Jacqueline Stevens v. United States Department of S

20 F.4th 337

• Court: Court of Appeals for the Seventh Circuit
• Decided: December 9, 2021
• Docket: 20-3504
• Status: Published

Opinion

In the

United States Court of Appeals For the Seventh Circuit ____________________ No. 20‐3504 JACQUELINE STEVENS, Plaintiff‐Appellant, v.

UNITED STATES DEPARTMENT OF STATE, Defendant‐Appellee. ____________________

Appeal from the United States District Court for the Northern District of Illinois, Eastern Division. No. 17 C 2494 — John Z. Lee, Judge. ____________________

ARGUED OCTOBER 1, 2021 — DECIDED DECEMBER 9, 2021 ____________________

Before EASTERBROOK, MANION, and WOOD, Circuit Judges. WOOD, Circuit Judge. Jacqueline Stevens, a Professor of Po‐ litical Science at Northwestern University, is conducting re‐ search into the relations between the foreign campuses of American universities, the federal government, and private‐ sector entities. As part of that research, Professor Stevens sub‐ mitted three Freedom of Information Act requests to the De‐ partment of State. A lawsuit followed, terminating in entry of summary judgment for the Department. On appeal, Professor 2 No. 20‐3504

Stevens contends that summary judgment was improper both because the Department’s searches were inadequate and be‐ cause its withholdings were unwarranted. We are not per‐ suaded by either point, and so we affirm. I The events giving rise to this case began in February 2015, when Professor Stevens filed three FOIA requests with the Department. Request 3180 (the Northwestern request) sought all materials from the Department’s headquarters and the Qa‐ tar U.S. consulate referring to “Northwestern University’s Qatar campus.” Request 3181 (the Campuses request) sought policy and planning materials relating to the establishment of “U.S. university campuses in Qatar, Abu Dhabi, South Korea, China, and Singapore.” Finally, Request 3575 (the USAID/MEPI request) sought documents sent to or from the U.S. Agency for International Development (USAID) and documents produced, received, or maintained by the Middle East Partnership Initiative (MEPI) relating to “U.S. Govern‐ ment funds transferred to the Independent Center of Journal‐ ists”; “Northwestern University and its components, includ‐ ing the Medill School of Journalism”; and “the Center of Jour‐ nalism Excellence.” Dissatisfied with the pace and quantity of the Depart‐ ment’s productions, Professor Stevens filed suit in March 2017. Years of negotiations and motions practice followed. Over that time, the Department provided Professor Stevens with 128 complete records and 350 partial records responsive to the Northwestern request, 29 complete records and two partial records responsive to the USAID/MEPI request, and no records responsive to the Campuses request. It also No. 20‐3504 3

withheld 22 records responsive to the Northwestern request and two records responsive to the USAID/MEPI request. In May 2019, the Department moved for summary judg‐ ment. In support of the motion, it submitted a 35‐page decla‐ ration describing its search processes. Eric Stein, the Director of the Department’s Office of Information Programs and Ser‐ vices, prepared the declaration. The Department also submit‐ ted a Vaughn index describing each withheld document and the grounds for withholding it. See Vaughn v. Rosen, 484 F.2d 820, 826–28 (D.C. Cir. 1973). For the most part, the district court granted the summary‐ judgment motion, but it ordered the Department to conduct follow‐up Northwestern searches in the records of two of its subdivisions. (The Department had searched those records for “Northwestern University” rather than the broader “Northwestern.”) The Department complied, producing two additional records, and the district court then entered final judgment. This appeal followed. II Before turning to the FOIA standards and their applica‐ tion, we must address a docket‐management dispute. In Au‐ gust 2017, the Department approached Professor Stevens with a proposal. The Department’s preliminary searches had turned up a very large number of documents it deemed “po‐ tentially responsive” to the Northwestern request. It asked Professor Stevens to provide a set of keywords that could be used to streamline productions from those documents. She agreed and furnished the Department with a list of 20 or so terms. In so doing, Professor Stevens made clear, she was as‐ sisting with “prioritization” but was not waiving her right to 4 No. 20‐3504

an adequate production of documents responsive to the orig‐ inal Northwestern request. In its September 2017 status re‐ port, the Department memorialized its understanding of the deal. If Professor Stevens were unsatisfied with the adequacy of the filtered production, then the parties would discuss how to evaluate any remaining documents. Professor Stevens voiced no objection to the status report’s account or proposed method of proceeding. The Department moved forward with the Northwestern production, employing the agreed‐upon keywords to filter the already‐identified universe of potentially responsive doc‐ uments. It informed Professor Stevens that it had finished its resulting bulk productions in February 2018. Professor Ste‐ vens expressed no dissatisfaction with its reliance on the key‐ word list at the time. The Department completed some small follow‐up productions in early May. Again, Professor Stevens voiced no concerns about its use of the keywords. Nor did she say anything a few weeks later, when the Department filed a Vaughn index and summary‐judgment motion relying on the keyword agreement. Only in September 2018—a year after the status report and months after the Department’s produc‐ tions were complete—did Professor Stevens first object. She asserted that the keywords were meant only to help the De‐ partment prioritize its work, not to limit its ultimate obliga‐ tions. When she renewed that argument at summary judg‐ ment, the district court rejected it, holding that her objections to the Department’s use of the list had come too late. As we often have said, district courts have “considerable discretion to manage their dockets and to require compliance with deadlines.” Keeton v. Morningstar, Inc., 667 F.3d 877, 884 (7th Cir. 2012). It is true that in this case the parties never set No. 20‐3504 5

an explicit deadline for Professor Stevens to object to the ade‐ quacy of the filtered results. But that fact did not strip the dis‐ trict court of all power to manage the case; to the contrary, it remained entitled to take reasonable steps to keep the case moving forward. And given the timeline just described, the court was well within its rights to hold that Professor Ste‐ vens’s objections came too late. Accordingly, in assessing the adequacy of the Department’s Northwestern search, we do so in light of Stevens’s acquiescence to the Department’s reliance on the keyword list. It is worth noting, in this connection, that nothing we de‐ cide today precludes Professor Stevens from filing a new FOIA request seeking documents not captured by the keyword list. Agreements constraining the universe of documents to be searched in response to one request do not collaterally estop or otherwise bar requesters from filing future requests outside the scope of the earlier search. (Nonetheless, the loser in one FOIA suit cannot simply file the identical request again and start over.) Whether Professor Stevens chooses to take further permissible action is up to her. III We are now ready to discuss the merits of the appeal. We acknowledge at the outset that we have not been consistent in the standard of review we have applied to grants of summary judgment in FOIA cases. Higgs v. U.S. Park Police, 933 F.3d 897, 903 (7th Cir. 2019) (collecting cases). But our recent practice has been to distinguish between two parts of the inquiry, and to apply a different standard to each one. Henson v. Dep’t of Health & Hum. Servs., 892 F.3d 868, 875–76 (7th Cir. 2018). The first is whether the district court properly granted summary judgment. For that, we take a de novo look at the decision and 6 No. 20‐3504

ask whether there is any genuine issue of material fact about the adequacy of the agency’s search. Rubman v. U.S. Citizen‐ ship & Immigr. Servs., 800 F.3d 381, 387 (7th Cir. 2015). But when we turn to the propriety of withholdings, a more com‐ plex standard is necessary. We begin by considering de novo whether, as a threshold matter, the district court had an ade‐ quate factual basis for its decisions. Id. at 388. This step most often turns on the specificity of the agency’s affidavit, its Vaughn index, and whether the district court conducted an in camera review of the contested materials. Id. Our purpose at this stage is to decide whether the district court had enough in front of it to make a legally sound decision about whether an exemption applies. If it did, we move to the second step, reviewing for clear error its conclusion that the exemption did apply. Id. The thinking is that the district court is best situated to conduct the comprehensive, record‐by‐record review that FOIA withholdings may require. Becker v. IRS, 34 F.3d 398, 402 n.11 (7th Cir. 1994). A We begin with the adequacy of the Department’s searches. A search is adequate if it is the result of “a good faith effort” and is also “reasonable in light of the request.” Rubman, 800 F.3d at 387. “Good faith is presumed, and it can be bolstered by evidence of the agency’s efforts to satisfy the request”; rea‐ sonableness “is a flexible and context‐dependent standard.” Id. Professor Stevens takes issue with the presumption of good faith that agencies enjoy in FOIA litigation. But that rule is well settled. See, e.g., Henson, 892 F.3d at 875; Rubman, 800 F.3d at 387. And there are good reasons for this approach. It essentially requires us to credit the agency’s sworn statements No. 20‐3504 7

in the absence of evidence to the contrary. Administrative reg‐ ularity is the baseline assumption throughout our law, FOIA included. Here, the Department has documented its exten‐ sive, good‐faith efforts to satisfy the three requests in a thor‐ ough declaration prepared by its chief FOIA officer. And it has produced hundreds of documents over the course of four years of negotiation and engagement with Professor Stevens. Nothing in this history is incompatible with good faith. In assessing the reasonableness of the search, we look first to the processes described in Director Stein’s declaration. An agency is entitled to support its searches through affidavits, so long as they are “non‐conclusory” and “reasonably de‐ tailed.” Henson, 892 F.3d at 875. An affidavit is sufficiently de‐ tailed if it “set[s] forth the search terms used” and “the kind of search[es] performed,” and “aver[s] that all files likely to contain responsive documents were searched.” Id. The Stein declaration satisfies each of these criteria. Professor Stevens, however, sees things differently. She first contends that Director Stein’s declaration lacks crucial details, preventing us from evaluating whether the Depart‐ ment searched all databases likely to contain responsive doc‐ uments. Her main points of concern relate to the records‐man‐ agement system at the U.S. Embassy in Doha, the Depart‐ ment’s process for choosing databases to search, and the De‐ partment’s account of what she terms its “general file sys‐ tem.” Professor Stevens’s first problem is that she waived this argument by failing to raise it below. It should go without saying that “arguments not raised to the district court are waived on appeal.” Puffer v. Allstate Ins. Co., 675 F.3d 709, 718 (7th Cir. 2012). 8 No. 20‐3504

In any case, we do not share Professor Stevens’s concerns about the declaration. The Department’s general file system, according to the declaration, is distributed across “[e]ach of‐ fice within the Department, as well as each Foreign Service post and mission.” FOIA does not compel large, complex agencies such as the Department of State to maintain a spe‐ cific sort of centralized file system. As for the places searched at the Doha Embassy, the declaration indicates that the De‐ partment searched shared drives and print servers, the email records of employees in the three sections likely to have worked on matters responsive to the requests, and the ambas‐ sador’s post‐2015 email records. That is sufficient detail for our review. And the declaration explains that the Depart‐ ment’s officers made choices about what to search based on their “familiarity with the holdings of the Department’s rec‐ ords systems, applicable records disposition schedules, and the substantive and functional mandates of numerous De‐ partment offices and Foreign Service posts and missions.” Professor Stevens aptly notes that there is considerable discre‐ tion in this system. But to be fair to the Department, its records are spread across offices in over 200 countries; some discre‐ tion is inevitable. In our assessment, the Department’s ac‐ count of its methods provides sufficient detail for us to say that its methods were reasonable. Professor Stevens’s second complaint is that there are nu‐ merous locations, which she identifies, that she contends should have been searched but were not. But her list is a mix of locations that were in fact searched (e.g., “classified emails”), that no longer exist (e.g., the ambassador’s pre‐2015 records), or that could reasonably have been deemed unlikely to contain responsive emails (e.g., foreign national employees’ emails). None of the criticized omissions was unreasonable. No. 20‐3504 9

Before leaving this point, however, we must comment on one aspect of the district court’s reasoning. It struck Professor Stevens’s factual contentions in her complaint because it deemed them not to be based on her personal knowledge. See Fed. R. Civ. P. 56(c)(4). We have concerns about that reason‐ ing. Professor Stevens was asserting a predictive fact—that the locations in question were likely to contain responsive ma‐ terials. Rule 56 is concerned with facts “that would be admis‐ sible in evidence,” which we take to be of a different sort. A FOIA claimant is entitled to argue that a search was unrea‐ sonable because a likely storehouse was not searched. District courts are free to disagree with that prediction, but they should be cautious about treating Rule 56(c)(4) as a bar to such a prediction when the claimant has relied on accurate, albeit secondhand, information about the responsibilities of specific departments or officials. B If material is responsive, an agency must release it unless the agency can carry its burden to establish that the material falls under one of the nine FOIA exemptions. 5 U.S.C. § 552(b). Stevens challenges the Department’s reliance on three exemp‐ tions. The first of these, the national‐security exemption, applies to materials “specifically authorized under criteria estab‐ lished by an Executive order to be kept secret in the interest of national defense or foreign policy and [that] are in fact properly classified pursuant to such Executive order.” Id. § 552(b)(1). Here, the Department withheld material classified pursuant to Executive Order 13,526, which establishes the rules for identifying and handling classified material throughout the executive branch. Professor Stevens launches 10 No. 20‐3504

a facial assault on Order 13,526, asserting, among other things, that it “leads to the abdication of the role of the Judici‐ ary.” But it does no such thing. Using such sources as Vaughn indices, sworn affidavits, and in camera review, courts are able to assess national‐security withholdings made pursuant to Order 13,526, as this case illustrates. Professor Stevens’s more specific challenges to the Depart‐ ment’s national‐security withholdings fare no better. Docu‐ ment 27, for instance, would reveal confidential‐source infor‐ mation and risk the harassment of an education official. Doc‐ ument 28 would reveal information that a Qatari official shared in confidence. Document 34 would risk confidential‐ source information and harm to U.S.–Qatari relations. And so on. These and the other national‐security grounds for the challenged withholdings are articulated in convincing detail in the Vaughn index. Professor Stevens’s arguments to the contrary just reiterate her general concerns about document classification—a topic well beyond any particular FOIA re‐ quest. The second exemption Professor Stevens contends was misapplied covers materials “specifically exempted from dis‐ closure by statute.” 5 U.S.C. § 552(b)(3). Material “pertaining to the issuance or refusal of visas” falls under this heading. 8 U.S.C. § 1202(f). Professor Stevens contests the Department’s reliance on these two statutes to support its withholding of three documents, arguing that records withheld under sec‐ tions 552(b)(3) and 1202(f) must pertain to a “past or pending request” and that the three documents in question do not. But her only authority for her presumed rule is an unpublished district‐court opinion. See Immigr. Just. Clinic v. U.S. Dep’t of State, 2012 WL 5177410, at *1–2 (S.D.N.Y. Oct. 18, 2012). And No. 20‐3504 11

even if we saw merit in reading “pertaining to” so narrowly, it would make no difference here. The document in dispute in Immigrant Justice Clinic concerned a person’s “activities while in immigration detention.” Id. at *1. Here, in contrast, two of the three withheld documents are emails titled “Help with U.S. Visa,” while the third is a letter “pertain[ing] directly to the issuance or refusal of a visa.” They would thus satisfy even the requirement that Professor Stevens urges us to adopt (an invitation, we note, that we are declining). We turn next to the exemption for “inter‐agency or intra‐ agency memorandums or letters that 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). We have understood this ex‐ emption to apply to documents that would be privileged in the government’s litigation against a private party. Enviro Tech Int’l, Inc. v. EPA, 371 F.3d 370, 374 (7th Cir. 2004). Here, the Department has invoked the deliberative‐process privi‐ lege to withhold cables and emails containing preliminary agendas for official visits, memo drafts, and the like. Professor Stevens does not specifically contest the deliberative nature of any of these documents. Instead, she makes two general ob‐ jections. First, she asserts that cables are never deliberative be‐ cause they always “constitute final agency action.” This is in‐ correct, as is evident from some of the cables described in the Vaughn index. Second, she argues that a document is only pre‐ decisional if it pertains “to the formation or adoption of an official agency policy” or “adjudication.” But the privilege is not so limited. A document is predecisional if it is generated before an agency’s final decision on a matter, whether that matter is official or not. Cf. U.S. Fish & Wildlife Serv. v. Sierra Club, Inc., 141 S. Ct. 777, 786 (2021). 12 No. 20‐3504

Finally, Professor Stevens suggests that some of the with‐ held documents could instead have been produced in part. But she provides no specifics. And our independent review of the Vaughn index, coupled with the Department’s production of portions of more than 350 segregable documents, leave us confident that the Department’s efforts to identify such docu‐ ments were sufficient. IV We AFFIRM the judgment of the district court.