Jeremy Graber v. Michael Boresky

59 F.4th 603

• Court: Court of Appeals for the Third Circuit
• Decided: February 10, 2023
• Docket: 21-1407
• Status: Published

Opinion

PRECEDENTIAL

UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT ______________

No. 21-1407 ______________

JEREMY GRABER,

v.

POLICE OFFICER JOHN DOE II, Badge No. in his individual and official capacity as an officer for the city of Philadelphia Police Department; POLICE OFFICER JOHN DOE III, Badge No. in his individual and official capacity as an officer for the city of Philadelphia Police Department; POLICE OFFICER JOHN DOE IV, Badge No. in his individual and official capacity as an officer for the city of Philadelphia Police Department; SPECIAL AGENT MICHAEL BORESKY, in his individual and official capacity as a Special Agent for the U.S. Secret Services; POLICE INSPECTOR JOEL DALES, in his individual and official capacity as an Inspector for the city of Philadelphia

Special Agent Michael Boresky, Appellant ______________

On Appeal from the United States District Court for the Eastern District of Pennsylvania (D.C. No. 2-18-cv-03168) U.S. District Judge: Honorable Cynthia M. Rufe ______________

Argued October 4, 2022 ______________

Before: HARDIMAN, SHWARTZ, and NYGAARD, Circuit Judges.

(Filed: February 10, 2023) ______________

Brian M. Boynton Joseph F. Busa Jaynie Lilley [ARGUED] United States Department of Justice Civil Division 950 Pennsylvania Avenue, N.W. Washington, DC 20530

Paul E. Werner United States Department of Justice Torts Branch, Civil Division P.O. Box 7146, Ben Franklin Station Washington, DC 20044

Counsel for Appellant

Paul J. Hetznecker [ARGUED] 1420 Walnut Street Suite 911 Philadelphia, PA 19102

2 Counsel for Appellee ______________

OPINION OF THE COURT ______________

SHWARTZ, Circuit Judge.

In Bivens v. Six Unknown Named Agents of the Federal Bureau of Narcotics, 403 U.S. 388 (1971), the Supreme Court held that a cause of action existed against federal agents who violated the Fourth Amendment. Relying on Bivens, Plaintiff Jeremy Graber sued Defendant Michael Boresky, a Special Agent for the United States Secret Service, asserting that Boresky violated his Fourth Amendment rights by arresting, detaining, and charging him with a crime without probable cause. In an order denying a motion to dismiss, the District Court held that a Bivens claim could be brought against Boresky. Thereafter, the Court dismissed Boresky’s motion for summary judgment without prejudice based upon qualified immunity because it found that discovery was needed to determine whether Boresky was entitled to qualified immunity. At oral argument before our Court, Boresky stated that he is not challenging the qualified immunity ruling but argued that we should review the District Court’s Bivens ruling. Because the Bivens ruling is not a final decision and is not appealable under the collateral order doctrine, we lack jurisdiction to consider that interlocutory ruling and we must dismiss this appeal.

3 I

A

In 2016, Philadelphia hosted the Democratic National Convention (the “Convention”). The Department of Homeland Security (“DHS”) designated the event as a National Special Security Event (“NSSE”). Once an event is designated an NSSE, federal agencies coordinate operational security with state and local law enforcement. Relevant here, the Secret Service “coordinate[d] the development and implementation of the overall operational security plan.” App. 53.

In the lead-up to the Convention, the Secret Service announced that access to certain areas around the Convention would be restricted (the “Restricted Area”). 1 The Restricted 0F

Area was surrounded by an eight-foot fence.

On the evening of July 27, 2016, Plaintiff joined political protests outside the Restricted Area. 2 Protestors 1F

breached the gated perimeter around the Restricted Area. The Philadelphia Police Department (“PPD”) apprehended those within the Restricted Area. Plaintiff was one of seven individuals taken into custody. PPD did not prepare any arrest paperwork for Plaintiff.

1 The Restricted Area was Broad Street from 7th Street to 20th Street, the cross-streets between Packer Avenue to I- 95, and the entirety of FDR Park. 2 Plaintiff, a paramedic, was also there to provide emergency medical aid to protestors.

4 Thereafter, the Assistant to the Special-Agent-in- Charge of the Secret Service in Philadelphia informed Boresky of the arrests and told him that the arrestees were to be charged with violating 18 U.S.C. § 1752, 3 and that Boresky would serve 2F

as the affiant for the criminal complaint. 4 The next morning, 3F

Special Agent Aaron McCaa e-mailed Boresky a synopsis of the events leading to the arrests as well as photographs of the fence and evidence seized from the arrestees.

Boresky appeared before a Magistrate Judge and signed an affidavit identifying Plaintiff as one of the seven individuals arrested inside the Restricted Area. Boresky attested that the contents of the affidavit were based upon his “personal knowledge, experience and training,” “information developed during the course of this investigation,” and “information . . . imparted to [him] by other law enforcement officers.” App. 77. Boresky admits that he was not present at the arrest and did not write the affidavit but reviewed it for accuracy based upon the information in McCaa’s synopsis. Boresky did not view any video evidence before swearing out the affidavit.

Plaintiff was held overnight at the Federal Detention Center. Plaintiff’s counsel thereafter provided Fox 29 News video clips to the Government confirming that Plaintiff never

3 18 U.S.C. § 1752 prohibits persons and groups from entering a restricted area where a Secret Service protectee is or will be visiting or an area restricted in conjunction with an event designated as a special event of national significance. 4 The night before Plaintiff’s arrest, Boresky served as the affiant for criminal complaints against four other individuals arrested for unlawfully entering the Restricted Area.

5 passed through the fence. Plaintiff was released and the charges were dismissed.

B

Citing Bivens, Plaintiff sued Boresky for false arrest, unlawful detention, and false charges. 5 Boresky moved to 4F

dismiss, arguing that Plaintiff could not pursue his Fourth Amendment claim against him under Bivens. Graber v. Dales, No. 18-CV-3168, 2019 WL 4805241, at *1-2 (E.D. Pa. Sept. 30, 2019) (“Graber I”). 6 The District Court employed the 5F

Supreme Court’s two-step framework set forth in Ziglar v. Abbasi, 137 S. Ct. 1843 (2017), which requires a court to first consider whether a plaintiff’s claim presents a context in which the Supreme Court had not previously recognized a Bivens claim and, if so, whether special factors counsel against extending Bivens to permit such a claim. Graber I, 2019 WL 4805241, at *3 (citing Ziglar, 137 S. Ct. at 1857-60). The Court denied Boresky’s motion to dismiss under Federal Rule of Civil Procedure 12(b)(6), concluding that Plaintiff’s Fourth Amendment claim, predicated on the assertion that he was arrested and charged without probable cause, did not present a “new Bivens context,” id. at *3, and thus presented a basis for

5 Plaintiff also asserted First Amendment and conspiracy claims against all officers. 6 The District Court dismissed the claims against Boresky in his official capacity for lack of subject-matter jurisdiction because the United States had not waived sovereign immunity for constitutional tort claims as well as Plaintiff’s First Amendment and civil conspiracy claims against Boresky in his individual capacity for failure to state a claim. Graber I, 2019 WL 4805241, at *2, 7-9.

6 relief. The Court further held that, even if Plaintiff’s claim arose in a new context, special factors did not counsel against extending Bivens to permit Plaintiff’s claim to proceed. Id. at *4-5.

The case then proceeded to discovery. Amid discovery disputes between the parties, Boresky filed a motion for summary judgment based on qualified immunity and asked to stay discovery pending resolution of that motion. Graber v. Dales, 511 F. Supp. 3d 594, 595 (E.D. Pa. 2021) (“Graber II”). Plaintiff responded by filing a declaration under Federal Rule of Civil Procedure 56(d), asserting that he needed discovery to respond to Boresky’s summary judgment motion.

The District Court concluded that Boresky’s entitlement to qualified immunity hinged on whether it was “objectively reasonable” for him to believe that there was probable cause to detain and charge Plaintiff, and this required consideration of “evidence surrounding the statements and communication upon which Defendant Boresky relied.” Id. at 599. Because Plaintiff had no opportunity to conduct any discovery, the Court concluded that it would be “wholly inequitable” to permit Boresky to rely upon affidavits and communications to which Plaintiff had no access and denied the qualified immunity motion without prejudice to permit discovery. Id. at 600.

Boresky appeals, waiving his challenge to the qualified immunity ruling and asking us to review whether the District Court erred in holding Plaintiff could bring a Bivens claim. Oral Argument at 5:52-6:02, Graber v. Boresky (Oct. 4, 2022) (No. 21-1407), https://www2.ca3.uscourts.gov/oralargument/audio/21-

7 1407Graberv.SpecialAgentMichaelBoresky.mp3.

II 7 6F

At the outset, we must ensure we have jurisdiction over this appeal. While we would have had jurisdiction to review an interlocutory appeal of the District Court’s qualified immunity order, 8 Mack v. Yost, 968 F.3d 311, 318 (3d Cir. 7F

7 The District Court had subject matter jurisdiction pursuant to 28 U.S.C. § 1331. 8 Where a district court defers ruling on qualified immunity to permit further fact discovery, as is the case here, “implicit in that ruling” is the legal conclusion that the plaintiff adequately pled a violation of clearly established law, and thus the order is immediately appealable. Oliver v. Roquet, 858 F.3d 180, 189 (3d Cir. 2017) (holding appellate jurisdiction existed despite factual component of the court’s qualified immunity ruling); see also In re Montgomery Cnty., 215 F.3d 367, 370 (3d Cir. 2000) (concluding that an “implicit denial” of immunity claims is “sufficient to confer appellate jurisdiction”). The District Court’s order dismissing the motion seeking summary judgment based on qualified immunity to allow for discovery contains the implicit legal conclusion that Bivens is available in this context. See Vanderklok v. United States, 868 F.3d 189, 197 (3d Cir. 2017) (characterizing Bivens remedy as a “threshold question of law” that “is directly implicated by the defense of qualified immunity”); see also Wilkie v. Robbins, 551 U.S. 537, 549 n.4 (2007) (recognizing appellate courts have jurisdiction over

8 2020), Boresky no longer challenges the qualified immunity ruling. As a result, we must determine whether we can review the Court’s Rule 12(b)(6) Bivens ruling untethered from a challenge to a qualified immunity ruling. Boresky contends that we have jurisdiction under the collateral order doctrine.

We have jurisdiction over “appeals from all final decisions of the district courts.” 28 U.S.C. § 1291. There are, however, “a small class of rulings, not concluding the litigation, but conclusively resolving claims of right separable from, and collateral to, rights asserted in the action.” Will v. Hallock, 546 U.S. 345, 349 (2006) (quotation marks and citation omitted). Such interlocutory orders are appealable under the collateral order doctrine if they: (1) “conclusively determine the disputed question”; (2) “resolve an important issue completely separate from the merits of the action”; and (3) are “effectively unreviewable on appeal from a final judgment.” Id. (citation omitted). The Supreme Court has described these elements as “stringent” to ensure that the collateral order doctrine does not “overpower the substantial finality interests § 1291 is meant to further.” Id. at 349-50; see also Digital Equip. Corp. v. Desktop Direct, Inc., 511 U.S. 863, 868 (1994) (describing the collateral order doctrine as a “narrow exception” to the final order requirement). Orders falling into this narrow group “are sufficiently important and collateral to the merits [such] that they should nonetheless be treated as final.” Will, 546 U.S. at 347 (quotation marks and citation omitted).

interlocutory appeals challenging a Bivens ruling in a qualified immunity appeal).

9 A Bivens ruling does not fall within this small group of orders that require interlocutory review under the collateral order doctrine: a Bivens ruling can be effectively reviewed after final judgment because, unlike various immunity doctrines, a Bivens ruling is not meant to protect a defendant from facing trial. The Supreme Court has identified several types of orders that are entered to protect a defendant from facing trial, and each would be effectively unreviewable if considered after final judgment is entered: orders denying absolute immunity, orders denying qualified immunity, orders denying Eleventh Amendment immunity, and adverse double jeopardy rulings. Id. at 350 (collecting cases). An order denying immunity (or double jeopardy protection) from suit cannot be “reviewed ‘effectively’ after a conventional final judgment,” id. at 351, because the suit has already occurred by the time the appeal is reviewed, and thus the purpose of the immunity (or double jeopardy protection) is defeated, see, e.g., Swint v. Chambers Cnty. Comm’n, 514 U.S. 35, 42 (1995) (“[A]n official’s qualified immunity is an immunity from suit rather than a mere defense to liability; and like an absolute immunity, it is effectively lost if a case is erroneously permitted to go to trial.” (quotation marks, citation, and emphasis omitted)).9 Moreover, the immunity doctrines in

9 The dissent is correct that the collateral order doctrine allows review of more than orders addressing assertions of immunity, and each of the additional examples the dissent cites is similarly reviewable under the collateral order doctrine because each involves issues that are “in danger of becoming moot,” and thus unreviewable, following a final judgment. United States v. Mitchell, 652 F.3d 387, 397 (3d Cir. 2011) (en banc) (citation omitted) (order prohibiting pretrial collection of defendant’s DNA sample); see also United States v. Bellille,

10 particular are meant to allow government officers to avoid “the burdens of litigation” and to carry out their duties without the threat of a “full trial . . . whenever they acted reasonably in the face of law that is not ‘clearly established.’” Will, 546 U.S. at 352. Thus, immediate review of orders denying immunity furthers the goal of the immunity doctrines, which is to avoid trial, while also “honoring the separation of powers,” “preserving the efficiency of government,” and encouraging “the initiative of its officials.” Id.

Bivens, however, is not an immunity doctrine. See, e.g., F.D.I.C. v. Meyer, 510 U.S. 471, 483-84 (1994) (observing that “whether there has been a waiver of sovereign immunity” is “analytically distinct” from whether substantive law upon which plaintiff relies provides a basis for relief). Rather, it is a judicially created cause of action that allows a plaintiff to sue

962 F.3d 962, 737-38 (3d Cir. 2020) (order denying motion to withdraw as counsel); Doe v. Coll. of N.J., 997 F.3d 489, 494 (3d Cir. 2021) (order denying motion to proceed anonymously); United States v. Wecht, 537 F.3d 222, 228-29 (3d Cir. 2008) (order denying the public’s right of access to a criminal trial). In other words, each of these issues, unlike a Bivens ruling, involves a right that would be “irretrievably lost” absent an immediate appeal. Praxis Props., Inc. v. Colonial Sav. Bank, S.L.A., 947 F.2d 49, 60 (3d Cir. 1991), as amended on denial of reh’g (Nov. 13, 1991) (order denying statutory right to a 90-day stay). The issue of whether there is a cognizable Bivens claim does not become moot following a final judgment. Instead, a litigant can continue to assert a defense under Bivens and seek review of that defense after the entry of a final judgment because his right not to be held liable is not “irretrievably lost” absent an interlocutory appeal.

11 a federal officer for damages for constitutional violations. Bivens actions are very limited, and new ones cannot be created where “there is any rational reason (even one) to think that Congress is better suited to ‘weigh the costs and benefits of allowing a damages action to proceed.’” Egbert v. Boule, 142 S. Ct. 1793, 1805 (2022) (emphasis omitted) (quoting Ziglar, 137 S. Ct. at 1858). The Court’s focus in determining whether such a claim can be brought, therefore, is on whether courts should be in the business of creating avenues for liability, which is distinct from whether a defendant is immune from suit altogether.

The Supreme Court itself has recognized this difference and the impact it has on the ability to seek immediate review of a Bivens ruling. The Court stated that “if simply abbreviating litigation troublesome to Government employees were important enough for [collateral order] treatment, [then] collateral order appeal would be a matter of right whenever the Government lost a motion to dismiss under the Tort Claims Act, or a federal officer lost one on a Bivens action, or a state official was in that position in a case under 42 U.S.C. § 1983, or Ex parte Young.”10 Will, 546 U.S. at 353-54 (rejecting

10 The dissent characterizes the Supreme Court’s statement as dicta. Even if that label is accurate, we have held that statements made by the Supreme Court in “dicta are highly persuasive.” Galli v. N.J. Meadowlands Comm'n, 490 F.3d 265, 274 (3d Cir. 2007). We have observed that, because the “Supreme Court uses dicta to help control and influence the many issues it cannot decide because of its limited docket,” failing to follow those statements could “frustrate the evenhanded administration of justice by giving litigants an outcome other than the one the Supreme Court would be likely

12 application of the collateral order doctrine to the Federal Tort Claims Act’s judgment bar). In short, the Supreme Court has recognized that a Bivens ruling is different from an immunity ruling and is not eligible for interlocutory appeal under the collateral order doctrine. Id.11

Accordingly, Boresky’s assertion that there is no cause of action under Bivens is simply a defense to liability, which can be effectively reviewed after the entry of final judgment. Cf. Swint, 514 U.S. at 41-43 (holding that a denial of a motion for summary judgment on a Monell claim is not appealable under the collateral order doctrine because the defendant’s argument would amount to a “mere defense to liability” that could be “reviewed effectively on appeal from final judgment”). Unlike an immunity ruling, any error in a Bivens

to reach were the case heard there.” Official Comm. of Unsecured Creditors of Cybergenics Corp. v. Chinery, 330 F.3d 548, 561 (3d Cir. 2003) (quoting In re McDonald, 205 F.3d 606, 612-613 (3d Cir. 2000)). 11 The dissent’s observations about separation of powers are well-taken but they do not support creating an avenue for interlocutory review of an issue that goes directly to liability. The Supreme Court’s Bivens jurisprudence cautions courts not to create new causes of action, as that is the job of the legislature. See Egbert, 142 S. Ct. at 1802 (2022). However, whether a court has created a cause of action or, in the language of Bivens, a plaintiff presented a context in which such a Bivens claim had not been recognized, presents questions about whether the plaintiff has stated a claim upon which relief may be granted, which is the type of ruling regularly reviewed after the entry of final judgment.

13 ruling can be cured on appeal at the end of the case.12 Thus, an order denying a motion to dismiss or for summary judgment based upon Bivens, untethered to an order denying qualified immunity, is not appealable under the collateral order doctrine. 138F

A sister circuit court reached the same conclusion. In Himmelreich v. Federal Bureau of Prisons, 5 F.4th 653 (6th Cir. 2021), the Court of Appeals for the Sixth Circuit held that an appellate court lacks jurisdiction under the collateral order doctrine over a Bivens ruling absent an appealable qualified immunity order. Id. at 659. The court assumed without deciding that the Bivens order there, which permitted plaintiff to proceed on his First Amendment claim against a federal officer, was a conclusive ruling and that the order resolved an issue of separation of powers distinct from the plaintiff’s constitutional claim. Id. at 661. However, the court concluded that the issue could be adequately reviewed following final judgment. Id. at 662. Like us, the court also observed that Bivens provides a “remedy for unconstitutional conduct” but “does not grant defendants entitlement not to stand trial.” Id.

12 Because an order denying dismissal based upon Bivens fails the third factor from Will, we need not consider whether the first or second Will factors—whether the order “conclusively determine[d] the disputed question” or “resolve[d] an important issue completely separate from the merits of the action,” 546 U.S. at 349—are satisfied. 13 The Supreme Court counsels appellate courts to hesitate in enlarging the types of orders eligible for review under the collateral order doctrine that are not independently appealable nor certified for review under 28 U.S.C. § 1292(b). Swint, 514 U.S. at 47-48.

14 Because a Bivens ruling can be effectively reviewed after the entry of final judgment, it is not an order that falls within the small class of orders that are immediately appealable under the collateral order doctrine and, as a result, we lack appellate jurisdiction to review the District Court’s Bivens ruling.14

14 The availability of an alternative appellate mechanism pursuant to 28 U.S.C. § 1292(b) also counsels against the position taken by the Government and the dissent. For difficult questions of law “in exceptional cases,” parties may seek interlocutory review by a court of appeals. Milbert v. Bison Labs., Inc., 260 F.2d 431, 433 (3d Cir. 1958). A district court may grant a certificate of appealability under § 1292(b) when its order: “(1) involve[s] a ‘controlling question of law,’ (2) offer[s] ‘substantial ground for a difference of opinion’ as to its correctness, and (3) if appealed immediately ‘materially advance[s] the ultimate termination of the litigation.’” Katz v. Carte Blanche Corp., 496 F.2d 747, 754 (3d Cir. 1974) (en banc) (quoting § 1292(b)). Certification is available for purposes that address the Government’s concerns here: avoiding “a wasted[,] protracted trial” when “a pretrial order erroneously overrul[ed] a defense going to the right to maintain the action.” Id. (citing legislative history of § 1292(b)). One of § 1292(b)’s regular uses is to permit interlocutory appeal to decide whether a statute permits a private cause of action. See, e.g., Zeffiro v. First Pa. Banking & Tr. Co., 623 F.2d 290, 292 (3d Cir. 1980) (whether an injured investor has a federal cause of action under the Trust Indenture Act of 1939); Northstar Fin. Advisors, Inc. v. Schwab Invs., 615 F.3d 1106, 1114-15 (9th Cir. 2010) (whether there is a private right of action to enforce § 13(a) of the Investment Company Act of 1940); Love v. Delta Air

15 III

For the foregoing reasons, we will dismiss this appeal.

Lines, 310 F.3d 1347, 1350-51 (11th Cir. 2002) (whether the Air Carrier Access Act of 1986 implies a private right of action). That determination and the availability of a Bivens remedy present a question of law in the same category: can the plaintiff sue at all? The discretionary availability of § 1292(b)’s mechanism makes us hesitate to agree that an order allowing a Bivens claim to proceed is one within “that small class” of orders “too important to be denied review and too independent of the cause itself to require that appellate consideration be deferred until the whole case is adjudicated.” Cohen v. Beneficial Indus. Loan Corp., 337 U.S. 541, 546 (1949).

16 Jeremy Graber v. Special Agent Michael Boresky et al., No. 21-1407 ______________

HARDIMAN, Circuit Judge, dissenting.

This appeal is unusual. Special Agent Michael Boresky could have filed a meritorious interlocutory appeal of the District Court’s order dismissing without prejudice his motion for summary judgment on qualified immunity. Instead, Boresky asks us to be the first appellate court to hold that an order denying a motion for summary judgment that challenges the existence of a Bivens cause of action is appealable before a final judgment is entered. This gambit implicates two conflicting trends of Supreme Court jurisprudence: the Court’s careful policing of the expansion of the collateral order doctrine established in Cohen v. Beneficial Industrial Loan Corporation, 337 U.S. 541 (1949), and its repeated refusal to allow new constitutional tort actions against federal officers under Bivens v. Six Unknown Agents of Federal Bureau of Narcotics, 403 U.S. 388 (1971). It’s hard to predict how the Supreme Court would resolve this conflict. But because I think the Court will allow interlocutory appeals in cases like this one—where the constitutional separation of powers is imperiled—I respectfully dissent.

I

In Swint v. Chambers County Commission, 514 U.S. 35 (1995), the Supreme Court identified appealable collateral orders as those that (1) are “conclusive,” (2) “resolve important questions separate from the merits,” and (3) are “effectively unreviewable on appeal from the final judgment.” 514 U.S. at 42. The majority opinion holds that Boresky failed to satisfy the third criterion. In doing so, it focuses on four types of collateral orders the Supreme Court has recognized, observing that each rejects a defense the purpose of which is to avoid suit altogether. Maj. Op. 10–12. The majority then notes that Bivens is “not an immunity doctrine” but rather addresses “whether courts should be in the business of creating avenues for liability.” Id. at 11–12. I agree with those propositions.

But an immunity is neither sufficient nor necessary for an order denying a claim to be “effectively unreviewable on appeal.” That criterion is met if an order denies a potentially dispositive pretrial defense that implicates a sufficiently important public value. See Will v. Hallock, 546 U.S. 345, 352– 53 (2006). And we have recognized collateral orders that do not involve immunity defenses at all, much less immunity from suit. See, e.g., United States v. Mitchell, 652 F.3d 387, 398 (3d Cir. 2011) (en banc) (orders prohibiting pretrial collection of a criminal defendant’s DNA sample); United States v. Bellille, 962 F.3d 731, 737–38 (3d Cir. 2020) (orders denying motions to withdraw as counsel in criminal cases); Doe v. Coll. of N.J., 997 F.3d 489, 494 (3d Cir. 2021) (orders denying motions to proceed anonymously); Chehazeh v. Att’y Gen., 666 F.3d 118, 139 (3d Cir. 2012) (sua sponte BIA orders to reopen removal proceedings); United States v. Wecht, 537 F.3d 222, 229 (3d Cir. 2008) (orders denying the public’s right of access to a criminal trial); Praxis Properties, Inc. v. Colonial Sav. Bank, S.L.A., 947 F.2d 49, 61 (3d Cir. 1991), as amended on denial of reh’g (Nov. 13, 1991) (orders denying requests for a litigation stay under 12 U.S.C. § 1821(d)(12)). The Supreme Court has done likewise. See Eisen v. Carlisle & Jacquelin, 417 U.S. 156, 170–72 (1974) (orders allocating the costs of providing notice to class members); Stack v. Boyle, 342 U.S. 1, 6 (1951) (orders denying motions to reduce bail). And several

2 kinds of collateral orders involve interests that are less weighty than the constitutional value imperiled by the District Court’s Bivens authorization: the separation of powers. See, e.g., Cohen, 337 U.S. at 546–47 (orders rejecting the applicability of security laws enacted after the initiation of derivative shareholder suits); Praxis Properties, 947 F.2d at 61 (orders denying requests for a litigation stay).

The majority leans on the Supreme Court’s decision in Will v. Hallock, stating: “the Supreme Court has recognized that a Bivens ruling is different from an immunity ruling and is not eligible for interlocutory appeal under the collateral order doctrine.” Maj. Op. 13 (emphasis added). Will did not so hold. The issue there was whether a “refusal to apply the judgment bar of the Federal Tort Claims Act is open to collateral appeal.” 546 U.S. at 347. The Court held it was not. Id.

What’s more, Will characterized the interest supporting the FTCA’s judgment bar as “avoidance of litigation for its own sake.” Id. at 353. The Court contrasted this “mere avoidance” of trial generally with avoidance of a trial that would “imperil a substantial public interest”; the latter is what counts under Swint’s third criterion. Id. “[I]f,” the Court concluded, “simply abbreviating litigation troublesome to Government employees were important enough for Cohen treatment,” a 28 U.S.C. § 1291 appeal would lie whenever a federal officer lost a motion to dismiss “on a Bivens action”—or the Tort Claims Act, 42 U.S.C. § 1983, or Ex parte Young, 209 U.S. 123 (1908). Id. at 353–54.

The majority cites this sentence from Will as evidence of the Supreme Court’s “recogni[tion]” that an order authorizing a Bivens cause of action is ineligible for interlocutory appeal. Maj. Op. 13. I don’t put as much stock as my colleagues in the

3 Court’s drive-by dictum about Bivens, primarily because of several substantive points Will made. First, the Court noted that the FTCA’s judgment bar isn’t important enough to merit interlocutory appeal of orders denying its applicability because it resembles the defense of claim preclusion, which “has not been thought to protect values so great that only immediate appeal can effectively vindicate them.” Will, 546 U.S. at 355. Will also contrasted the judgment bar’s “essential procedural element”—the bar can be raised “only after a case under the Tort Claims Act has been resolved in the Government’s favor”—with a qualified immunity defense, which is “timely from the moment an official is served with a complaint.” Id. at 354. The defense that no Bivens cause of action lies is just like qualified immunity in this respect. Finally, Will acknowledged that “honoring the separation of powers” and “preserving the efficiency of government and the initiative of its officials” were “particular value[s] of a high order” sufficient to warrant § 1291 interlocutory review. Id. at 352. Those are precisely the values imperiled by erroneous Bivens authorizations.

II

I agree with my colleagues that we must police the parameters of the collateral order class “stringent[ly].” Maj. Op. 9 (quoting Will, 546 U.S. at 349). That class must remain of “modest scope.” Will, 546 U.S. at 350. Yet we do recognize new collateral orders. See, e.g., Bellille, 962 F.3d at 737–38; Doe, 997 F.3d at 494. Our task is to honor the collateral order doctrine’s “internal logic” and “strict[ly] appl[y]” the Cohen criteria restated in Swint. Ashcroft v. Iqbal, 556 U.S. 662, 672 (2009).

4 A

The majority concludes that we lack jurisdiction because the order in question resolves an issue that would not be “effectively unreviewable on appeal” after final judgment. I understand that criterion—Swift’s third—differently than my colleagues.

The touchstone for that criterion is the importance of the values imperiled by an erroneous ruling. See Will, 546 U.S. at 351–52 (“only some orders denying an asserted right to avoid the burdens of trial qualify” under Cohen, namely those involving interests judged sufficiently valuable); Sell v. United States, 539 U.S. 166, 177 (2003) (the “importance of the constitutional issue” can distinguish appealable from non- appealable collateral orders); Wecht, 537 F.3d at 229 (asking whether the “value” of immediate vindication is “significant enough to justify [interlocutory] review”); Lauro Lines s.r.l. v. Chasser, 490 U.S. 495, 503 (1989) (Scalia, J., concurring) (post-judgment vindication is “enough” when the interest in question is not “sufficiently important” to overcome the policies underlying the final judgment rule). So it’s not enough—though it is necessary—for Boresky to invoke an interest that will be “essentially destroyed” if its vindication awaits post-trial review. Lauro Lines, 490 U.S. at 499. He must also invoke a “particular value of a high order” or a “substantial public interest” to tip the scale. Will, 546 U.S. at 352–53. Whether delayed review would imperil such a value is the collateral order doctrine’s “decisive consideration.” Mohawk Industries, Inc. v. Carpenter, 558 U.S. 100, 107 (2009).

5 As those precedents suggest, not only those prerogatives flying under the banner of “immunity” can be collateral.1 First, not every denial of an immunity defense warrants interlocutory review. See, e.g., Van Cauwenberghe v. Biard, 486 U.S. 517, 524 (1988) (denial of a “claim of immunity from civil service of process” is not a collateral order); We, Inc. v. City of Philadelphia, 174 F.3d 322, 326 (3d Cir. 1999) (denial of “Noerr-Pennington immunity” is not a collateral order).

Second, not every collateral order denies an immunity claim. Cohen itself held that an order rejecting the applicability of a security law enacted after the initiation of a derivative shareholder suit was “final” under § 1291. 337 U.S. at 546–47. The Supreme Court has also recognized orders allocating the costs of providing notice to class members as collateral. Eisen, 417 U.S. at 170–72. And we have recognized collateral orders implicating other rights or interests that are not immunities. It’s true that whether a claimed entitlement is better characterized as an immunity from suit or defense against liability is a key consideration under the collateral order doctrine. Robinson v. Hartzell Propeller, Inc., 454 F.3d 163, 171 (3d Cir. 2006). But that’s because whereas a pure “right not to be tried” necessarily satisfies Swint’s third criterion, not every “right whose remedy requires the dismissal” of a claim before trial does. United States v. Hollywood Motor Car Co., 458 U.S. 263, 269 (1982).

The majority cites favorably the Sixth Circuit’s decision in Himmelreich v. Federal Bureau of Prisons, 5 F.4th 653 (6th Cir. 2021), which denied that an order allowing a Bivens action

1 Since there is sometimes “no obviously correct way” to characterize the value or interest at issue, it is not always clear whether we are dealing with an “immunity.” Lauro Lines, 490 U.S. at 500.

6 to proceed was collateral under § 1291. Respectfully, that decision also misunderstands Swint’s third criterion. Himmelreich assumed that the district court’s order conclusively determined an important issue separate from the merits. 5 F.4th at 661. The court then denied that such an order was “effectively unreviewable” on appeal from final judgment. Id. at 662. It reasoned: “[u]nlike qualified immunity, Bivens provides a plaintiff’s remedy for unconstitutional conduct. It does not grant defendants an entitlement not to stand trial.” Id. But Swint’s third criterion does not ask only whether a defendant has an entitlement not to stand trial. And Himmelreich, like the majority here, mistakenly concluded that Will foreclosed the argument that “the collateral order doctrine extends to standalone appeals of district court orders recognizing a Bivens remedy.” Id.

In short, Swint’s third criterion does not look to whether an immunity is asserted. It focuses on the importance of the values involved in the order under review.

I would hold that interlocutory appeals from orders denying motions for summary judgment that challenge the cognizability of a Bivens cause of action are “final” under § 1291 and the collateral order doctrine. Though we have rejected the application of the collateral order doctrine to non- final orders in “the vast majority of cases,” Robinson, 454 F.3d at 170, the order at issue here satisfies Swint’s criteria. The Supreme Court’s recent opinions delimiting the scope of Bivens underscore the importance of the values jeopardized when district courts wrongly allow such claims to proceed. And orders denying summary judgment motions arguing that no Bivens cause of action is cognizable conclusively determine

7 an important question of law distinct from a Bivens claim’s merits.

Orders like those just mentioned are effectively unreviewable on appeal because they imperil a “particular value of a high order” and “substantial public interest,” Will, 546 U.S. at 352–53: the Constitution’s separation of the legislative and judicial powers.

A cause of action is a “remedial mechanism.” Bivens, 403 U.S. at 397. It permits a plaintiff to “appropriately invoke the power of [a] court” to hear his suit and grant relief. Davis v. Passman, 442 U.S. 228, 240 n.18 (1979). The choice to allow a Bivens action to proceed against a federal officer requires consideration of “a number of economic and governmental concerns,” including the “time and administrative costs” run up by the discovery and trial process and the extent to which “monetary and other liabilities should be imposed upon” officers who violate the Constitution. Ziglar v. Abbasi, 137 S. Ct. 1843, 1856 (2017). Congress is “best positioned” to reach that sort of judgment. Hernandez v. Mesa, 140 S. Ct. 735, 742 (2020) (Hernandez II). So it’s a “significant step under separation-of-powers principles” for an Article III court to authorize a Bivens action. Ziglar, 137 S. Ct. at 1856. That’s why doing so is a “disfavored judicial activity.” Id. at 1857 (internal quotation marks and citation omitted).

The Supreme Court recently put the point more directly: “creating a cause of action is a legislative endeavor,” pure and simple. Egbert v. Boule, 142 S. Ct. 1793, 1802 (2022). The Judiciary’s power to authorize a Bivens cause of action at all is “uncertain”—so much so that a court should not extend Bivens

8 if there is any rational reason to think Congress better positioned to decide whether to create a cause of action. Id. at 1803. That hurdle is a high one. So too the cost of wrongly clearing it: Congress cannot undo judicially created constitutional remedies. Dongarra v. Smith, 27 F.4th 174, 181 (3d Cir. 2022).

The Court has called the separation of powers a “particular value of a high order” that satisfies Swint’s third criterion. Will, 546 U.S. at 352. In my view, “protect[ing] the constitutional command of separation of powers” against the “impermissible assertion” of authority by “the federal courts” is an imperative worthy of immediate enforcement. Helstoski v. Meanor, 442 U.S. 500, 505–06 (1979). We should show “special solicitude” toward “threatened breach[es]” of the “separation of powers.” Nixon v. Fitzgerald, 457 U.S. 731, 743 (1982).

It’s true that forcing a private litigant to shoulder the burden of a legally unwarranted trial is often consistent with the calculus underlying the final judgment rule. Cf. Robinson, 454 F.3d at 171–72. But the Bivens defendant always is a federal officer. Also, unlike the defenses invoked under res judicata and statutes of limitation, which protect only the “interest in not being held ultimately liable” on some claim, Bell Atl. v. Penn. Pub. Util. Comm’n, 273 F.3d 337, 344 (3d Cir. 2001), the defendant sued under Bivens asserts that the court cannot entertain the claim in the first place. See Elhady v. Unidentified CBP Agents, 18 F.4th 880, 884 (6th Cir. 2021) (“Plaintiffs . . . often have no cause of action unless we extend Bivens. And if there is no cause of action, courts should stop there.”); Vanderklok v. United States, 868 F.3d 189, 197 (3d Cir. 2017) (existence of Bivens cause of action is a “threshold question of law”). And though “privately negotiated” or privately “conferred” rights—such as entitlements allocated in

9 settlement agreements—often fail to “rise to the level of importance” required by Swint’s third criterion, Digital Equipment Corp. v. Desktop Direct, Inc., 511 U.S. 863, 876, 878–79 (1994), the Bivens defendant’s right not to be subject to a claim for which no cause of action lies is assured by the Constitution.2

Judicial creation of a cause of action against federal officers places “great stress on the separation of powers.” Nestlé USA, Inc. v. Doe, 141 S. Ct. 1931, 1938 (2021). And orders “rais[ing] questions of clear constitutional importance” are sufficiently important to warrant immediate review. Sell, 539 U.S. at 176; see also Chehazeh, 666 F.3d at 138–39 (an interest raising substantial concerns that implicate constitutional safeguards is “compelling” under Swint’s third criterion). A court’s decision to authorize a Bivens cause of action also generates “substantial costs” for Executive officers. Ziglar, 137 S. Ct. at 1855. Expansion of Bivens in violation of the separation of powers thus disrupts effective governance, subjecting officers to the same “distraction from duty” that qualified immunity is meant to foreclose. Digital Equipment, 511 U.S. at 881 (cleaned up). That harm cannot be undone even

2 Though that entitlement is not express in the Constitution or federal law, the Supreme Court long ago discarded the rule that a value or interest must “rest[] upon an explicit statutory or constitutional guarantee” to warrant interlocutory review. Midland Asphalt Corp. v. United States, 489 U.S. 794, 801 (1989) (emphasis added). The value need only “originat[e] in the Constitution or statutes,” or be “embodied in” those sources—as the separation of powers does and is. Digital Equipment, 511 U.S. at 879.

10 if the officer is acquitted. See Sell, 539 U.S. at 177. The problem with erroneous Bivens extensions, then, is “not limited to liability for money damages,” Mitchell v. Forsyth, 472 U.S. 511, 526 (1985); a more serious risk is hindrance of government interests. See Egbert, 142 S. Ct. at 1805 (courts inevitably “impair governmental interests” when they misapply the Bivens special factors analysis) (cleaned up). The Supreme Court has signaled that we should proactively mitigate those harms—including on our own initiative. See id. at 1806 n.3 (courts have a sua sponte “responsibility” to “evaluate any grounds that counsel against Bivens relief,” even those not raised by the parties, because “recognizing a Bivens cause of action is an extraordinary act”) (cleaned up). We’ve done the same by setting aside party waiver to correct Bivens errors. See Bistrian v. Levi, 912 F.3d 79, 88 (3d Cir. 2018) (overlooking waiver to reach the cognizability of a Bivens cause of action because “[t]o rule otherwise would be to allow new causes of action to spring into existence merely through the dereliction of a party”).

In sum, a court’s wrongful arrogation of the legislative power to create a cause of action for claims of constitutional torts against federal officers violates the constitutional separation of powers and disrupts effective governance. Because those harms are immediate and those interests essential, an order wrongly authorizing a Bivens claims to proceed is “effectively unreviewable” on appeal after final judgment.

Having explained why Boresky satisfies Swint’s third criterion, I proceed to discuss the first two.

11 A decision authorizing a Bivens cause of action resolves an important question of law separate from the claim’s merits. Whether a plaintiff can show that a federal officer committed a constitutional tort against him is legally distinct from whether his claim is cognizable under Bivens. See Dongarra, 27 F.4th at 177 (explaining that a Bivens plaintiff must clear two distinct “hurdles” to recover damages: show an invasion of his legal rights, and show that “Bivens lets him sue”). That’s why the District Court could analyze the Bivens question here without adverting once to Fourth Amendment doctrine. See Graber v. Dales, 2019 WL 4805241, at *2–6 (E.D. Pa. Sept. 30, 2019). And the threshold question of cognizability does not merge with the merits question: the “fact that an issue is outcome determinative does not mean that it is not ‘collateral’ for purposes of the Cohen test.” Mitchell, 472 U.S. at 529 n.10.

Bivens analysis does require comparing the facts of an alleged constitutional violation to the facts of cases in which the Supreme Court authorized Bivens causes of action. But other legal issues that we review on interlocutory appeal under § 1291 involve similar comparisons. Double jeopardy challenges, for instance, require us to determine whether successive prosecutions are for the same offense—yet whether the Double Jeopardy Clause bars the suit is distinct from whether the accused committed a crime. See id. at 528. In cases implicating qualified immunity, similarly, whether “a particular complaint sufficiently alleges a clearly established violation of law cannot be decided in isolation from the facts pleaded.” Ashcroft, 556 U.S. at 673. And qualified immunity analysis looks to precedent for law enshrining “clearly established” rights. Kisela v. Hughes, 138 S. Ct. 1148, 1152 (2018) (per curiam). Yet whether qualified immunity precludes the suit is also distinct from whether the official’s actions were

12 unlawful. Mitchell, 472 U.S. at 528–29. With qualified immunity as with Bivens, a plaintiff must do more than establish the merits of his tort claim to receive the requested relief. The majority puts the point well: though the merits question here is whether Boresky violated Graber’s Fourth Amendment rights, the Bivens question is “can [Graber] sue at all?” Maj. Op. 16 n.14.

Finally, a decision authorizing a Bivens cause of action conclusively determines whether the claim can be maintained: but for that decision, a court should “reject” the claim. Hernandez II, 140 S. Ct. at 743. Where no Bivens cause of action lies, courts should “stop there.” Elhady, 18 F.4th at 884. And once the district court rules on the issue at the summary judgment stage, the defendant typically will take “no further steps” to dismiss the claim on this ground. Abney v. United States, 431 U.S. 651, 659 (1977). That’s because the defendant’s pleadings, discovery, and summary judgment record almost always will show whether the facts of the case mirror those of the Supreme Court’s Bivens authorizations. Under these circumstances, a court will not “meaningfully reconsider” its Bivens authorization after summary judgment. See Doe, 997 F.3d at 493.

* * *

Egbert wasn’t the death knell for Bivens, but it nearly rang it. See Egbert, 142 S. Ct. at 1810 (Gorsuch, J., concurring in the judgment) (Egbert leaves “barely implicit” the conclusion that the “right answer” to whether to authorize a Bivens cause of action “will always be no”). The Supreme Court’s deep skepticism toward Bivens and its progeny highlights the profound separation of powers implications of every erroneous expansion of Bivens by federal courts. The “crucial question”

13 here is whether deferring until final judgment our review of an order allowing a Bivens cause of action to proceed “so imperils” the separation of powers as to justify immediate appeal as of right. Mohawk, 558 U.S. at 108. Because I believe it does, I would recognize such orders as “final” under § 1291 and the collateral order doctrine.3

3 The “discretionary availability” of interlocutory certification under 28 U.S.C. § 1292(b) makes the majority “hesita[nt] to agree” with this conclusion because that mechanism “counsels against” expanding the class of collateral orders. Maj. Op. 16 n.14. Though that hesitation may be prudent as a general matter, the majority does not explain why § 1292(b) counsels against collateral recognition of orders authorizing Bivens causes of action. First, it doesn’t follow from an issue’s appropriateness for § 1292(b) certification that the issue is unsuitable for collateral appeal under § 1291. Section 1292(b) authorizes parties to request that district courts certify, and empowers the Courts of Appeals to grant, interlocutory appeals involving “controlling question[s] of law” on which there is “substantial ground for difference of opinion,” provided the appeal may “materially advance” the litigation’s termination. 28 U.S.C. § 1292(b). If (collateral) orders denying a challenge to a Bivens cause of action satisfy those criteria, so do (collateral) orders denying qualified immunity or a double jeopardy defense. So even if the Bivens order under review satisfied § 1292(b), that fact wouldn’t support the majority’s holding. In any event, the order before us likely would not satisfy § 1292(b). After Egbert, the answer to the question whether courts can authorize a Bivens cause of action will almost always be “no.” See Egbert, 142 S. Ct. at 1810 (Gorsuch, J., concurring in the

14 III

Having explained why appellate jurisdiction lies, I turn to the merits. Did the District Court err when it authorized a Bivens cause of action against Boresky for swearing out a warrant that lacked probable cause? It did.

We ask two questions to determine whether a Bivens cause of action is cognizable. Does the claim arise in a new context by differing “in a meaningful way” from previous Bivens causes of action the Supreme Court has authorized? Ziglar, 137

judgment); cf. Maj. Op. 12 (“Bivens actions are very limited.”). So there’s little ground for difference of opinion as to whether authorization is permitted. Second, the majority notes that one of § 1292(b)’s “regular uses is to permit interlocutory appeal to decide whether a statute permits a private cause of action.” Maj. Op. 15 n.14. But none of the cases the majority cites to support that statement involves a federal defendant and thus the threat to effective governance that Bivens authorizations pose. See Ziglar, 137 S. Ct. at 1856. Courts have tools other than the collateral order doctrine to facilitate interlocutory appeals of important legal issues. For instance, 28 U.S.C. § 1292(e) and § 2072(c) authorize the Supreme Court to prescribe rules governing interlocutory appeals, including by designating certain classes of orders “final” under § 1291. The existence of alternative mechanisms for interlocutory appeal gives us reason to mark the boundary of the class of collateral orders “stringent[ly].” Digital Equipment, 511 U.S. at 883. But that proposition does not forbid us from recognizing new collateral orders, and our Court continues to recognize them notwithstanding those alternative mechanisms.

15 S. Ct. at 1859. If so, we then ask if any “special factors counsel[] hesitation” before extending Bivens into that new context. Id. at 1857. If there are, the cause of action cannot proceed. Those two inquiries often resolve into one: is there “any reason to think that Congress might be better equipped to create a damages remedy[?]” Egbert, 142 S. Ct. at 1803 (emphasis added). The court should not authorize the Bivens cause of action if there is.

Graber’s claim against Boresky arises in a new context. Among the circumstances “meaningful enough to make a given context a new one” are differences in the constitutional right at issue and the risk of disruptive intrusion by the Judiciary into the functioning of coordinate branches. See Ziglar, 137 S. Ct. at 1859–60. Graber’s allegations differ in at least these two respects from Bivens.

The defendants in Bivens conducted a warrantless search during which they “manacled” a man in front of his family, threatened to arrest the family, booked the man at the federal courthouse, and subjected him to a strip search. Bivens, 403 U.S. at 389. Here, Boresky has been sued for charging Graber based on a warrant that purportedly lacked probable cause. So Graber invokes a different constitutional provision than Mr. Bivens did. Compare U.S. Const., amend. IV (guaranteeing that “no Warrants shall issue, but upon probable cause”), with id. (proscribing “unreasonable searches and seizures”). And the Supreme Court has “repeatedly refused” to extend Bivens “beyond the specific clauses of the specific amendments for which a cause of action has already been implied.” Vanderklok, 868 F.3d at 200 (emphasis added). If that weren’t enough, our intrusion into the Secret Service’s management of the

16 government’s response to security breaches occurring at National Special Security Events would also disrupt the workings of the political branches. These differences from Bivens establish that Graber’s claim arises in a new context.

Second, multiple special factors counsel hesitation in authorizing a new Bivens cause of action for claims like Graber’s. We have noted that two Ziglar factors are “particularly weighty”: the “existence of an alternative remedial structure and separation-of-powers principles.” Bistrian, 912 F.3d at 90. Another special factor is “whether national security is at stake.” Id. All these factors militate against allowing the Bivens claim to proceed against Boresky.

An alternative remedial process is available to plaintiffs like Graber. The Secret Service is a component of the Department of Homeland Security. See 6 U.S.C. § 381. Graber can report alleged civil rights abuses by the Secret Service to DHS’s Office of the Inspector General. See Hotline, Office of the Inspector General, https://www.oig.dhs.gov/hotline. Congress has provided for a senior official within the Office to receive and review complaints about and to investigate alleged civil rights abuses. 5. U.S.C. App. 3 § 8I(f)(1)–(2). That procedure need not involve complainant participation or the right to judicial review. Egbert, 142 S. Ct. at 1806. What matters is that Congress or the Executive has created a remedial process it deems sufficient to secure deterrence of wrongful conduct. Id. at 1807. We cannot “second-guess that calibration by superimposing a Bivens remedy.” Id. Doing so would raise obvious separation of powers concerns.

17 Authorizing a Bivens cause of action here also would require us to interfere with sensitive Executive-branch functions. See Ziglar, 137 S. Ct. at 1861; Mack v. Yost, 968 F.3d 311, 323 (3d Cir. 2020) (declining to authorize a Bivens cause of action because “judicial intervention” in “administrative decisions would improperly encroach upon the executive’s domain”). Those functions—coordinating the government’s security plan for keeping high-level officers and candidates safe at a National Special Security Event—involve national security. Whether to create a “new substantive legal liability” for Secret Service agents participating in a coordinated response to security breaches is the sort of choice Congress, not the courts, should make. Ziglar, 137 S. Ct. at 1857 (cleaned up). Our failure to heed that counsel would embroil us in policy judgments we are ill-suited for. See Egbert, 142 S. Ct. at 1804–05. And we cannot “predict the ‘systemwide’ consequences” that would follow if we were to expand Bivens to allow suits like this one against Secret Service agents. See id. at 1803–04. A “Bivens cause of action may not lie where . . . national security is at issue.” Id. at 1805.

The District Court’s decision to authorize Graber’s Bivens cause of action was contrary to a spate of recent Supreme Court decisions. I would vacate its order and remand with instructions to dismiss Graber’s amended complaint against Boresky.