HOWARD, Circuit Judge.
Though clouded by an airline disaster and claims of a government cover-up, this case ultimately turns on a relatively straightforward question: must the government release certain information? Plaintiff-Appellant Thomas Stalcup brought this Freedom of Information Act ("FOIA") suit against the Central Intelligence Agency ("CIA"), seeking two documents from an investigation into the crash of TWA Flight 800. Stalcup also sought the names of the eyewitnesses interviewed during the investigation. The district court rejected Stalcup's requests, concluding that FOIA permitted the agency to withhold the sought-after material. Because we agree with each of the district court's conclusions, we affirm its decision to grant summary judgment for the CIA.
I.
On July 17, 1996, TWA Flight 800 exploded in mid-air and crashed eight miles south of Long Island, New York. Pursuant to its obligations under 49 C.F.R. § 800.3(a), the National Transportation Safety Board ("NTSB") launched an investigation into the tragedy. The Board quickly arrived at three possible explanations for the crash: a bomb, a missile, or a mechanical failure.
Given the possibility of criminal or terrorist activity, the FBI joined the probe. A central component of the Bureau's task was to interview eyewitnesses. Many of the 244 individuals who were interviewed described a streak of light rising up to the plane just before the explosion. Given the consistency of that narrative, the FBI asked the CIA to analyze the accounts and explore the likelihood of a missile strike.
The CIA reviewed the eyewitness reports along with raw flight and radar data. It concluded that the eyewitnesses had not seen a missile soaring towards the plane but, instead, had observed the burning aircraft in various stages of dismantling. On March 28, 1997, the CIA passed this analysis along to the FBI, which ultimately reached the same conclusion. In November 1997, the CIA publicized these results in a video entitled: "TWA Flight 800: What Did the Eyewitnesses See?"
As new data emerged, the CIA continued its work. For instance, in 1998 it produced a 17-page draft report analyzing new radar tracking data ("Analysis of Radar Tracking"). At that time, it also created an 18-page draft report assessing the plane's flight path ("Dynamic Flight Simulation"). Both documents contained recommendations to the agency about how the newly acquired data should impact the analysis. In 1999, the CIA relayed this new evaluation to a NTSB-sponsored group studying the eyewitness accounts.
On August 23, 2000, the investigation, which had been the largest and most expensive in the NTSB's history, reached its terminus. The Board adopted the CIA's assessment of the eyewitness accounts and concluded that a mechanical explosion in the center wing fuel tank had caused the crash. The NTSB distributed a final report detailing these findings.
A decade later, theorizing that the CIA was covering up that the true cause of the crash was a missile strike, Stalcup sent the CIA a letter requesting "copies of all data, images, video, documents and/or other information related to or a product of the CIA's involvement in the TWA Flight 800 investigation." He also asked for the "`Technical Analysis Briefing: TWA Flight 800'...; ... all eyewitness documents, reports, videos, images, and/or audio provided to the CIA ... [and] any and all correspondence ... regarding the CIA's ... analysis of the eyewitness evidence."
The CIA first disclosed twenty-five documents that it had previously released in response to a similar FOIA request. Unsatisfied with the CIA's response, Stalcup brought this FOIA action. 5 U.S.C. § 552. The complaint, filed in the District of Massachusetts, asked the court to order the CIA to disclose additional material. As the litigation progressed, the CIA provided Stalcup with forty-nine documents, a DVD, eighty-nine partially-redacted documents, and fourteen documents created by other agencies. The agency also filed a Vaughn index with the court detailing its redactions and withholdings.
Nonetheless, Stalcup demanded more. He requested unredacted versions of the 1998 Analysis of Radar Tracking document (only the technical data, graphs, and certain headings were initially provided); the 1998 Dynamic Flight Simulation analysis (only the headings had been released); and the names of the eyewitnesses interviewed by the FBI.
In due course, the CIA moved for summary judgment, which the district court granted. The court concluded that the agency had properly withheld the requested documents under the deliberative process exemption of the law, 5 U.S.C. § 552(b)(5)(hereinafter "exemption 5"), and had appropriately redacted the eyewitness names pursuant to the law enforcement exemption of the act, 5 U.S.C. § 552(b)(7)(C)(hereinafter "exemption 7(C)"). The court also rejected Stalcup's contentions that disclosure of the information was required in light of alleged government misconduct. Finally, the court concluded that the CIA had performed an adequate search in response to the FOIA request. This timely appeal followed.
II.
We review a district court's grant of summary judgment in a FOIA case de novo. Moffat v. U.S. Dep't of Justice, 716 F.3d 244, 250 (1st Cir.2013). Accordingly, we draw all reasonable inferences in favor of the non-moving party, and will only affirm the district court's decision if no genuine dispute of material fact exists and a party is entitled to judgment as a matter of law. Fed.R.Civ.P. 56.
III.
FOIA is an important tool in holding the government accountable because it provides citizens a means to "know what their government is up to." Carpenter v. U.S. Dep't of Justice, 470 F.3d 434, 437 (1st Cir.2006) (internal quotation marks and citation omitted). By establishing a presumption in favor of agency disclosure, Congress aimed to "expose the operations of federal agencies to public scrutiny." Providence Journal Co. v. U.S. Dep't of Army, 981 F.2d 552, 556 (1st Cir.1992). The need for transparency, however, must be balanced with the goal of the "efficient administration of government." Carpenter, 470 F.3d at 438. Accordingly, Congress provided a number of exemptions that permit an agency to withhold certain documents from release. To fulfill the broad purposes of FOIA, we construe these exemptions narrowly. FBI v. Abramson, 456 U.S. 615, 630, 102 S.Ct. 2054, 72 L.Ed.2d 376 (1982) (citation omitted).
Two exemptions take center stage in this appeal: exemption 5, the deliberative process exemption, and exemption 7(C), the law enforcement exemption. We note at the outset that the Ninth Circuit has recently addressed a nearly identical challenge to the exact materials at issue here. Lahr v. Nat'l Transp. Safety Bd., 569 F.3d 964 (9th Cir.2009). After an in camera review, that court determined that the government had properly withheld the material. Though we are not bound by that conclusion, and would independently arrive at the same result based on this record alone, we do find Lahr to be persuasive. After considering the Ninth Circuit's thoughtful analysis, the CIA's extensive declaration in this case, and the absence of any viable argument to the contrary, we are left satisfied that the CIA's response to the FOIA request accorded with the law.
A. Exemption 5: The Deliberative Process Exemption
Exemption 5 of FOIA, the deliberative process exemption, permits an agency to withhold "inter-agency or intra-agency memorandums or letters which would not be available by law to a party other than an agency in litigation with the agency." 5 U.S.C. § 552(b)(5). It thus generally exempts from disclosure documents containing work product, attorney-client correspondence, or material that is otherwise "privileged in the civil discovery context." NLRB v. Sears, Roebuck & Co., 421 U.S. 132, 149, 154, 95 S.Ct. 1504, 44 L.Ed.2d 29 (1975). The government carries the burden of establishing the applicability of the exemption and must show: (1) that the withheld material is an inter- or intra- agency memorandum — an uncontested issue here; (2) that the document is deliberative; and (3) that it is predecisional. Providence Journal Co., 981 F.2d at 557.
The government asserts that it properly withheld the 1998 Analysis of Radar Tracking and the 1998 Dynamic Flight Simulation under the exemption. Stalcup, however, contends that the documents are neither deliberative nor predecisional. Though the analysis intersects at times, the two criteria are distinct prongs and must be examined separately.
To satisfy the "deliberative" element of the exception, a document must reflect "the give-and-take of the consultative process." Petroleum Info. Corp. v. U.S. Dep't of Interior, 976 F.2d 1429, 1434 (D.C.Cir.1992) (citations and internal quotation marks omitted). More specifically, the document must: "(i) form[] an essential link in a specified consultative process, (ii) reflect[] the personal opinions of the writer rather than the policy of the agency, and (iii) if released, ... inaccurately reflect or prematurely disclose the views of the agency." Providence Journal Co., 981 F.2d at 559 (internal quotation marks and citations omitted). Conversely, a document "consisting only of compiled factual material or purely factual material contained in deliberative memoranda and severable from its context" is subject to disclosure. EPA v. Mink, 410 U.S. 73, 87-88, 93 S.Ct. 827, 35 L.Ed.2d 119 (1973).
Stalcup speculates that the two documents primarily contain factual information respecting the cause of the Flight 800 crash. He postulates that neither focuses on any policy issue or legal question. The district court, as he sees it, thus failed to differentiate between "materials reflecting deliberative or policymaking processes on the one hand, and purely factual, investigative matters on the other." Id. at 89, 93 S.Ct. 827.
Stalcup's contention is dubious. Though the Supreme Court in Mink did distinguish between purely factual material and policy prescriptions, the Court also emphasized that "Congress sensibly discarded a wooden exemption that could have meant disclosure of manifestly private and confidential policy recommendations simply because the document containing them also happened to contain factual data." Mink, 410 U.S. at 90, 93 S.Ct. 827. The question is not merely whether the documents contain factual information — or even whether the document is predominantly comprised of findings of fact — but rather the degree to which the facts are indissolubly linked to the broader analysis. Thus, although the two documents may contain certain facts, that alone does little to advance the analysis.
Instead, the issue hinges on whether the documents were "prepared to facilitate and inform a final decision or deliberative function entrusted to the agency." Providence Journal Co., 981 F.2d at 560. In Lahr, the Ninth Circuit reviewed the documents at issue here and concluded that each contains a preliminary analysis of newly acquired data and that both discuss recommendations for agency management to consider. Lahr, 569 F.3d at 983. For instance, the Dynamic Flight Simulation "exposes in detail the thought process of the CIA analysts involved in calculating the simulated flight path, as well as language reflecting the decisionmaking process." Id. Moreover, if released, both documents would "expose the agency's internal deliberations in such a way that would discourage candid discussion and effective decisionmaking." Id. Although factual findings are indeed included in the documents, they are tethered to broader policy considerations. Id.
The Ninth Circuit's conclusion coheres with the CIA's declaration and Vaughn index in this case. The Analysis of Radar Tracking, for example, has the label "draft" on it, was shown to the NTSB but never finalized, contains analytical opinions, assessments, and judgments, and was passed on to agency management for possible policy changes. The Dynamic Flight Simulation, meanwhile, is also labeled "draft", includes opinions and information relevant to calculating the flight path, identifies challenges to conducting the analysis, and was also provided to agency management but never finalized. Stalcup has provided no reason to question those descriptions.
Further, to the extent that the documents contain purely factual information severable from the analysis, any worries can be put to rest. The district court in Lahr had ordered the CIA to provide any factual material that could be separated. Lahr, 569 F.3d at 983. In turn, the CIA released — and provided to Stalcup in this case — certain radar data, graphs, and headings from those documents. Stalcup has failed to raise a genuine dispute that any additional information is severable.
In his reply brief, Stalcup presents a number of new arguments. Among them is his contention that the passage of time undermines the basis for enforcing the exemption. See National Security Archive v. Central Intelligence Agency, 752 F.3d 460 (D.C.Cir.2014). We need not consider that claim here since it, like his other new arguments, is waived. See, e.g., Carpenter, 470 F.3d at 440 n. 9.
Stalcup next takes aim at the predecisional prong of the requirement. To satisfy this element, a document must be prepared "in order to assist an agency decisionmaker in arriving at his [or her] decision." Town of Norfolk v. U.S. Army Corps of Eng'rs, 968 F.2d 1438, 1458 (1st Cir.1992) (citations omitted). An agency claiming the exemption must "(i) pinpoint the specific agency decision to which the document correlates, (ii) establish that its author prepared the document for the purpose of assisting the agency official charged with making the decision, and (iii) verify that the document precedes, in temporal sequence, the decision to which it relates." Providence Journal, 981 F.2d at 557 (internal quotation marks and citations omitted).
Stalcup argues that the CIA determined the cause of the crash in 1997, but only created the two documents at issue in 1998 after it had reached that decision. From his perspective, the documents were produced solely to reconcile new data that did not support the agency's initial conclusion. Thus, he asserts, the CIA improperly attempted to "explain agency action already taken or an agency decision already made." Sears, Roebuck & Co., 421 U.S. at 153, 95 S.Ct. 1504.
The roadblock in Stalcup's path is that the CIA's task did not end in 1997 when it reached its initial conclusion. Instead, as would any reasonable government entity presented with new data, it undertook to determine whether its prior assessment was accurate or whether it needed to change its position.
Again, the Lahr court describes the documents well. The Dynamic Flight Simulation is "based on additional and more complete data that became available over the course of the investigation." Lahr, 569 F.3d at 982-83. The court further observes that, "[a]lthough it is dated after the November 1997 CIA animation, it was clearly prepared for the specific purpose of aiding the agency in its determination of the likely flight path of the aircraft following the explosion, a determination central to the CIA's task of explaining what the eyewitnesses actually saw." Id. at 983. The Analysis of Radar Tracking, too, "contains conclusions and thoughts of CIA analysts concerning the viability and accuracy of certain radar data," and therefore forced the CIA to ask whether it needed to reconsider its decision. Id. at 983. Stalcup offers us no basis upon which to reject that reasoning.
Finally, Stalcup maintains that, even if the twin justifications of exemption 5 are met, release of the documents is nevertheless appropriate. Emphasizing his belief that the CIA participated in a cover-up, he urges us to endorse a government misconduct waiver. Drawing from the civil litigation context, Texaco P.R., Inc. v. Dep't of Consumer Affairs, 60 F.3d 867, 885 (1st Cir.1995), he argues that the exemptions should be waived if the evidence "warrant[s] a belief by a reasonable person that the alleged government impropriety might have occurred." Nat'l Archives & Records Admin. v. Favish, 541 U.S. 157, 174, 124 S.Ct. 1570, 158 L.Ed.2d 319 (2004). He points to the approach of the District of D.C., which has employed a narrow waiver in FOIA cases presenting "extreme government wrongdoing." Neighborhood Assistance Corp. of Am. (NACA) v. U.S. Dep't of Hous. & Urban Dev., ___ F.Supp.3d ___, ___, 2013 WL 5314457, at *8 (D.D.C. Sept. 24, 2013); see also Enviro Tech Int'l, Inc. v. EPA, 371 F.3d 370, 376 (7th Cir.2004) (assuming that a waiver might apply for ultra vires actions by an agency). Contra Appleton Papers, Inc. v. EPA, 702 F.3d 1018, 1022 (7th Cir.2012); Hoover v. U.S. Dep't of Interior, 611 F.2d 1132, 1142 (5th Cir.1980).
This is a road we need not travel. To the extent that we might similarly recognize a narrow waiver doctrine in the FOIA context, it would not apply in this case. Even assuming that Stalcup could show a scintilla of support for his claim, he still fails to connect the requested materials to the alleged government misconduct. Courts that have adopted such a waiver in the FOIA milieu also require a party to establish a nexus between the misconduct and the requested documents. See Judicial Watch of Fla. v. U.S. Dep't of Justice, 102 F.Supp.2d 6, 15 (D.D.C.2000). This is a sound way to avoid a litigant's mere fishing expedition into government action. Stalcup's silence as to how the requested documents would shed any new light on the alleged misconduct is fatal to his claim.
Ultimately, the CIA properly withheld the materials under exemption 5. Both documents are deliberative and predecisional in nature, and any misconduct waiver would be inapplicable.
B. Exemption 7(C): The Law Enforcement Exemption
In addition to the two documents, Stalcup seeks the names of the eyewitnesses interviewed by the FBI as part of the investigation. Although the CIA withheld the names under multiple exemptions — it asserted that exemption 6 protecting "personnel and medical files" would also apply — the parties and the district court correctly focused on exemption 7(C). See U.S. Dep't of Justice v. Reporters Comm. for Freedom of the Press, 489 U.S. 749, 756, 109 S.Ct. 1468, 103 L.Ed.2d 774 (1989) (noting that the protections of 7(C) are more extensive than those of 6). Exemption 7(C) shields information compiled for law enforcement purposes when the release of such records "could reasonably be expected to constitute an unwarranted invasion of personal privacy." 5 U.S.C. § 552(b)(7)(C). That analysis requires a court to balance the relevant public and private interests implicated by disclosure. Maynard v. CIA, 986 F.2d 547, 566 (1st Cir.1993).
Stalcup asserts that the eyewitnesses have an insignificant privacy interest. Generally, he says, the names of witnesses to crimes or accidents are made public and, therefore, the eyewitnesses in this investigation should be treated no differently. Since the NTSB had the authority to call the witnesses to testify, he adds, the witnesses have a reduced expectation of anonymity.
This argument ignores the Supreme Court's observation that an individual's privacy interest is "at its apex" when he or she is involved in a law enforcement investigation. Favish, 541 U.S. at 166, 124 S.Ct. 1570 (citation omitted). Indeed, as we have said before, "[t]his Court has long protected the identities of witnesses and informants in law enforcement records," even when the individual is not the subject of the investigation. Carpenter, 470 F.3d at 439.
The other aspect of this argument, that the NTSB's subpoena power minimized the privacy interest, is equally unavailing. It mistakenly assumes that the mere possibility of being called as a witness is somehow equivalent to an individual voluntarily abdicating his or her privacy. Stalcup offers no authority supporting that position. Moreover, even assuming that a witness had been required to testify, that does not necessarily diminish his or her privacy interest. Moffat, 716 F.3d at 251 (stating that "prior revelations of exempt information do not destroy an individual's privacy interest").
Gaining no traction there, Stalcup turns to United States v. Weber Aircraft Corp., 465 U.S. 792, 104 S.Ct. 1488, 79 L.Ed.2d 814 (1984). In Weber, the Supreme Court noted that the government provided the witnesses a guarantee that their testimony would not be released outside of the relevant investigation. Weber, 465 U.S. at 797 n. 11, 104 S.Ct. 1488. In Stalcup's view, this guarantee of confidentiality greatly enhanced the privacy interests at stake. He then posits that, absent such a promise in this case, no significant privacy interest can be established.
Even if the cited footnote in Weber had been central to the Court's analysis (it was not, because Weber was an exemption 5 case), Stalcup's inversion of the proposition is a textbook logical fallacy. More fundamentally, it is based on a faulty assumption. The argument presumes that individuals start with a minimal threshold of privacy and gain more through government action. Not only is there an absence of authority supporting that proposition, it ignores an individual's inherent privacy interest irrespective of any government intervention. See Carpenter, 470 F.3d at 438 (emphasizing the broad nature of the privacy interest covered by exemption 7(C)).
Finally, relying on the testimony of a single witness who felt "intimidated" from speaking out, Stalcup asserts that the eyewitnesses in this case actually wish to go public with their observations. He claims that the CIA has either destroyed their credibility or threatened retaliation if they come forward. He thus distinguishes this case from one such as Forest Service Employees for Environmental Ethics v. U.S. Forest Service, 524 F.3d 1021 (9th Cir. 2008), in which the witnesses' decision to remain silent for years indicated a desire to remain private.
Although the district court is required to make inferences in favor of the non-moving party at summary judgment, it is only required to do so if the inference asserted is reasonable. Mulero-Rodriguez v. Ponte, Inc., 98 F.3d 670, 672 (1st Cir.1996). The inference here requires a leap that is simply not justified by the evidence in this record. Even crediting the testimony regarding intimidation, that provides no basis to conclude that the government also "intimidated" 243 other eyewitnesses or, more benignly, that those witnesses had a desire to discuss the case. Although this single witness may wish to speak about the crash, as far as we know, the remainder may prefer to remain private.
Given that some privacy interest is at stake, a significant public interest must be present to nevertheless warrant disclosure of the witnesses' names. Favish, 541 U.S. at 172, 124 S.Ct. 1570. Here, again, Stalcup relies on the perceived cover-up as the core of the relevant public interest. The problem for Stalcup is that he fails to show how providing him the names — thus permitting him to further interview the witnesses — would yield any new information. That gap prevents us from concluding that release would further his purported public interest. Thus, given the presence of a privacy interest, and the complete absence of any public benefit, the balance between the two unquestionably banks against release. The district court correctly reached this conclusion.
C. A Final Note: The CIA's Search for the Records
We can quickly dispose of Stalcup's final contention. He asserts that the CIA failed to conduct a reasonable search after he sent his FOIA request, and should now be required to do more. He raises two points.
First, in a press release, the FBI referred to a photograph captured by an eyewitness and later analyzed by the National Imagery and Mapping Administration. The CIA, however, never produced this image. This failure, Stalcup says, is emblematic of the CIA's handling of his FOIA request.
Second, he posits that the CIA only searched one of the several departments, called "directorates," within the organization. He speculates that the work must have been done across the agency, particularly at the director level. In his view, this by itself renders the search inadequate.
Resolution of this claim turns on whether the agency made a good faith, reasonable effort "using methods which can be reasonably expected to produce the information requested." Oglesby v. U.S. Dep't of the Army, 920 F.2d 57, 68 (D.C.Cir. 1990). After an agency shows that it acted accordingly, which is generally accomplished through an affidavit, a rebuttable presumption that the agency acted in good faith emerges. Maynard, 986 F.2d at 560.
Robert Roland, the Information Review Officer of the CIA's Directorate of Intelligence, provided extensive detail on how the agency conducted its search. He also cogently explained why the CIA believed that a lone department, the directorate of intelligence, would house the responsive records. His declaration provided a reasonable explanation for the agency's process and, at a bare minimum, created a presumption that the CIA acted in good faith.
Stalcup's attempt to rebut that presumption goes nowhere. The absence of the single photograph (one, it should be noted, that was analyzed by an agency within the Department of Defense and not the CIA) does not warrant reversal. The omission of a single document in this case does not negate what is otherwise a reasonable inquiry. See Iturralde v. Comptroller of the Currency, 315 F.3d 311, 315 (D.C.Cir.2003). Consistent with its approach to the other issues in this case, the CIA handled the search in the required manner.
IV.
Finding the district court's conclusion to be fully supported, we affirm its order granting the CIA's motion for summary judgment.
