500(e)(1).