UNITED STATES DEPARTMENT OF JUSTICE, ET AL., PETITIONERS V. REPORTERS COMMITTEE FOR FREEDOM OF THE PRESS, ET AL. No. 87-1379 In the Supreme Court of the United States 4 October Term, 1988 On Writ of Certiorari to the United States Court of Appeals for the District of Columbia Circuit Reply Brief for the Petitioners The Freedom of Information Act (FOIA), 5 U.S.C. (& Supp. IV) 552, provides a broad mandate to "open agency action to the light of public scrutiny" (Department of the Air Force v. Rose, 425 U.S. 352, 372 (1976)). Yet its drafters recognized that allowing ready access to the federal government's vast accumulations of individualized data would invade personal interests that fall within the broad concepts of "privacy." Congress addressed that problem by providing an exemption that it described as protecting interests "equally important" as the basic policies of FOIA. /1/ And, unlike its approach to other FOIA exemptions, Congress directed that privacy interests be protected not by outright exemption of defined categories of information, but by case-by-case balancing of the privacy and public interests involved. Respondents attempt to truncate the balancing process by narrowly defining the range of "privacy" interests to be considered and by invoking a virtual per se rule that one has no "legitimate" interest in the practical obscurity of widely scattered and frequently unindexed "public records." And, although respondents appear to agree with much of our criticism of the court of appeals' approach to the "public interest" side of the balance, they again seek to avoid a real weighing of interests by insisting that any information regarding the backgrounds of government contractors readily outweighs the privacy interests at stake. In both respects, respondents' analysis fails to give full effect to the privacy protections Congress intended. 1. a. Although the term "privacy" is applied to a variety of legal interests, including common law rights and constitutional rights, the question before the Court in this case is a matter of statutory construction: What range of interests did Congress intend the courts to consider in determining whether a given disclosure would be an "unwarranted (or clearly unwarranted) invasion of personal privacy" (5 U.S.C. (& Supp. IV) 552(b)(6) and (7)(C))? Respondents attempt to restrict the range of such privacy interests. For example, they dismiss the "mere" possibility that the release of particular information may be embarrassing or even harmful to individuals, characterizing the "subjective" desire for privacy as unworthy of consideration (Resp. Br. 9-12, 27-28). Respondents insist that privacy interests in nondisclosure can be considered only when they relate to a "legitimate, objectively reasonable expectation" of secrecy (id. at 9). In accordance with that formulation, respondents invoke the law of torts or constitutional theories of privacy to show that individuals have no "right" of privacy with respect to their criminal history records (Resp. Br. 13-14, 18, 21-23, 29, 32; see also American Newspaper Publishers Association et al. (ANPA) Br. 13). Respondents' denial of a privacy "right" in this context is true in a sense but irrelevant to a proper FOIA analysis. We have never argued that the subjects of FBI criminal history records have a "right" to prevent the dissemination of such information. /2/ Indeed, the FBI disseminates that information for a number of statutorily authorized purposes. Nor could an individual with a criminal record invoke a tort-based "right" to prevent the press from publishing any records that it obtained. Cox Broadcasting Corp. v. Cohn, 420 U.S. 469 (1975). The lack of enforceable "rights," however, in no way negates the existence of strong privacy interests regarding such information, and respondents err in supposing that only privacy interests that give rise to such rights should be considered under FOIA. The fallaciousness of respondents' position is demonstrated by examples of matters that the courts have held to raise important "privacy" concerns under FOIA. The court in Baez v. United States Dep't of Justice, 647 F.2d 1328 (D.C. Cir. 1980), upholding an Exemption 7(C) claim, noted that "(t)here can be no clearer example of an unwarranted invasion of personal privacy than to release to the public that another individual was the subject of an FBI investigation" (id. at 1338 (quoting defense affidavit); see also Fund for Constitutional Gov't v. National Archives & Records Serv., 656 F.2d 856, 864-865 (D.C. Cir. 1981)). That correct result is inconsistent with respondents' focus on privacy "rights," for the subjects of FBI investigations have no "underlying right" to prevent government disclosure of such facts (which may be necessary to aid in apprehension or for other important law enforcement purposes), nor do they have any "legitimate expectation" of secrecy in such matters. See Paul v. Davis, supra; see also Pet. Br. 22 n.6. Thus, respondents' extensive discussions of cases involving the tort law of privacy concerning public records (e.g., Resp. Br. 21-23, 29, 33) are largely beside the point. Whether one focuses on the First Amendment, as in Cox Broadcasting, or on tort principles as such, those cases involve altogether different considerations, particularly the freedom of persons to speak of public record facts that they have learned. That such cases have found the subjects' interest in privacy outweighed by the great value our jurisprudence places on matters of "pure expression" (Cox, 420 U.S. at 495) does not mean that the privacy interests have no weight of their own. /3/ On the contrary, the willingness of at least some courts and commentators to let such privacy interests override even the right of free expression (see Pet. Br. 21-22 & n.5) indicates the importance of those interests. Nor does the balance that is struck in tort law or First Amendment cases dictate the outcome of the very different balance when persons seek to obtain information, rather than publish information they have already obtained. There is no general constitutional right of access to information held by the government. See generally Houchins v. KQED, Inc., 438 U.S. 1 (1978). The interest of persons in access to government information is, moreover, different from -- and weaker than -- that in publishing whatever information they obtain. See Press-Enterprise Co. v. Superior Court, 478 U.S. 1, 20 (1986) (Stevens, J., dissenting). The Court recognized that difference in Cox Broadcasting, indicating that, although the States cannot penalize the publication of public record information once obtained, the decision whether to release such information is a distinct policy question to be addressed by "political institutions" (420 U.S. at 496). Here, Congress has determined that release should not take place if it "could reasonably be expected to constitute an unwarranted invasion of personal privacy" (5 U.S.C. (Supp. IV) 552(b)(7)(C)). The legislative history of Exemption 6 evinces a broad purpose to exclude from mandatory disclosure "those kinds of files the disclosure of which might harm the individual." H.R. Rep. 1497, 89th Cong., 2d Sess. 11 (1966), FOIA Source Book 32 (emphasis added). Thus, contrary to respondents' contentions, the potential of information to harm or embarrass the individual subject is at the heart of the "privacy" concerns expressed by Congress. /4/ Although the interest of the individual in controlling the dissemination of information about himself may be outweighed in particular instances by a public interest in disclosure, there is no support for respondents' position that a court should not even consider that individual interest unless it rises to the level of a constitutional or tort-based "right." /5/ b. A second flaw that pervades respondents' arguments is their focus on underlying records of arrests, indictments, and convictions, rather than on the compilations at issue here, which raise distinct privacy concerns that respondents ignore. For example, respondents discuss at length the "public" nature of the various steps of the criminal justice process, arguing the importance of exposing those events to public view and decrying the possibility of "secret" criminal proceedings (Resp. Br. 13-20). Such arguments are misdirected, for we have never argued that an arrest -- much less an indictment or conviction -- is a "'private occurrence.'" /6/ It is an altogether different question, however, whether unlimited public access to a federal government file that collects millions of arrest and conviction records from thousands of scattered and often obscure sources implicates "personal privacy." Respondents deny that the large-scale compilation of criminal history records can make any difference for FOIA purposes (Resp. Br. 26). The simple notion that "(t)he focus of FOIA is on information" (ibid.), however, does not answer this complex question. Although this Court held in FBI v. Abramson, 456 U.S. 615 (1982), that records remain "law enforcement" records for Exemption 7 purposes even when they are incorporated into other kinds of records, it distinguished the case of internal agency records, which can lose their exempt status under Exemption 5, 5 U.S.C. 552(b)(5), by inclusion in final agency decisions (456 U.S. at 630). In each instance, the result is dictated by the purposes behind the particular exemption invoked (ibid.). Thus, the question here is whether the purpose of Exemptions 6 and 7(C) call for the courts to take account of or to ignore the distinct nature of large-scale compilations of data on individuals. Congress was specifically aware of the "privacy" concerns that were frequently voiced about government data banks, including criminal history data banks. See Pet. Br. 24-30. Before the 1974 FOIA amendments, Congress and others focused a great deal of attention on those issues -- consistently under the rubric of "privacy" (ibid.). Respondents miss the point of these developments (Resp. Br. 34-39). Of course, the consideration of various legislative proposals regarding criminal history records does not in itself show that Congress intended to bar disclosure under FOIA; but it does reflect an understanding of the range of interests comprised within the term "privacy," as Congress used it in Exemption 7(C). Respondents mischaracterize our discussion of the Privacy Act of 1974, 5 U.S.C. 552a, and its legislative history as an attempt to equate Privacy Act exemptions with FOIA exemptions (Resp. Br. 36-39). The Privacy Act is relevant specifically because its history reflects a clear congressional understanding that the term "privacy" includes the interest in limiting the dissemination of information collected in large government data banks, even when that information has its origin in "public" sources (see Pet. Br. 29-30; Pet. App. 44a (Starr, J., dissenting)). Respondents acknowledge that the Privacy Act's provisions are "not limited to private facts" (Resp. Br. 38), but they fail to grasp the significance of that proposition. Just as the Privacy Act is premised on a broader concept of "privacy" than respondents would prefer, Exemption 7(C) of FOIA -- passed contemporaneously with the Privacy Act -- is based on a similarly broad concept of privacy and should be interpreted accordingly. In failing to focus on the distinct problems of large-scale compilation of records, respondents also ignore or greatly underestimate the serious practical difficulties faced by both the government and the subjects of such files. Respondents' only answers to the well-documented and serious difficulties that the wide dissemination of criminal history creates for the individual subjects (Pet. Br. 33-34) are either to suggest speciously that considerations of "fairness" are not at issue (Resp. Br. 19) /7/ or to tout the "condemnation" of criminal violators -- including, apparently, the use of bare arrest records by private employers and credit bureaus -- as a positive virtue of their position (id. at 27, 30; see also ANPA Br. 14-15). Some degree of such harms may be inherent in the criminal justice process, but the question here is whether FOIA was intended to increase those harms by affording instant, nationwide availability to records that would otherwise be obscure. In light of Congress's express intent to bar disclosure of information that "might harm the individual" (H.R. Rep. 1497, supra, at 11, FOIA Source Book 32), it is difficult indeed to see such disclosures as consistent with FOIA, unless they serve the sort of public interests FOIA sought to advance. Moreover, the practical difficulties inherent in the maintenance of large data banks, such as the FBI's rap sheet files, are not answered by respondents' glib assertion that the FBI should simply make sure that there are no inaccuracies in its criminal history records (Resp. Br. 34 n.26). The reality is that mistakes will occur, and that the release of inaccurate information will be extremely harmful to the individuals affected. /8/ Respondents also do not face up to the administrative dilemma that their narrow view of "privacy" would create. Under the court of appeals' holding, the existence of any privacy interest in particular information would turn on whether, as a matter of fact, that information is "publicly available" (Pet. App. 41a-42a). However, for a variety of reasons -- varying local practices, unreported expungements, or even lost records -- that fact is often not known to the FBI when it receives an individual request. Protection of even those privacy interests that the court of appeals would acknowledge thus requires the agency to undertake new factual inquiries. Respondents' answer to this problem -- that the FBI should "presume" public availability (Resp. Br. 39-40 n.31) - would simply have the FBI ignore the privacy interests that Congress intended it to safeguard. /9/ The real answer to this dilemma is to recognize that it is the consequence of respondents' erroneously cramped view of "privacy." If, consistent with FOIA's legislative history, that term is read to encompass the distinct problems associated with compilations of obscure, scattered records, the FBI will be able to make a practical assessment of the privacy interests at stake without need for contrived presumptions of public availability. c. Respondents' position also is inconsistent with United States Dep't of State v. Washington Post Co., 456 U.S. 595 (1982), in which this Court rejected the argument -- made in a somewhat different form -- that matters of "public record" are beyond the scope of the FOIA privacy exemptions. As noted above, respondents criticize our invocation of broad definitions of "privacy" as entailing the interest of individuals in controlling the dissemination of information about themselves. See Resp. Br. 27; Pet. Br. 20-21. Although respondents themselves nowhere state a clear definition of "privacy," the terminology they invoke is that of "intimate details," recalling the threshold test they advanced, and this Court rejected, in Washington Post. Resp. Br. 10; see also ANPA Br. 11-12, 14. /10/ Washington Post focused not on the balancing test but on the "similar files" threshold of Exemption 6, which respondents and their supporting amici read as excluding any information reflected in "public records." The arguments made in support of that proposition, however, were very similar to those made now. In particular, respondents in that case argued that the "public records" at issue there were necessarily without "privacy" implications, urging that such records "would not be eligible for withholding under Exemption 6 no matter how that term (i.e., similar files") is interpreted" (Brief for the Respondent at 24-34, United States Dept. of State v. Washington Post Co., supra)." This Court rejected those arguments and remanded the case for consideration of "the effect of disclosure upon the privacy interests of (the subjects)" (456 U.S. at 603). Respondents now see in that disposition no impediment to their argument that there can be no privacy interest in "public record" information, but their view would render this Court's remand in that case a pointless formality: since the only information sought in that case was matter of public record "somewhere in the Nation," there would have been nothing for the lower courts to balance if respondents' view of "privacy" interests were correct. In seeking to avoid outright conflict with Washington Post, respondents struggle -- as did the court of appeals -- to articulate some limitation on their stance that there is no "legitimate" privacy interest in matters of "public record," no matter how obscure as a practical matter. Respondents suggest that "(t)here may be matters that are reflected in some public records that could properly be withheld (under FOIA)" (Resp. Br. 25-26 n.16), but they fail to specify any types of records that might qualify for such treatment or to explain how such a result could be squared with their "legitimate expectations" premise. Respondents also invoke the court of appeals' means of avoiding outright conflict with Washington Post, by suggesting that a showing of "particular harm" might give rise to a privacy interest (Resp. Br. 28). Apart from the inadequacy of that approach for reasons we have discussed (Pet. Br. 33 n.20), it is inconsistent with respondents' insistence elsewhere that harm to the individual is insufficient to create a "privacy" interest (Resp. Br. 11). In sum, both respondents' insistence on "legitimate expectations" of privacy "rights" and the court of appeals' dismissal of the privacy concerns posed by the dissemination of compilations of otherwise public records as "insignificant" (Pet. App. 20a) represent attempts to impose artificial limitations on the range of privacy interests to be considered. Both the legislative history of FOIA and this Court's approach in previous Exemption 6 cases, however, show that the proper approach is to take the full range of such interests into account, and to consider the "public record" status of information -- along with all other relevant factors -- in the judicious balancing of interests that Congress mandated. 2. Once the proper range of relevant privacy considerations is identified, the analysis under FOIA's privacy exemptions turns to a balancing of such considerations against the public interest in disclosure. See Rose, 425 U.S. at 373. As we have noted (Pet. 22; Pet. Br. 39), the court of appeals' most drastic departure from existing FOIA law in the present case was its holding that a court's assessment of the "public interest" for this purpose requires no evaluation of the particular information being sought, but simply a recognition of the overall disclosure policies of FOIA itself. In this Court, no one has attempted to defend the court of appeals' approach to this issue. The parties and amici are not entirely in agreement, however, as to the type of analysis that should be used in place of that of the court of appeals. Our own approach to this issue is founded on the policies of FOIA itself: in assessing whether a particular invasion of privacy is "unwarranted," a court should consider whether disclosure is supported by the essential policies of FOIA itself, of "open(ing) agency action to the light of public scrutiny" (see Rose, 425 U.S. at 372). This obviously requires specific consideration of the particular information sought: although disclosure of the FBI rap sheet of "one's otherwise inconspicuously anonymous next-door neighbor" (Pet. App. 45a (Starr, J., dissenting)) surely does not advance FOIA's goals, disclosure of the rap sheet of a high-ranking government official may do so, if it reflects crimes that call into question the official's ability to perform public duties competently and honestly. Respondents themselves do not appear to take issue with our basic approach to the public interest side of the balance, geared to the "core purposes" of FOIA. In arguing for disclosure in the present case, respondents argue that disclosure will serve the "basic purpose" of FOIA (Resp. Br. 41, 45 (citing NLRB v. Robbins Tire & Rubber Co., 437 U.S. 214 (1978)). Although we disagree strenuously with respondents' assessment of the extent to which the records at issue in the present case touch on FOIA's core purposes, we and they are in apparent agreement as to the essential test of the "public interest." /12/ Some amici, however, do attack our focus on FOIA's core purposes in assessing the public interest in disclosure (e.g., NARFE Br. 11-13). Yet, in their attempts to deny the existence of ascertainable "core purposes" of FOIA, they only succeed in restating them. The need to obtain information "'to evaluate federal programs and formulate wise policies'" (id. at 11, quoting Soucie v. David, 448 F.2d 1067, 1080 (D.C. Cir. 1971)), for example, is among the core purposes we have articulated. And, although the legislative history of FOIA indeed invokes James Madison's words about the power of knowledge (see NARFE Br. 11-12), the purpose of that knowledge -- i.e., to ensure that the people can "be their own governors" -- shows again that FOIA is directed chiefly to illuminating the workings of government to enable citizens to make informed political choices. Contrary to NARFE's repeated assertions that a core purposes test is "unworkable," Congress itself has articulated just such a test in the provision of FOIA that waives or reduces fees "if disclosure of the information is in the public interest because it is likely to contribute significantly to public understanding of the operations or activities of the government" (5 U.S.C. (Supp. IV) 552(a)(4)(A)(iii)). That standard shows that, in Congress's view at least, there is indeed a principled and workable means of assessing the public interest in the release of particular information. In denying the usefulness of this approach, NARFE falls into the same trap as the court of appeals by failing to recognize that a court can meaningfully assess "the interest of the general public in release of the records themselves" (Pet. App. 26a). Amici Public Citizen et al. argue that various "weighty objectives" apart from the purposes of FOIA itself may be served by the disclosure of information subject to the privacy exemptions (Pub. Citizen Br. 18-19). It is unclear, however, why FOIA should be construed to further those objectives, which by hypothesis were not a purpose of FOIA's enactment, at the expense of the personal privacy concerns that Congress expressly acknowledged and addressed in the statute. The awkwardness, standardlessness, and unpredictability of allowing federal judges to invade personal privacy whenever they think any countervailing objective is sufficiently "weighty" is what properly led the court below to seek a less "idiosyncratic" approach (Pet. App. 21a-23a). The court went too far in insisting that every approach to this issue is necessarily standardless, but the court was hardly unjustified in eschewing the approach that amici now advocate, in which the "public interest" depends not on identifiable concerns of FOIA itself but on the ingenuity of counsel in identifying an objective that will strike the judge as worthy. A focus on the policies of FOIA itself solves this problem and provides a principled structure for the public interest analysis (see Pet. Br. 45-47). 3. As noted above, respondents themselves invoke the "basic purpose" of FOIA, and they attempt to characterize the materials they seek as information the disclosure of which is "needed to check against corruption and to hold the governors accountable to the governed" (Resp. Br. 41, 45 (quoting NLRB v. Robbins Tire & Rubber Co., 437 U.S. at 242)). These attempts fail, however, since respondents can establish no logical connection between such vaunted public interests and any information actually at issue in this case. Respondents repeatedly emphasize that the ultimate target of their investigations was a matter that would raise a significant public interest -- uncovering possible corruption by a Member of Congress. That fact alone, however, falls far short of establishing that any and all information respondents seek serves that public interest sufficiently to warrant a significant invasion of privacy. The district court properly recognized that, in order for the release of information to be warranted under Exemption 6 or 7(C), there must be some logical connection between the information at issue and the public interest (Pet. App. 57a). Respondents criticize the district court for drawing the conclusion that the value to the public of any information actually at issue in this case is insignificant. They argue that the public should "decide for itself" the significance of any information (Resp. Br. 42 (quoting Washington Post Co. v. United States Dep't of HHS, 690 F.2d at 264)). Respondents' argument, however, proves too much. The case they cite (unlike the present one) involved a situation where the information sought plainly had a logical connection to a distinct public interest, /13/ and the court merely pointed out that the goal of FOIA is to let the public decide "for itself" whether the government was taking proper steps in light of this information. When the question is whether there is any such logical connection at all, however, FOIA obviously does not require that the public decide the relevance of every item "for itself"; that would be to require disclosure to the public as a step in the very balancing process mandated by the statute to determine whether disclosure is warranted. Congress has instead entrusted the task of balancing the public and private interests at stake to the agencies, subject to de novo review by the courts. /14/ In their attempt to substitute a per se rule for the careful assessment of the public interest in disclosure mandated by the statute, respondents assert that there was a significant public interest in "(a)ny evidence" of a criminal record by any of the principles of Medico Industries (Resp. Br. 47). That unsupported assertion sweeps far too broadly and would eviscerate the protections of the FOIA privacy exemptions for large classes of persons. To use a hypothetical example, a 30-year-old arrest for disorderly conduct, which did not lead to an arrest, would not bear on the fitness of a firm controlled by the arrestee to perform government contracts. For the reasons discussed above, release of such information would significantly invade that arrestee's privacy interests, and it is plain that such an invasion of privacy would in no way be "warranted" by any discernible public interest. But respondents' approach to this issue would require release of such information, and presumably of any personal information involving any government contractor or any government employee. Such a ready finding of a public interest in disclosure would mock Congress's determination to temper FOIA's policies of disclosure with protection of "certain equally important rights of privacy" (S. Rep. 813, 89th Cong., 1st Sess. 3 (1965), FOIA Source Book 38). Respondents also stress the "newsworthy" nature of the events surrounding Representative Flood and the Medicos (Resp. Br. 43-44). As with their privacy discussion, however, here respondents confuse the freedom to publish information with their interests in obtaining it. We would not dispute the proposition that the Medicos were "newsworthy" in some sense, and, as noted above, respondents doubtless enjoy a broad right to publish such information as they can obtain. Those considerations do not, however, show that there is any connection between the information they seek and the sort of public interests served by FOIA. We cannot, of course, state here what information, if any, is to be found in the records respondents seek. /15/ But there is no reason to think that the judges who have actually examined any withheld records -- and rejected respondents' position -- have ignored the interests that respondents propound. A connection between the records and the public interest would be most obvious if the subject himself were a public official, but it is also conceivable that the requisite public interest would be present if there were any records tending to show that contracts were improperly awarded, that Medico Industries was unqualified to receive government contracts, that contract performance was inadequate, or that there was "corruption" of any sort surrounding any government contracts. If, however, a court reviewing the material in camera -- as the district court did -- can satisfy itself that the information at hand has no logical connection to any such public interests, then it should not order the disclosure of information that touches on important privacy concerns. That is precisely what the district court did in the present case (Pet. App. 57a-58a, in a judgment confirmed by the only member of the court of appeals who purported to engage in a balancing of interests (Pet. App. 49a). This Court too is fully capable of examining any withheld records in light of respondents' arguments and reaching what we believe to be the inevitable conclusion: that there are no records in this case imbued with a sufficient public interest to overcome the privacy interest of Charles Medico. For the foregoing reasons and the reasons stated in our opening brief, the judgment of the court of appeals should be reversed. Respectfully submitted. CHARLES FRIED Solicitor General SEPTEMBER 1988 /1/ See S. Rep. 813, 89th Cong., 1st Sess. 3 (1965), reprinted in Subcomm. on Administrative Practice and Procedure, Senate Comm. on the Judiciary, 93d Cong., 2d Sess., Freedom of Information Act Source Book: Legislative Materials, Cases, Articles 38 (Comm. Print 1974) (hereinafter FOIA Source Book). The uniquely important personal interests served by Exemption 6, 5 U.S.C. 552(b)(6), and the even broader Exemption 7(C), 5 U.S.C. (Supp. IV) 552(b)(7)(C), render inapt respondents' insistence that the exemptions at issue here be "narrowly construed" (Resp. Br. 27 (quoting United States Dep't of Justice v. Julian, No. 86-1357 (May 16, 1988), slip op. 6)). Moreover, as this Court has also noted, all of the FOIA exemptions are integral parts of the Act itself and serve distinct congressional policies. See EPA v. Mink, 410 U.S. 73, 80 (1973); Baldrige v. Shapiro, 455 U.S. 345, 352 (1982). /2/ The court of appeals implied the existence of a constitutionally based right of privacy in this area in Menard v. Mitchell, 430 F.2d 486, 490-492 (D.C. Cir. 1970), but that implication is effectively discredited by Paul v. Davis, 424 U.S. 693 (1976). /3/ Similarly, the academic writings that ANPA cites (ANPA Br. 12-15) questioning the advisability of recognizing a general "right" to control information about oneself are beside the point, since they focus on tort law, and not on access to information. Moreover, the author of the article on which ANPA principally relies acknowledged that his stance on these issues is well out of the mainstream, and bemoaned the "legislative trend" toward greater protection of personal privacy. See Posner, The Right of Privacy, 12 Ga. L. Rev. 393, 404, 406 (1978). /4/ Although respondents invoke the principle -- borrowed from Fourth Amendment jurisprudence -- of one's "legitimate expectation of privacy," they make only a brief and unsuccessful effort to relate that principle to FOIA itself (Resp. Br. 11-12). The cases respondents cite do not support the proposition that the balancing of interests under Exemptions 6 and 7(C) is limited to areas of such "legitimate expectations"; on the contrary, these cases themselves focus on whether the information at issue is of an "embarrassing" nature. See Washington Post Co. v. United States Dep't of HHS, 690 F.2d 252, 262 (D.C. Cir. 1982); International Bhd. of Elec. Workers v. United States Dep't of Housing & Urban Dev., 593 F. Supp. 542, 544 (D.D.C. 1984), aff'd, 763 F.2d 435 (D.C. Cir. 1985). Moreover, the commentary that respondents cite acknowledges that protection of individuals from harm and embarrassment was the chief goal of Congress in enacting Exemption 6. See Kronman, The Privacy Exemption to the Freedom of Information Act, 9 J. Legal Studies 727, 739 (1980). /5/ In arguing that a focus on the embarrassing nature of information "proves too much" (Resp. Br. 28), respondents merely illustrate how readily the proper balancing test can be applied. Their examples of "embarrassing" facts that should not qualify for withholding under Exemption 6 -- those of a government official who has abused public funds, and of a citizen who has sold government secrets -- implicate the exact sort of public interests in responsible government that FOIA was meant to further. The invasion of privacy is real and deserves to be considered, but it is plainly "warranted." /6/ See Resp. Br. 17 (quoting W. Rehnquist, Is an Expanded Right of Privacy Consistent with Fair and Effective Law Enforcement?, Nelson Timothy Stephens Lectures, University of Kansas Law School, Pt. I, at 13 (Sept. 26-27, 1974)). The lecture that respondents quote points out, immediately after the passage quoted by respondents, that "to conclude that an event is not wholly private does not mean that an individual has no interest in limiting disclosure or dissemination of information." See also Rehnquist, Is an Expanded Right of Privacy Consistent with Fair and Effective Law Enforcement?, 23 Kan. L. Rev. 1, 9 (1974). /7/ Although respondents suggest that "education" will resolve any problems of unwarranted actions based on arrest records (Resp. Br. 19-20), the extravagantly speculative inferences respondents themselves draw from William Medico's arrest records (id. at 47-48) belie their attempt to characterize the problem of unfairness in the use of such records as unimportant. The misuse or misunderstanding of limited information is also reflected in ANPA's unsupported characterization of Charles Medico as a "reputed organized crime figure" (ANPA Br. 3). /8/ Both respondents and ANPA argue that the problem of inaccuracies will be alleviated by public release of such records (Resp. Br. 34 n.26; ANPA Br. 16). But public dissemination of such information is the very harm to be avoided. Respondents' argument, if credited, could be used to justify any invasion of privacy, on the ground that it would allow the public to monitor the government's collection of inaccurate or embarrassing information. The interest of individuals in assuring accuracy is far better served by the Justice Department's regulations, which allow an individual to obtain access to his own rap sheet. See 28 C.F.R. 20.34. /9/ Indeed, respondents' footnote 31 now appears to abandon the limitation of their request to "matters of public record" (J.A. 33). Respondents now show that they do not really want petitioners to undertake outside research into the public availability of the records -- outside research that FOIA does not mandate (see Pet. Br. 36-39) -- but instead want petitioners to avoid such research by presuming public availability (and thereby negating the "public record" limitation that respondents placed on their request). Thus, although we believe that withholding of the requested records in this case (if any exist) is proper even if we assume that every record at issue is a "matter of public record" somewhere, this Court should not be misled by assertions of respondents and their amici (e.g., ANPA Br. 9-10) that only "public" records will be affected by the decision in this case. /10/ The similarities between the present case and Washington Post include not only the issues but the parties as well. The Washington Post, an amicus in the present case, was the respondent in that case, represented by the same law firm representing the present respondents. Respondent Reporters Committee appeared as amicus in the earlier case, joined by ANPA. /11/ Respondents' current arguments parallel those advanced in Washington Post in other respects. Here, as there, respondents stress the view that harm or embarrassment to the subjects of government information should be ignored, and that the courts should instead focus on "legitimate expectations" of privacy. See Brief for Respondents at 20-23, United States Dep't of State v. Washington Post Co., supra. The present arguments regarding the "public" nature of criminal proceedings, moreover, are closely analogous to those made with respect to citizenship proceedings in Washington Post (id. at 24-27, 31-33). See also Brief Amici Curiae of the American Newspaper Publishers Association, et al., at 3, United States Dep't of State v. Washington Post Co., supra ("such uniquely public information (records of citizenship) can never be withheld from the public in the name of 'privacy'"). /12/ In one respect, however, respondents go further than we do in disagreeing with the analysis of the public interest issue by the court of appeals. Respondents attempt to insert into the analysis consideration of the "purpose" of the specific FOIA request (Resp. Br. 41; cf. Pet. App. 23a-24a). For reasons we have summarized (Pet. Br. 47 n.35), we do not think that factor has an appropriate place in Exemption 6 or Exemption 7(C) balancing. Instead, in our view, Judge Leventhal got it right long ago when he (and the court) repudiated the D.C. Circuit's initial false step on this issue (Getman v. NLRB, 450 F.2d 670, 675, 677 n.24 (1971)) and explained that "(a) general balancing of the privacy loss from release to the public and the public interest in disclosure to the public would carry out the discernible intent of the Senate Report without conflicting with the Act's broad purpose to enable 'any person' to request disclosure without making a showing that he was 'properly and directly concerned(') with the information." Ditlow v. Shultz, 517 F.2d 166, 172 n.21 (D.C. Cir. 1975) (citations omitted; emphasis added). Because respondents claim no special interest in the requested information other than their interest in disseminating it to the general public, it is doubtful that anything in this case turns on the difference between the Getman approach that respondents advocate and the Ditlow approach that we advocate. Nevertheless, the farfetched claims of some amici call for brief response. The claim of amici Public Citizen et al. (Br. 21-22) that this issue was resolved in Department of Justice v. Julian, supra, is wrong. Julian fashioned a narrow exception -- in the context of applying Exemption 5 to a request by the subject of the records -- to the general principle that a party's rights under FOIA "are neither increased nor decreased by reason of the fact that it claims an interest in the (requested records) greater than that shared by the average member of the public" (NLRB v. Sears, Roebuck & Co., 421 U.S. 132, 143 n.10 (1975); see also EPA v. Mink, 410 U.S. at 86). Public Citizen's position in this case would require the Court to fashion a new and entirely different exception to that general principle. Likewise, the claim by the National Association of Retired Federal Employees (NARFE) that in all cases other than this one the courts have followed the Getman approach (Br. 6) is not at all accurate. Virtually the only reasoned discussion of the issue to be found anywhere in the case law is in Getman v. NLRB, supra; Ditlow v. Shultz, supra; Washington Post Co. v. United States Dep't of HHS, 690 F.2d 252, 258, 259 n.17 (D.C. Cir. 1982); and United States Dep't of the Air Force v. FLRA, 838 F.2d 229, 233 (7th Cir. 1988), petition for cert. pending, No. 88-354. Apart from Getman, which has been repudiated within its own circuit, all of those cases support our approach. /13/ That case involved financial disclosures made by consultants employed by the National Cancer Institute, which had a strong and direct connection to the public interest in ascertaining any conflicts of interest by such persons. See 690 F.2d at 264. /14/ In light of the availability of de novo review, the charge of amicus ANPA that our analysis would "imbue the FBI with virtually unlimited discretion" (ANPA Br. 18) is baseless. /15/ Contrary to the assertions of amicus Public Citizen, there is nothing in the least "absurd" (Pub. Citizen Br. 23) about our continued insistence that any confirmation that Charles Medico does or does not have a criminal record would itself give away information that the FOIA privacy exemptions seek to protect. The court of appeals did not, as Public Citizen claims, "explicitly confirm()" the existence of a criminal record (ibid.), and any passages in the court's opinion that might seem to imply that such a record exists (see Pet. App. 16a, 22a) are merely unfortunate -- and not necessarily accurate -- lapses from the court's duty to safeguard, pendente lite, the very information whose public availability is at issue. Furthermore, if anything is "absurd," it is the sort of public disclosure in the course of the FOIA litigation itself that Public Citizen apparently favors -- a listing of any rap sheet entries, indicating type, age, and nature of the offense (Pub. Citizen Br. 24) -- which would preemptively destroy the very privacy interest that is at issue. Indeed, the simple statement that an individual "has a criminal record" is itself highly damaging information, which may be all the more harmful because it is incomplete.