Clarissa Gilmore v. Georgia Department of Corrections, et al, 144 F.4th 1246

Before W ILLIAM P RYOR , Chief Judge, and J ORDAN , R OSENBAUM , J ILL P RYOR , N EWSOM , B RANCH , G RANT , L UCK , L AGOA , B RASHER , A BUDU , K IDD , and T JOFLAT , Circuit Judges.

J ORDAN , Circuit Judge, delivered the opinion of the Court in which W ILLIAM P RYOR , Chief Judge, and N EWSOM , B RANCH , G RANT , L AGOA , B RASHER , A BUDU , and T JOFLAT , Circuit Judges, joined, and in which L UCK Circuit Judge, joined as to Parts II.C, III.B.2, and IV. W ILLIAM P RYOR , Chief Judge, filed a concurring opinion in which L AGOA and T JOFLAT , Circuit Judges, joined. , Circuit Judge, filed a concurring opinion in which J ILL

P RYOR and K IDD , Circuit Judges, joined. Opinion of the Court

T JOFLAT Circuit Judge, filed a concurring opinion in which L AGOA and B RASHER , Circuit Judges, joined, and in which L UCK , Circuit Judge, joined as to Parts III, IV, and V.

J ORDAN , Circuit Judge:

Clarissa Gilmore sued several Georgia correctional officers under 42 U.S.C. § 1983 for subjecting her to a strip search in Febru- ary of 2017 when she visited her then-husband at Smith State Prison in Georgia. She alleged that the officers did not have any suspicion to conduct a strip search, that they coerced her consent by threatening her with detention, that they failed to give her the option to forgo her visit and leave the facility, and that the strip search involved physical touching of intimate body parts and a vis- ual body-cavity inspection.

The district court granted summary judgment to the officers on qualified immunity grounds, and a panel of this court affirmed. The panel, viewing the evidence in the light most favorable to Ms. Gilmore, concluded that the officers violated the Fourth Amend- ment because they lacked reasonable suspicion for the strip search but agreed with the district court that the officers were entitled to qualified immunity because Supreme Court and Eleventh Circuit precedent did not clearly establish at the time of the strip search that reasonable suspicion was required. See Gilmore v. Ga. Dept. of Corr. , 111 F.4th 1118, 1130–36 (11th Cir. 2024).

Two members of the panel wrote separate concurrences. Judge Rosenbaum explained that our cases, contrary to Supreme Court precedent, failed to look to “a robust consensus of Opinion of the Court persuasive authority” in determining whether the law was clearly established for qualified immunity purposes. And she suggested that the case be reheard en banc to conform our caselaw to the Su- preme Court’s teachings. See id. at 1136–38 (Rosenbaum, J., con- curring). Judge Newsom generally agreed with Judge Rosen- baum’s concern and wrote to set out some “oddities” in qualified immunity jurisprudence. See id. at 1138–41 (Newsom, J., concur- ring).

We voted to rehear the case as a full court, see Gilmore v. Ga. Dept. of Corr. , 119 F.4th 839 (11th Cir. 2024) (en banc), and asked the parties to brief two issues:

1. Whether Marsh v. Butler County , 268 F.3d 1014, 1032 n.10 (11th Cir. 2001) (en banc), and Thomas ex rel. Thomas v. Roberts , 323 F.3d 950, 955 (11th Cir. 2003), should be over- ruled to the extent that they hold or state that no amount of out-of-circuit authority can clearly establish the law for purposes of qualified immunity.

2. If so, whether a “robust consensus of persuasive author- ity” clearly established that the strip search violated Ms. Gilmore’s Fourth Amendment rights.

After hearing oral argument, we asked the parties to brief two ad- ditional issues:

3. Whether a jury could find that the strip search violated the Fourth Amendment if it credits Ms. Gilmore’s ver- sion of events. Opinion of the Court

4. If so, whether the Fourth Amendment violation was one of “obvious clarity” such that the officers are not entitled to qualified immunity.

We now answer the last two questions affirmatively.

First, if it credits the version of events presented by Ms. Gil- more, a jury could find under the totality of the circumstances that the officers who conducted the strip search violated her Fourth Amendment rights. The strip search was not justified at its incep- tion because the officers (1) lacked even reasonable suspicion that Ms. Gilmore was involved in any illegal activity, (2) coerced her con- sent through a threat of detention, and (3) failed to give her the option to forgo her visit and leave the facility. The search was also unreasonable in scope because it involved the physical touching of intimate body parts and a visual body-cavity inspection.

Second, for all of the reasons summarized above, the Fourth Amendment violation was one of “obvious clarity.” As a result, the officers who conducted the strip search are not entitled to qualified immunity at this stage of the litigation.

As for the first two questions, Marsh and Thomas do not hold that cases from our sister circuits cannot be considered in deter- mining whether a constitutional violation was one of “obvious clarity” for purposes of qualified immunity. To the extent that lan- guage in Marsh , Thomas , and other Eleventh Circuit cases can be read to suggest that out-of-circuit authority is irrelevant in deter- mining whether the law was clearly established, we now clarify that Opinion of the Court such authority may indeed be considered in an “obvious clarity” scenario.

We leave for another day the broader questions of what con- stitutes a “robust consensus of persuasive authority” and whether such a consensus can alone constitute clearly established law in the absence of Supreme Court or Eleventh Circuit precedent.

I Many of the facts here are contested. At the summary judg- ment stage, however, we resolve any conflicts in favor of Ms. Gil- more. We do so not only to decide whether a jury could find in her favor on the Fourth Amendment claim, but also to determine whether the officers who conducted the strip search are entitled to qualified immunity. See Tolan v. Cotton , 572 U.S. 650, 657 (2014). Viewed through this prism, here is the evidence in the light most favorable to Ms. Gilmore, taken from the panel opinion and the record. See Gilmore , 111 F.4th at 1124–25.

A Twice a month, Ms. Gilmore visited her then-husband, Mulik Sheets, at Smith State Prison in Georgia. On February 26, 2017, she arrived, as she had roughly fifty times before, and success- fully proceeded through the initial security screening. That meant undergoing three different types of searches: a pat-down search, a metal-detector wand search, and an electromagnetic-radia- tion/body-scan search. Opinion of the Court

Smith State Prison did not have any signs warning visitors that they might be (or would be) subject to strip searches. There was a sign posted outside the sallyport, but it only told visitors that by seeking to enter they “consent[ed]” to “a search of their person and property at any time,” including “searches by the use of X-ray devices, metal detectors, body scanners, and pat down searches.”

Once Ms. Gilmore cleared the initial security screening, cor- rectional officers escorted her to a second building where the visit- ation room was located. Officer Sabrini Lupo assigned Ms. Gil- more to a table, where Mr. Sheets joined her, and their visit began. Officer Lupo and Lieutenant Alberta Milton remained present in the visitation room during the visit.

About thirty minutes into the visit, Ms. Gilmore noticed that Lieutenant Milton was staring at her. Ms. Gilmore stared back for “one to two minutes.” In apparent response, Lieutenant Milton walked past Ms. Gilmore and returned to the front of the room, where she spoke to other officers. Then Lieutenant Milton left the visitation room.

When Lieutenant Milton returned with Officer Christina Iri- zarry, she told Ms. Gilmore to go with her. Lieutenant Milton and Officer Irizarry took Ms. Gilmore into the hallway and handed her a strip-search approval form. That form was blank and lacked ap- proval signatures from prison officials. Ms. Gilmore asked why the officers were going to strip search her, but Lieutenant Milton re- fused to tell her. Ms. Gilmore also asked Lieutenant Milton if she Opinion of the Court could speak to her supervisor, but Lieutenant Milton responded that she was the officer in charge that day.

Lieutenant Milton and Officer Irizarry insisted that Ms. Gil- more sign the strip-search approval form. If she didn’t, they said, she would be sent to jail and would be unable to visit her husband again. Not only that, Lieutenant Milton told Ms. Gilmore that the officers would “search [her] anyway.” Ms. Gilmore “didn’t feel like [she] had an option,” so she signed the form.

After Ms. Gilmore signed the form, Lieutenant Milton and Officer Irizarry led her to an empty bathroom and instructed her to remove all of her clothes, including her bra and underwear. Ms. Gilmore complied. Officer Irizarry examined Ms. Gilmore’s cloth- ing for contraband but found nothing.

When Officer Irizarry finished, at Lieutenant Milton’s direc- tion, she conducted a manual search of Ms. Gilmore. Officer Iri- zarry first manipulated Ms. Gilmore’s breasts, lifting each breast and looking underneath it. Lieutenant Milton then ordered Ms. Gilmore to “[t]urn around,” “bend over,” and “open [her] butt cheeks.” Ms. Gilmore did as instructed, and Officer Irizarry “felt in between” Ms. Gilmore’s buttocks with her gloved hand.

Lieutenant Milton and Officer Irizarry next instructed Ms. Gilmore to spread her vagina, which they visually inspected. Find- ing no contraband, they told Ms. Gilmore to put her clothes back on and allowed her to resume her visit.

Officer Irizarry led Ms. Gilmore back to the visitation room and told her that she was “so sorry.” Ms. Gilmore stayed until Opinion of the Court visitation ended, but she and Mr. Sheets barely spoke because she was upset and “tearing up.” Ms. Gilmore left Smith State Prison and cried through her drive home.

B Two days later, Ms. Gilmore called Deputy Warden Tamarshe Smith to complain and ask why she had been strip- searched. Deputy Warden Smith seemed unaware of the incident. He told Ms. Gilmore that he would look into it and call her back. When Ms. Gilmore spoke to Deputy Warden Smith again, he apol- ogized and “said that he did not see anything on the video” footage of the visitation room “that would warrant a strip search.”

Officer Lupo testified that she smelled marijuana on Ms. Gil- more during her visit to Smith State Prison, and that she shared her observation with Lieutenant Milton. Officer Lupo also testified that she found it suspicious that Ms. Gilmore was staring at her and Lieutenant Milton. But the witness statements of Lieutenant Mil- ton and Officer Irizarry, recorded on the day of the incident, did not mention any marijuana odor or suspicious eye contact. Nor did the strip-search approval form, which stated only that Ms. Gil- more was “[u]nder suspicion for carrying contraband.”

Lieutenant Milton testified that she called Deputy Warden Smith before conducting the strip search, and he gave her verbal approval. But Lieutenant Milton’s contemporaneous statement contains no reference to any such call. And duty records show that Deputy Warden Smith was not working the day of the search. For summary judgment purposes, then, a reasonable inference is that Opinion of the Court Lieutenant Milton did not obtain verbal approval from Deputy Warden Smith for the strip search. ^[1]

For her part, Ms. Gilmore denied consuming, possessing, or smelling like marijuana at any point before or during her visit. She also denied staring at Officer Lupo in the visitation room, and only stared back at Lieutenant Milton in response to the latter’s initial staring.

II Our discussion relates only to the Fourth Amendment claims against Lieutenant Milton and Officer Irizarry, who con- ducted the strip search of Ms. Gilmore. We first address whether a jury could find that the strip search violated the Fourth Amend- ment. ^[2]

A The Fourth Amendment, as relevant here, provides that the “right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated[.]” U.S. Const. amend. IV. It applies to the states through Opinion of the Court the Due Process Clause of the Fourteenth Amendment. See, e.g., City of Ontario v. Quon , 560 U.S. 746, 750 (2010). It is undisputed that the non-consensual strip search of Ms. Gilmore, which in- cluded both physical touching of intimate body parts and a visual body-cavity inspection, was both a search and a seizure under the Fourth Amendment. See Florence v. Bd. of Chosen Freeholders , 566 U.S. 318, 325 (2012) (upholding, as reasonable under the Fourth Amendment, a county jail’s policy requiring that all arrestees be strip-searched before being admitted to general population); Torres v. Madrid , 592 U.S. 306, 317 (2021) (explaining that the “appropriate inquiry” for determining whether there has been a Fourth Amend- ment seizure “is whether the challenged conduct objectively mani- fests an intent to restrain”) (emphasis in original).

As a general matter, Fourth Amendment reasonableness takes into account the totality of the circumstances. See, e.g., County of Los Angeles v. Mendez , 581 U.S. 420, 427 (2017) (use of force); Samson v. California , 547 U.S. 843, 848 (2006) (search); Florida v. Bostick , 501 U.S. 429, 437 (1991) (seizure). The Supreme Court has applied that same holistic approach to strip searches in the school context, and so have we. See Safford Unified Sch. Dist. No. 1 v. Redding , 557 U.S. 364, 375 (2009) (analyzing the strip search of a teenage student at a middle school); T.R. by and through Brock v. La- mar Cnty. Bd. of Ed. , 25 F.4th 877, 883 (11th Cir. 2022) (same). We therefore consider the totality of the circumstances in evaluating the reasonableness of Ms. Gilmore’s strip search. Opinion of the Court

“[W]hat is reasonable depends on the context within which a search takes place.” New Jersey v. T.L.O. , 469 U.S. 325, 337 (1985). In the prison context, the Supreme Court has explained that the “test of reasonableness under the Fourth Amendment is not capa- ble of precise definition or mechanical application. In each case it requires a balancing of the need for the particular search against the invasion of personal rights that the search entails. Courts must consider the scope of the particular intrusion, the manner in which it is conducted, the justification for initiating it, and the place in which it is conducted.” Bell v. Wolfish , 441 U.S. 520, 528, 559 (1979) (upholding a detention center policy which required all pretrial de- tainees to “expose their body cavities for inspection following con- tact visits”).

“Determining the reasonableness of any search involves a twofold inquiry: first, one must consider ‘whether the . . . action was justified at its inception[;]’ second, one must determine whether the search as actually conducted ‘was reasonably related in scope to the circumstances which justified the interference in the first place[.]’” T.L.O. , 469 U.S. at 341 (citations omitted). We discuss these two matters below.

B The “amount of suspicion required to justify a particular search depends on the intrusiveness of that search.” United States v. Pino , 729 F.2d 1357, 1359 (11th Cir. 1984). The Supreme Court has not addressed the level of suspicion necessary for a strip search of a prison visitor. And prior to the panel’s decision, see Gilmore , Opinion of the Court 111 F.4th at 1128, nor had we. But the nine circuits to have faced the issue have all held that such a search requires reasonable suspi- cion. See Wood v. Clemons , 89 F.3d 922, 928–29 (1st Cir. 1996); Var- rone v. Bilotti , 123 F.3d 75, 79 (2d Cir. 1997); Calloway v. Lokey , 948 F.3d 194, 202 (4th Cir. 2020); Thorne v. Jones , 765 F.2d 1270, 1276 (5th Cir. 1985); Daugherty v. Campbell , 935 F.2d 780, 787 (6th Cir. 1991); Burgess v. Lowery , 201 F.3d 942, 945 (7th Cir. 2000); Hunter v. Auger , 672 F.2d 668, 674 (8th Cir. 1982); Cates v. Stroud , 976 F.3d 972, 985 (9th Cir. 2020); Romo v. Champion , 46 F.3d 1013, 1020 (10th Cir. 1995).

The strip search at issue here involved both the physical touching of intimate body parts (the breasts and the buttocks) and a visual body-cavity inspection. We need not definitively decide whether this type of strip search requires reasonable suspicion or probable cause. But for the following reasons we conclude that, at the very least, correctional officers must have reasonable suspicion that a visitor is concealing contraband (e.g., drugs or weapons) be- fore they subject her to a strip search.

A visitor who seeks admission to a prison has a diminished expectation of privacy and can expect to have her person and prop- erty searched because contact visits can “open the institution to the introduction of drugs, weapons, and other contraband,” Block v. Rutherford , 468 U.S. 576, 586 (1984), but she is not in the same posi- tion as an incarcerated detainee or inmate for purposes of the Fourth Amendment. As the panel correctly put it, “a free person visiting a prison is in a different position for Fourth Amendment Opinion of the Court purposes than a prisoner or detainee.” Gilmore , 111 F.4th at 1130. See also T.L.O. , 469 U.S. at 338 (stating that civilians and prisoners “stand in wholly different circumstances, separated by the harsh facts of criminal conviction and incarceration”) (quoting Ingraham v. Wright , 430 U.S. 651, 669 (1977)); Blackburn v. Snow , 771 F.2d 556, 563 (1st Cir. 1985) (“[T]hose visiting a prison cannot credibly claim to carry with them the full panoply of rights they normally enjoy. But neither may they constitutionally be made to suffer a wholesale loss of rights—nor even one commensurate with that suffered by inmates.”).

We reject the broad contention of Lieutenant Milton and Officer Irizarry that, for purposes of determining the Fourth Amendment reasonableness of a strip search, a prison visitor’s sta- tus is irrelevant. See Appellees’ Supp. Br. at 7–9. A visitor’s status may not be determinative, but the claim that the prison setting is the only thing that matters—regardless of the type of search that is conducted—is a constitutional bridge too far. Such a bright-line rule goes against the balancing that the Supreme Court has called for in the prison context. See Bell , 441 U.S. at 559. Accord Scott v. Harris , 550 U.S. 372, 383 (2007) (“Although respondent’s attempt to craft an easy-to-apply legal test in the Fourth Amendment context is admirable, in the end we must still slosh our way through the factbound morass of ‘reasonableness.’”).

The one prison visitor/strip search case that Lieutenant Mil- ton and Officer Irizarry cite in support of their argument, State v. Martinez , 580 P.2d 1282 (Haw. 1978), is distinguishable on an Opinion of the Court important point. In Martinez the visitor had been strip-searched before “on several previous visits” to the prison. See id. at 1284. Based on these prior experiences, the Hawaii Supreme Court con- cluded that the visitor’s “consent to her search [wa]s . . . established in the present case, and the reasonableness of the search [had to] be judged in the light of that circumstance.” Id. at 1286. Here there is no evidence whatsoever that Ms. Gilmore had been strip- searched on any of her approximately fifty prior visits to Smith State Prison. Nor is there any evidence that posted signs warned visitors that they might (or would be) subject to a strip search. Mar- tinez simply cannot bear the jurisprudential weight that Lieutenant Milton and Officer Irizarry seek to place on it. ^[3]

There is “no authority for the proposition that strip searches of prison visitors are per se reasonable.” Thorne , 765 F.2d at 1276. In the words of a leading Fourth Amendment treatise, “a rule Opinion of the Court requiring all prison visitors to submit to a body cavity strip search, without any predicate requirement of individualized suspicion or showing of special and highly unusual institutional need, cannot satisfy the Fourth Amendment.” 5 Wayne R. La Fave, Search and Seizure: A Treatise on the Fourth Amendment § 10.7(b) (6th ed. & Nov. 2024 update) (quoting Blackburn , 771 F.2d at 562). Accord 2 Michael B. Mushlin, Rights of Prisoners § 9:21 (5th ed. & Nov. 2024 update) (“[R]outine strip searches of visitors are unconstitu- tional.”).

This should not be surprising, for a strip search is not a rou- tine or minimally intrusive means of maintaining prison security with respect to visitors. As we have explained, “a strip search rep- resents a serious intrusion upon personal rights[;]” it is “demean- ing, dehumanizing, undignified, humiliating, terrifying, unpleas- ant, embarrassing, [and] repulsive, signifying degradation and sub- mission.” Justice v. Peachtree City , 961 F.2d 188, 192 (11th Cir. 1992) (internal quotation marks and citation omitted). See also Redding , 557 U.S. at 374–75 (“embarrassing, frightening, and humiliating”). And that intrusion is magnified when—as here—there is “physical contact between the searcher and the person searched” and/or there is “exposure of intimate body parts[.]” United States v. Touset , 890 F.3d 1227, 1234 (11th Cir. 2018) (internal quotation marks and citation omitted).

We hold, therefore, that Lieutenant Milton and Officer Iri- zarry needed at least reasonable suspicion to subject Ms. Gilmore Opinion of the Court to a strip search. And if they lacked reasonable suspicion, they nec- essarily lacked probable cause, which is a higher standard.

C Considering the totality of the circumstances, and viewing the evidence in the light most favorable to Ms. Gilmore, the strip search conducted by Lieutenant Milton and Officer Irizarry vio- lated the Fourth Amendment at its inception. There was no suspi- cion whatsoever, the consent was coerced, there was no option to refuse consent and leave the facility, and the strip search was unrea- sonable in scope.

No suspicion . The “concept of reasonable suspicion is somewhat abstract,” and the Supreme Court has resisted efforts to reduce it to a “neat set of legal rules.” United States v. Arvizu , 534 U.S. 266, 274 (2002) (internal quotation marks and citations omit- ted). Nevertheless, reasonable suspicion is concerned with proba- bilities, and “[t]he officer must be able to articulate more than an inchoate and unparticularized suspicion or ‘hunch’ of criminal ac- tivity.” Illinois v. Wardlow , 528 U.S. 119, 123–24 (2000) (internal quo- tation marks and citation omitted).

Lieutenant Milton and Officer Irizarry did not have any sus- picion for the strip search. Ms. Gilmore had successfully cleared an initial security screening that consisted of a pat-down search, a metal-detector wand search, and an electromagnetic-radia- tion/body-scan search. She testified that she did not stare at Officer Lupo when she was in the visitation room, that she stared back at Lieutenant Milton only after the latter stared at her first, and that Opinion of the Court she did not smell of marijuana. There was no reasonable suspicion, and certainly no probable cause, for a strip search.

Coerced consent . Lieutenant Milton and Officer Irizarry coerced Ms. Gilmore’s consent. They told her that if she did not sign the strip-search approval form, she would be taken into cus- tody and would be strip-searched anyway. It has long been the law that the “Fourth and Fourteenth Amendments require that a con- sent not be coerced, by explicit or implicit means, by implied threat or covert force.” Schneckloth v. Bustamonte , 412 U.S. 218, 228 (1973). And “a search conducted pursuant to a coerced consent is not valid because the consenting party is forced to allow the search.” United States v. Rackley , 742 F.2d 1266, 1271 (11th Cir. 1984).

No option to leave . A fair inference from the record is that Lieutenant Milton and Officer Irizarry—by telling Ms. Gilmore that she would be detained and would be strip-searched anyway— did not give her the option to forgo her visit, leave Smith State Prison, and avoid the strip search. The Sixth, Seventh, and Ninth Circuits have held that a prison visitor cannot be forced to undergo a strip search if she wants to leave the facility, and we agree with them. See Spear v. Sowders , 71 F.3d 626, 632 (6th Cir. 1995) (en banc) (“While a person may consent to less invasive searches merely by entering the facility, we do not think that a person consents to a strip and body cavity search by simply appearing at a visiting center. Instead, the same logic that dictates that such a search may be con- ducted only when there is reasonable suspicion also demands that the person to be subjected to such an invasive search be given the Opinion of the Court opportunity to depart.”); Burgess , 201 F.3d at 947 (agreeing with Spear and explaining that “a general conditioning of prison visita- tion on subjection to a strip search is manifestly unreasonable”); Cates , 976 F.3d at 984 (“Even if there was reasonable suspicion that Cates was seeking to bring drugs into the prison (a question we do not reach), [the officer] violated her rights under the Fourth Amendment by subjecting her to a strip search without giving her the option of leaving the prison rather than being subjected to the search.”). Accord Blackburn 771 F.2d at 568 (conditioning access to jail “upon sacrifice of [the] right to be free of an otherwise unrea- sonable strip search” is “constitutionally intolerable”). ^[4]

Unreasonable Scope . The strip search was, moreover, un- reasonable in scope. This was not a strip search simpliciter , i.e., one which merely consists of the person’s “clothes [being] removed.” Search , Black’s Law Dictionary 1623 (12th ed. 2024). It was, instead, constitutionally more intrusive in two significant ways.

The strip search involved the physical touching of intimate body parts. Officer Irizarry lifted Ms. Gilmore’s breasts to look un- derneath them and felt in between Ms. Gilmore’s buttocks with her Opinion of the Court gloved hand. A “[p]hysically invasive inspection is simply more in- trusive than [a] purely visual inspection.” Bond v. United States , 529 U.S. 334, 337 (2000) (discussing the physical manipulation of a pas- senger’s carry-on bag). And where there is physical touching like there was here, a “greater amount of suspicion is necessary.” Pino , 729 F.2d at 1359. See also Amaechi v. West , 237 F.3d 356, 363 (4th Cir. 2001) (“Public exposure of the genitalia accompanied by physical touching is far more intrusive than directing an arrestee to remove her clothing in private for the purpose of ‘visually inspecting’ the arrestee’s genitalia.”).

In addition, the strip search involved a visual body-cavity in- spection which required Ms. Gilmore to bend over and spread her vagina. Needless to say, this exposition of an intimate area was more invasive and implicated even greater privacy concerns: “[V]is- ual body cavity searches are even more intrusive [than strip searches]. They require a [person] not only to strip naked in front of a stranger, but also to expose the most private areas of her body to others. This is often, as here, done while the person . . . is re- quired to assume degrading and humiliating positions.” Sloley v. VanBramer , 945 F.3d 30, 38 (2d Cir. 2019) (internal quotation marks and citation omitted). See generally William E. Ringel, Searches and Seizures, Arrests, and Confessions § 16:20 (2d ed. & March 2025 up- date) (“When more intrusive strip searches are involved . . . the governmental interests must be more closely scrutinized.”). Opinion of the Court

III The remaining question is whether Lieutenant Milton and Officer Irizarry are entitled to qualified immunity. “Qualified im- munity attaches when an offic[er’s] conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known. Because the focus is on whether the officer had fair notice that her conduct was unlawful, reasonable- ness is judged against the backdrop of the law at the time of the conduct.” Kisela v. Hughes , 584 U.S. 100, 104 (2018) (internal quota- tion marks and citation omitted).

“A right may be clearly established for qualified immunity purposes in one of three ways: (1) case law with indistinguishable facts clearly establishing the constitutional right; (2) a broad state- ment of principle within the Constitution, statute, or case law that clearly establishes a constitutional right; or (3) conduct so egre- gious that a constitutional right was clearly violated, even in the total absence of case law.” T.R. , 25 F.4th at 883 (internal quotation marks and citation omitted). Under the third method, “a general constitutional rule . . . may apply with obvious clarity to the spe- cific conduct in question, even though the very action in question has not previously been held unlawful.” Hope v. Pelzer , 536 U.S. 730, 741 (2002) (internal quotation marks and citation omitted) (deny- ing qualified immunity to prison guards who tied a shirtless inmate to a hitching post in the hot sun for approximately seven hours, gave him water only once or twice, and denied him any bathroom breaks). Opinion of the Court

We conclude that the strip search here, considering all the circumstances, was so clearly prohibited that a reasonable officer would have known of its unconstitutionality in February of 2017. The Fourth Amendment violation was, in other words, one of “ob- vious clarity.”

A Before returning to the record, which we view in the light most favorable to Ms. Gilmore, we consider a recent qualified im- munity case in which the Supreme Court found that a constitu- tional violation was of “obvious clarity” despite the lack of cases with similar facts.

In Taylor v. Riojas , 592 U.S. 7 (2020), a state inmate, Taylor, asserted an Eighth Amendment claim against several correctional officers. In his verified complaint, he alleged that he did not eat or drink for nearly four days while he was housed in a cell that was “covered, nearly floor to ceiling, in massive amounts of feces: all over the floor, the ceiling, the window, the walls, and even packed inside the water faucet.” Id. at 7–8 (internal quotation marks and citation omitted). He also alleged that he was moved to a “frigidly cold cell, which was equipped with only a clogged drain in the floor to dispose of bodily wastes,” and that he “held his bladder for over 24 hours, but he eventually (and involuntarily) relieved himself, causing the drain to overflow and raw sewage to spill across the floor.” Id. at 8. “Because the cell lacked a bunk, and because Taylor was confined without clothing, he was left to sleep naked in sew- age.” Id. Opinion of the Court

The Fifth Circuit affirmed the district court’s grant of sum- mary judgment in favor of the officers on qualified immunity grounds. Although it held that the conditions alleged by Taylor violated the Eighth Amendment, the Fifth Circuit concluded that the law was not clearly established that placing an inmate in cells “teeming with human waste” for “only six days” was unconstitu- tional, and as a result the officers lacked “fair warning.” See id. at 8 (quoting Taylor v. Stevens , 946 F.3d 211, 222 (5th Cir. 2019)).

In a short opinion, the Supreme Court summarily reversed the Fifth Circuit’s qualified immunity ruling and explained, citing to Hope , that a general constitutional rule identified in the caselaw could apply with “obvious clarity” to conduct lacking an exact fac- tual parallel. See id. at 8–9. Here is how the Court explained its decision:

[N]o reasonable correctional officer could have con- cluded that, under the extreme circumstances of this case, it was constitutionally permissible to house Tay- lor in such deplorably unsanitary conditions for such an extended period of time. The Fifth Circuit identi- fied no evidence that the conditions of Taylor’s con- finement were compelled by necessity or exigency. Nor does the summary-judgment record reveal any reason to suspect that the conditions of Taylor’s con- finement could not have been mitigated, either in de- gree or duration. And although an officer-by-officer analysis will be necessary on remand, the record sug- gests that at least some officers involved in Taylor’s ordeal were deliberately indifferent to the conditions 24 Opinion of the Court

of his cells. Confronted with the particularly egre- gious facts of this case, any reasonable officer should have realized that Taylor’s conditions of confinement offended the Constitution.

Id. (citations omitted and paragraph structure altered).

B Here, considering all of the circumstances described below, the unconstitutionality of Ms. Gilmore’s strip search was of “obvi- ous clarity.” In other words, a reasonable officer in February of 2017 would have had fair notice that the strip search violated the Fourth Amendment at its inception and in its scope.

Before examining the relevant circumstances, we explain that we can consider persuasive out-of-circuit authority to deter- mine whether a violation was of “obvious clarity.”

The Supreme Court has said on numerous occasions that the decisions of other circuits are a relevant consideration in deter- mining whether the law is clearly established. “To be clearly estab- lished, a legal principle must have a sufficiently clear foundation in then-existing precedent. The rule must be ‘settled law,’ which means it is dictated by ‘controlling authority’ or ‘ a robust consensus of cases of persuasive authority [.]’” District of Columbia v. Wesby , 583 U.S. 48, 63 (2018) (citations omitted and emphasis added). Accord Plumhoff v. Rickard , 572 U.S. 765, 780 (2014); Ashcroft v. al-Kidd , 563 U.S. 731, 742 (2011); Wilson v. Layne , 526 U.S. 603, 617 (1999). Opinion of the Court

Our decisions in Marsh and Thomas —which predate all of the Supreme Court cases cited above except for Wilson —contain language suggesting that out-of-circuit authority should not be considered in determining whether a right was clearly established. The language in Marsh , an en banc decision, was limited to the first method of determining clearly established law, a method which is based on prior cases that are on point. See Marsh , 268 F.3d at 1032 n.10 (“When case law is needed to ‘clearly establish’ the law appli- cable to the pertinent circumstances, we look to decisions of the U.S. Supreme Court, the United States Court of Appeals for the Eleventh Circuit, and the highest court of the pertinent state.”). The language in Thomas , however, was looser and potentially broader:

Plaintiffs insist that “consensus or persuasive author- ity” from other circuits may create clearly established law. Plaintiffs then direct us to six opinions from other circuits that deal with strip searches. As we have stated, only Supreme Court cases, Eleventh Circuit caselaw, and Georgia Supreme Court caselaw can “clearly establish” law in this circuit. In Marsh . . . we implicitly reaffirmed that position when we stated that we do not understand [ Wilson v. Layne , 526 U.S. 603 (1999)] to have held that a consensus of cases of persuasive authority would be able to establish law clearly.”

Thomas , 323 F.3d at 955 (citations and footnote omitted). Despite the language they contain, Marsh and Thomas do not hold, and should not be read to suggest, that cases from our sister circuits Opinion of the Court cannot be considered in determining whether a constitutional vio- lation was one of “obvious clarity” for purposes of qualified im- munity.

In fact, at least one of our cases points in the opposite direc- tion. In Chandler v. Baird , 926 F.2d 1057 (11th Cir. 1991), which con- cerned an Eighth Amendment/deliberate indifference claim in- volving prison conditions, we relied in part on decisions issued by our sister circuits in determining both (1) that the plaintiff had pre- sented sufficient evidence to withstand summary judgment and (2) that the correctional officers who had been sued were not entitled to qualified immunity. The inmate in Baird had testified that he had been detained in a “cold cell [with the temperature at 60 degrees] with no clothes except undershorts and with a plastic-covered mat- tress without bedding; [there was] filth on the cell’s floor and walls; [he was] depriv[ed] of toilet paper for three days; [he was] depriv[ed] of running water for two days; [there was a] lack of soap, toothbrush, toothpaste, and linen; and . . . the cell [had previously been occupied] by an inmate afflicted with an HIV virus.” Id. at 1063. He explained that he “slept on the floor [of his cell] and on occasion huddled with a roommate, sleeping between two mat- tresses.” Id.

In reversing the district court’s grant of summary judgment in favor of the officers, we noted in part that “[o]ther circuits ha[d] for some time recognized the temperature factor in assessing con- ditions of confinement.” Id. at 1064. Citing and discussing cases from the Second, Fourth, Seventh, Eighth, and Tenth Circuits, we Opinion of the Court held that the evidence, viewed in the light most favorable to the plaintiff, allowed a jury to find an Eighth Amendment violation: “We conclude from this body of caselaw that plaintiff is entitled to have the trier of fact determine whether the conditions of his ad- ministrative confinement, principally with regard to the cell tem- perature and the provision of hygiene items, violated the minimal standards required by the Eighth Amendment.” Id. at 1065. We then held, without further explanation, that the officers were not entitled to qualified immunity: “We also conclude, although the district court did not reach the issue, that the right of a prisoner not to be confined in a cell at so low a temperature as to cause severe discomfort and in conditions lacking basic sanitation was well es- tablished in 1986. The defendants therefore were not entitled to summary judgment on the basis of qualified immunity.” Id. at 1065–66.

Baird supports our view that reliance on out-of-circuit au- thorities is permitted in determining whether a violation was of “obvious clarity.” We have explained that the Baird panel, “in ren- dering its judgment on qualified immunity, was concerned entirely with the law related to excessive cold claims,” Chandler v. Crosby , 379 F.3d 1278, 1296 (11th Cir. 2004), and the cases cited in Baird in- volving cold prison cells came from other circuits.

Nevertheless, we recognize that some of our post- Marsh /post- Thomas cases can be read to say (or suggest) that out- of-circuit authority is not relevant under any of the three methods Opinion of the Court for determining whether a right was clearly established. We discuss two such cases as examples.

In Mercado v. City of Orlando , 407 F.3d 1152 (11th Cir. 2005), a Fourth Amendment excessive force case, we indicated that under each of the three methods of determining whether a right was clearly established, out-of-circuit cases cannot be considered:

Mercado can demonstrate that his right was clearly established in a number of ways. First, he can show that a materially similar case has already been de- cided, giving notice to the police. He could also show that a broader, clearly established principle should control the novel facts in this situation. Finally, he could show that this case fits within the exception of conduct which so obviously violates [the] constitu- tion that prior case law is unnecessary. To make this showing, Mercado must point to law as interpreted by the Supreme Court, the Eleventh Circuit, or the Supreme Court of Florida.

Id. at 1158–59 (citations omitted and emphasis added). On its face, this language from Mercado could be construed to preclude consid- eration of cases from other circuits under any of the methods for determining clearly established law, including the “obvious clarity” formulation. This is because the highlighted text seems to apply to all three ways of demonstrating clearly established law.

Terrell v. Smith , 668 F.3d 1244 (11th Cir. 2012), which also in- volved a Fourth Amendment excessive force claim, contains similar language. In that case, we held that the officer’s use of deadly force Opinion of the Court was reasonable, see id. at 1252–55, but then chose to also address the issue of qualified immunity. After setting out the three ways of showing that a right was clearly established—including that a vio- lation was of “obvious clarity”—we made the following general statement: “Under controlling law, the plaintiffs must carry their burden [as to clearly established law] by looking to the law as inter- preted at the time by the United States Supreme Court, the Elev- enth Circuit, or the Florida Supreme Court.” Id. at 1255. And after analyzing the three methods—again including the “obvious clarity” formulation—we concluded with this language: “In short, the clearly established law as interpreted by the United States Supreme Court, this Court, and the Florida Supreme Court would not have given [the officer] fair notice that his actions would violate the Fourth Amendment.” Id. at 1258. Because these two passages in Terrell are linked to all three ways of demonstrating clearly estab- lished law, they can be read as precluding consideration of persua- sive out-of-circuit authority in assessing whether a violation was one of “obvious clarity.” ^[5]

Opinion of the Court Not surprisingly, some district courts in the Eleventh Circuit have interpreted our precedent to bar consideration of out-of-cir- cuit authority in determining clearly established law. And they are not alone; a number of civil rights treatises have read our cases the same way. See, e.g., Stafford v. City of Argo , 514 F. Supp. 3d 1353, 1363 (N.D. Ala. 2021) (“[T]he Supreme Court has stated that in the ab- sence of controlling authority, a robust consensus of persuasive au- thority may provide fair and clear notice particular conduct violates the Constitution. . . . However, the Eleventh Circuit repeatedly has instructed district courts within its bounds that only decisions is- sued by the United States Supreme Court, the Eleventh Circuit it- self, and the highest court of the relevant state may [be consid- ered].”) (citations and internal quotation marks omitted); Jackson v. McCurry , 303 F. Supp. 3d 1367, 1375 n.5 (M.D. Ga. 2017) (explaining that, based on Eleventh Circuit precedent, “the Court does not ad- dress the cases from other circuits and district courts that Plaintiffs submitted to carry their burden of showing that Defendants vio- lated clearly established law”), aff’d , 762 F. App’x 919, 925–26 (11th Cir. 2019) (“A plaintiff may ‘demonstrate that the contours of the right were clearly established in one of three ways.’ First, a plaintiff may establish that ‘a materially similar case has already been de- cided.’ Second, the plaintiff may ‘point to a broader, clearly law can be clearly established for qualified immunity purposes by non-binding precedent.”); Amnesty Int’l, USA v. Battle , 559 F.3d 1170, 1184 (11th Cir. 2009) (“In this Circuit, only the caselaw of the Supreme Court, the Eleventh Circuit or the law of the highest court of the state where the events took place—in this case, Florida—can ‘clearly establish’ constitutional rights.”). Opinion of the Court

established principle that should control the novel facts of the situ- ation.’ Third, ‘the conduct involved in the case may so obviously violate the [C]onstitution that prior case law is unnecessary.’ The precedents that clearly establish law for these purposes are those of the Supreme Court, this Court, and the highest court of the state where the challenged action occurred.”) (citations omitted); 2 Ivan E. Bodensteiner & Rosalie Berger Levinson, State and Local Gov- ernment Civil Rights Liability § 2:8 (May 2024 update) (“The Elev- enth Circuit . . . will not consider case law from other circuits in deciding whether the law was clearly established.”); 2 Sheldon H. Nahmod, Civil Rights & Civil Liberties Litigation: The Law of Sec- tion 1983 § 8:22 n.2 (Sept. 2024 update) (citing Eleventh Circuit cases standing for the proposition that only decisions from the United States Supreme Court, the Eleventh Circuit, and the su- preme court of the state in question can clearly establish law for purposes of qualified immunity); Michael Avery et al., Police Mis- conduct: Law and Litigation § 3.9 n.22 (Dec. 2024 update) (citing an Eleventh Circuit case for the same proposition).

Cases like Mercado and Terrell are not, of course, the sum and substance of our qualified immunity law. Indeed, in some cases we have explained that a plaintiff can make an “obvious clarity” show- ing when there are no similar decisions from the Supreme Court, the Eleventh Circuit, or the relevant state supreme court, thereby indicating that the “obvious clarity” analysis is not limited to these three buckets of caselaw. See, e.g., Dukes v. Eaton , 852 F.3d 1035, 1043 (11th Cir. 2017) (“Because no precedent of the Supreme Court, our Circuit, or the Supreme Court of Georgia has addressed the Opinion of the Court constitutionality of flashbangs, Dukes must establish that ‘a gen- eral constitutional rule already identified in the decisional law . . . appl[ies] with obvious clarity’ to Deaton’s conduct.”) (citation omitted).

We therefore take this opportunity to clarify an aspect of our qualified immunity jurisprudence. Notwithstanding language in cases like Marsh , Thomas Mercado , and Terrell , persuasive decisions from other circuits can be considered in determining whether a vi- olation was one of “obvious clarity” for purposes of qualified im- munity. ^[6]

To those who might suggest that considering the decisions of sister circuits in an “obvious clarity” scenario is practically mean- ingless—because it does not take published caselaw to make certain constitutional violations “obvious”—we point out that it is not al- ways obvious that a violation was one of “obvious clarity.” See, e.g., Corbitt v. Vickers , 929 F.3d 1304, 1321–23, 1324–26 (11th Cir. 2019) (2-1 decision disagreeing about whether a Fourth Amendment vio- lation was one of “obvious clarity”); Coffin v. Brandau , 642 F.3d 999, 1014–18, 1027, 1029–30 (11th Cir. 2011) (en banc) (8-4 decision dis- agreeing about the same issue). In such cases persuasive authority Opinion of the Court 33 from other circuits may prove helpful in determining whether qualified immunity applies.

“[T]he salient question” for us is “whether the state of the law in [February of 2017] gave [Lieutenant Milton and Officer Iri- zarry] fair warning that their alleged treatment of [Ms. Gilmore] was unconstitutional.” Hope , 536 U.S. at 741. Given all of the cir- cumstances set out below, we answer that question affirmatively: the Fourth Amendment violation here, if Ms. Gilmore’s version of events is credited, was one of “obvious clarity.”

First, at the time of the conduct at issue here the seven cir- cuits that had confronted the issue (the First, Second, Fifth, Sixth, Seventh, Eighth, and Tenth Circuits) had all held that a strip search of a prison visitor must be supported by reasonable suspicion. See Wood , 89 F.3d at 928–29; Varrone , 123 F.3d at 79; Thorne , 765 F.2d at 1276; Daugherty , 935 F.2d at 787; Burgess , 201 F.3d at 945; Hunter , 672 F.2d at 674; Romo , 46 F.3d at 1020. The unanimous view of a ma- jority of the regional circuits on the minimum level of suspicion needed for a strip search is a relevant and important consideration in the “obvious clarity” analysis.

Lieutenant Milton and Officer Irizarry did not have reason- able suspicion, much less probable cause, for a strip search. Ms. Gilmore, who had successfully cleared a security screening which involved three different types of searches, did not stare at Officer Lupo while in the visitation room, stared back at Lieutenant Milton only in response, and did not smell of marijuana. Opinion of the Court

Second, Lieutenant Milton and Officer Irizarry coerced Ms. Gilmore to consent to the strip search by threatening her with de- tention. Again, “a search conducted pursuant to a coerced consent is not valid because the consenting party is forced to allow the search.” Rackley , 742 F.2d at 1271. And by telling Ms. Gilmore that she would be strip-searched anyway, Lieutenant Milton and Officer Irizarry did not give her the option to forgo her visit and leave.

Third, Lieutenant Milton did not obtain verbal approval for the strip search from Deputy Warden Smith as required by the reg- ulations of the Georgia Department of Corrections. This fact also weighs in favor of an “obvious clarity” violation. See Hope , 536 U.S. at 744 (considering, as a factor in the “fair notice”/qualified im- munity analysis, the correctional officers’ failure to comply with an Alabama Department of Corrections regulation that required a log whenever a hitching post was used: “A course of conduct that tends to prove that the requirement was merely a sham, or that respond- ents could ignore it with impunity, provides equally strong support for the conclusion that they were fully aware of the wrongful char- acter of their conduct.”).

Fourth, the strip search was clearly (i.e., obviously) unrea- sonable in scope. Lieutenant Milton and Officer Irizarry did not just require Ms. Gilmore to take off her clothes. At Lieutenant Mil- ton’s direction, Officer Irizarry lifted Ms. Gilmore breasts and felt in between her buttocks with her gloved hand. Following this phys- ical touching of intimate body parts, Ms. Gilmore was subjected to a visual body-cavity search for which she had to bend over and Opinion of the Court spread her vagina. These more intrusive aspects implicated addi- tional privacy concerns and required a “greater amount of suspi- cion.” Pino , 729 F.2d at 1359. Cf. D.H. by Dawson v. Clayton Cnty. Sch. Dist. , 830 F.3d 1306, 1318 (11th Cir. 2016) (denying qualified im- munity to an assistant principal who strip-searched a 14-year old student in front of other school officials and some of his peers: “Viewing all reasonable inferences in favor of D.H. [the student], we conclude that a reasonable official in [Assistant Principal] McDowell’s position would not have believed that requiring D.H. to strip down to his fully naked body in front of several of his peers was lawful in light of the clearly established principle that a student strip search, even if justified in its inception, must be ‘reasonably related to the objectives of the search and not excessively intrusive in light of the age and sex of the student and the nature of the in- fraction.’”) (quoting Redding , 557 U.S. at 386).

To recap, based on all the circumstances, the unreasonable- ness of Ms. Gilmore’s strip search under the Fourth Amendment (at its inception and in its scope) would have been obvious to any reasonable correctional officer in February of 2017. There was no suspicion whatsoever; the consent was coerced; there was no op- tion to leave; the strip search was not administratively approved; and the strip search involved physical touching of intimate areas and a visual body-cavity inspection. A reasonable officer would have had clear notice that the search was obviously unconstitu- tional. Opinion of the Court

IV We reverse the district court’s grant of summary judgment to Lieutenant Milton and Officer Irizarry on qualified immunity grounds and remand the case to the panel for further proceedings.

REVERSED AND REMANDED. W ILLIAM P RYOR , C.J., Concurring W ILLIAM P RYOR Chief Judge, joined by L AGOA and T JOFLAT , Cir- cuit Judges, concurring:

Our foremost job is to decide appeals correctly. En banc re- hearing facilitates this task by allowing us to consider each appeal anew from every angle. Starting over may also clarify what must be decided instead of what may be decided.

Here, our review with fresh eyes worked as intended. It turned a difficult question—whether and how to adopt Supreme Court dicta about a “robust consensus of cases of persuasive au- thority”—into a simple one—whether the alleged constitutional vi- olation is one of obvious clarity. And the majority opinion answers that simple question correctly.

Our concurring colleague nevertheless would have us an- swer the difficult question. See Rosenbaum Concurring Op. at 1. She does not deny that the majority opinion decides the simple question correctly. But in her eagerness to confront what may, but not must, be decided, she overlooks a problem—besides those raised by Judge Tjoflat, see Tjoflat Concurring Op. at 20–29—with holding that a “robust consensus” of persuasive authority alone can clearly establish the law. That is, the so-called “robust consensus” might be wrong.

Powell v. Barrett provides an example of this problem. 541 F.3d 1298 (11th Cir. 2008) (en banc). There, we held that jail officers may strip-search arrestees without reasonable suspicion before placing them in a general population of detainees. Id. at 1300, 1314. And we reached that holding despite the unanimous disagreement W ILLIAM P RYOR , C.J., Concurring of our sister circuits. Id. at 1314–16 (Barkett, J., dissenting) (citing nine sister circuits holding that reasonable suspicion was required before guards could strip-search an arrestee). Later, the Supreme Court adopted our view, not the consensus of our sister circuits. See Florence v. Bd. of Chosen Freeholders of Cnty. of Burlington , 566 U.S. 318, 325–26, 330, 339 (2012). In the interim, jail officials in our Cir- cuit were not misled that they were obliged to follow our sister cir- cuits’ robust—and erroneous—consensus.

The lesson from Powell may apply here too, though in an opposite way: by affording prison officers less, not more, deference. Several circuits have held that prison officers need only reasonable suspicion to strip-search a visitor without her consent, see Majority Op. at 12. None have held to the contrary. Yet, I am not convinced that reasonable suspicion, as opposed to probable cause, is the cor- rect standard, especially when the strip-searches of visitors involve body-cavity inspection and touching. Of course, we need not de- cide that question today. Under any standard, the officers’ alleged conduct violated Ms. Gilmore’s constitutional right to be free from an unreasonable seizure. But the majority’s prudent approach—or what our colleague describes as “skirt[ing] the reason we took the case en banc,” see Rosenbaum Concurring Op. at 1—will permit us to decide in a different appeal, if we must, whether probable cause, instead of reasonable suspicion, is the proper standard. We might then again disagree with all our sister circuits, and faced with that split, the Supreme Court might again reject the so-called “robust consensus.” What matters for now is that Ms. Gilmore will get her W ILLIAM P RYOR , C.J., Concurring day in court regardless of the correct answer to the difficult ques- tion that we avoid. R OSENBAUM , J., Concurring in part Circuit Judge, joined by J ILL P RYOR and K IDD , Circuit

Judges, concurring in part and in the judgment:

More than a quarter-century ago, the Supreme Court deter- mined that a “robust consensus of cases of persuasive authority” can clearly establish the law, making qualified immunity inappro- priate. The Court has since repeated this principle at least three times. So it’s unsurprising that every circuit has recognized and accepted this principle to govern their qualified-immunity analyses.

Except us. To be sure, the Supreme Court’s determination that a robust consensus of cases of persuasive authority can clearly establish the law binds us, too. But somehow, we expressly rejected that conclusion in an en banc decision twenty-four years ago. See Marsh v. Butler County , 268 F.3d 1014, 1032 n.10 (11th Cir. 2001) (en banc). Two years later, we doubled down on that position in Thomas ex rel. Thomas v. Roberts , 323 F.3d 950, 955 (11th Cir. 2003). And since then, we have continued acting like this Supreme Court rule doesn’t apply to us.

I thought we might finally come into compliance with Su- preme Court precedent when we voted Plaintiff-Appellant Clarissa Gilmore’s case en banc. I was wrong. Instead, after en banc oral argument on whether a robust consensus of cases can clearly es- tablish the law, we directed the parties to brief two other issues that allow us to skirt the reason we took the case en banc. And now, in a two-step move, we once again exempt ourselves from binding precedent. First, we cabin any “functionality” of the robust-con- sensus-of-persuasive-authority principle to cases of obvious J., Concurring in part clarity—that is, when, by definition, it’s so obvious that conduct violates clearly established law that no precedent (controlling or persuasive) is necessary. And second, we decide not to determine whether a robust consensus of persuasive authority can clearly es- tablish the law by precedent.

In my view, that’s not why we took this case en banc. And it does next to nothing to bring our precedent into compliance with Supreme Court precedent. It also leaves officers uncertain about governing precedent in this Circuit.

This case presents an opportunity to correct our precedent and clarify the law. I would use it to do both. So though I concur in the Court’s ultimate judgment, I write separately to respectfully explain what I think we should have done as an en banc court.

In particular, we granted en banc review to decide these questions:

(1) Should we overrule in part Marsh v. Butler County , 268 F.3d 1014 (11th Cir. 2001) (en banc), and Thomas ex rel. Thomas v. Roberts , 323 F.3d 950 (11th Cir. 2003), so that a robust consensus of cases of persuasive au- thority may clearly establish law for purposes of ab- rogating an officer’s qualified immunity; and (2) if so, did a robust consensus of cases of persuasive authority clearly establish that Plaintiff-Appellant Clarissa Gilmore’s Fourth and Fourteenth Amend- ment rights were violated? J., Concurring in part

As to the first question, the answer is no doubt “yes.” We should overrule Marsh and Thomas to the extent that they hold that a “robust consensus of cases of persuasive authority” can’t clearly establish the law because Supreme Court precedent says that it can. The Tjoflat Concurrence resists this answer. But Supreme Court precedent requires it. No need to take my word for it—as I’ve noted, every other circuit acknowledges that fact. We should, too.

As to the second question, by 2017, at the time Defendants strip searched Gilmore, seven circuits had held that the Fourth Amendment requires reasonable suspicion to strip search a prison visitor, and none had reached the opposite conclusion. Gilmore v. Georgia Dep’t of Corr. , 111 F.4th 1118, 1135 (11th Cir.) (collecting cases), reh’g en banc granted, opinion vacated , 119 F.4th 839 (11th Cir. 2024). By any measure, that’s a robust consensus of cases of per- suasive authority that clearly establishes this principle by prece- dent.

And it slams the door shut on the Tjoflat Concurrence’s in- sistence that it’s impossible to know what comprises a “robust con- sensus.” Indeed, ten of our sister circuits have defined the term. And under each circuit’s definition, seven unanimous cases from the federal courts of appeals amount to a robust consensus of cases of persuasive authority. We should also hold as much.

To be sure, the Tjoflat Concurrence offers a few reasons why we shouldn’t adopt the “robust consensus” standard and apply it in Gilmore’s case. And the William Pryor Concurrence piggybacks J., Concurring in part on the Tjoflat Concurrence, to provide an assist. But the Concur- rences’ reasoning doesn’t withstand scrutiny.

First, the Tjoflat Concurrence says Gilmore’s situation pre- sents a case of obvious clarity, so it’s not necessary to overturn Marsh and Thomas and recognize that a “robust consensus” can clearly establish the law. But using an en banc proceeding that re- quires us to determine whether Defendants are entitled to qualified immunity to announce the governing legal framework for evaluat- ing a qualified-immunity claim is hardly ground-breaking. To the contrary, it’s appropriate. After all, it prevents future defendants who violate clearly established rights, according to a “robust con- sensus of cases of persuasive authority,” from escaping liability for their unconstitutional acts.

In fact, to further precisely this interest, the Supreme Court has authorized lower courts to address the merits of a Section 1983 claim, even when a merits determination is avoidable and unneces- sary because the federal right was not clearly established. See Camreta v. Greene , 563 U.S. 692, 706 (2011). When it comes to qual- ified immunity, we’re supposed to settle the law sooner, rather than later, so that officials don’t “persist[] in the challenged practice” knowing that they “can avoid liability . . . because the law has still not been clearly established.” Id.

Second, the Tjoflat Concurrence argues we shouldn’t adopt the “robust consensus” standard here because doing so wouldn’t have provided Defendants with notice that they would be held lia- ble for conduct that seven other circuits (though not we) had J., Concurring in part unanimously held unconstitutional before Defendants’ challenged actions. But at the same time, the Tjoflat Concurrence declares that the law was clearly established as to the Officers here because it presented a case of obvious clarity. Meanwhile, the William Pryor Concurrence looks at the “robust consensus” standard that the Supreme Court has repeatedly invoked and exactly every other Circuit has adopted, and somehow it concludes that whether to adopt that standard presents a “difficult question” that we should avoid answering.

My colleagues can’t have it both ways. If this is a case of obvious clarity, it offers the perfect opportunity to announce the correct legal framework for determining whether a right is clearly established going forward. Defendants had fair notice because of the obvious clarity. And at the same time, no future plaintiff with a plausible claim would have to endure officers’ violations of their rights with impunity because we hadn’t yet adopted the “robust consensus” standard.

Finally, the Tjoflat and William Pryor Concurrences argue I’ve “overlook[ed]” the fact that the “robust consensus” rule of the other circuits might be wrong. Pryor Conc. Op. at 1; see also Tjoflat Conc. Op. at 24–25, 27–28. And the Pryor Concurrence supposes that the “robust consensus” rule here might be wrong because it seems to have coalesced around a reasonable-suspicion standard, not a probable-cause one. See Pryor Conc. Op. at 2. Then the Pryor Concurrence seems to suggest that if we adopted the “robust con- sensus” standard, plaintiffs (including Gilmore) and would-be J., Concurring in part plaintiffs would fare worse in cases when we adopt a stricter rule than our sister circuits do.

Not so. Under the “robust consensus” standard, if we think the “robust consensus” rule that officers need only reasonable sus- picion to perform the intrusive search here is wrong, then we can reach a different conclusion. We can instead hold that probable cause is necessary.

And if we adopted the “robust consensus” standard but did not adopt a “robust consensus” rule (here, that an officer must have reasonable suspicion (instead of probable cause) to conduct an in- vasive search of a prison visitor), the officers’ pre-suit conduct and the plaintiff’s ability to recover in the case of first impression would be the very same as if we didn’t adopt the “robust consensus” standard.

In both situations (if we adopted the “robust consensus” standard and if we didn’t), if officers had reasonable suspicion, but lacked probable cause, then they’d be entitled to qualified immun- ity because no decisions, not even those of our sister circuits, would have yet informed the officers that their conduct was unconstitu- tional. And in both situations, the unanimity of our sister circuits in adopting a rule that we ultimately didn’t, would show that the officer was reasonable in acting as he did. So rejecting our sister circuits’ rule would have precisely no effect on parties in the case before us if we adopted a stricter rule than our sister circuits. Nor would it affect any conceivable future case. But adopting the “ro- bust consensus” standard would protect people’s rights and J., Concurring in part authorize liability against officers who violate the “robust consen- sus” rule when we adopt the same rule as the other circuits. Again, it’s for this very reason the Supreme Court has allowed lower courts to settle constitutional issues that “ may be decided” before they technically “ must be decided,” id. at 1, so officers don’t violate the Constitution merely because we have not yet clearly established the law, see Greene , 563 U.S. at 706.

I break my discussion into three parts. Part I summarizes why the Court’s holding today does not accomplish much. Part II explains why we should hold that a “robust consensus of cases of persuasive authority” can clearly establish the law by precedent. And Part III shows that a “robust consensus of cases of persuasive authority” clearly establishes by precedent that the Fourth Amend- ment requires reasonable suspicion to strip search a prison visitor.

I. The Court’s holding today makes no effective differ- ence in our qualified-immunity jurisprudence.

Qualified immunity attempts to balance “the need to hold public officials accountable when they exercise power irresponsibly and the need to shield officials from harassment, distraction, and liability when they perform their duties reasonably.” Pearson v. Cal- lahan, 555 U.S. 223, 231 (2009). To accomplish these dual goals, the doctrine protects government officials engaged in discretionary functions and sued in their individual capacities unless they violate “clearly established federal statutory or constitutional rights of which a reasonable person would have known.” Keating v. City of Miami , 598 F.3d 753, 762 (11th Cir. 2010). The “clearly established” J., Concurring in part requirement shields from liability “all but the plainly incompetent or one who is knowingly violating the federal law.” Lee v. Ferraro , 284 F.3d 1188, 1194 (11th Cir. 2002) (citation omitted).

Three limitations on the legal authorities that plaintiffs may use to advance their claims ensure that the governing law is, in fact, “clearly established” before a plaintiff may overcome a qualified- immunity defense. See Gervin v. Florence , 139 F.4th 1236, 1260–61 (11th Cir. 2025).

First, we limit the substance of the law a plaintiff may use. She “must point to (1) ‘case law with indistinguishable facts,’ (2) ‘a broad statement of principle within the Constitution, statute, or case law,’ or (3) ‘conduct so egregious that a constitutional right was clearly violated, even in the total absence of case law.’” Crocker v. Beatty , 995 F.3d 1232, 1240 (11th Cir. 2021) (quoting Lewis v. City of West Palm Beach , 561 F.3d 1288, 1291–92 (11th Cir. 2009)). We sometimes refer to the third category as one of “obvious clarity”— that is, “the conduct at issue so obviously violated the Constitution that prior case law is unnecessary.” JW ex rel. Williams v. Birmingham Bd. of Educ. , 904 F.3d 1248, 1260 (11th Cir. 2018).

Second, we limit the timing of the relevant case law: plain- tiffs may rely on only the case law issued at the time of the official’s act, not on law that developed later. Harlow v. Fitzgerald , 457 U.S. 800, 818 (1982). “If objective observers cannot predict—at the time the official acts—whether the act was lawful or not, . . . the official deserves immunity from liability for civil damages.” Foy v. Holston , 94 F.3d 1528, 1534 (11th Cir. 1996). J., Concurring in part

And third, we limit the jurisdictions from which a plaintiff may identify applicable law: the plaintiff must point “to binding de- cisions of the Supreme Court of the United States, this Court, [or] the highest court of the relevant state.” Glasscox v. City of Argo , 903 F.3d 1207, 1217 (11th Cir. 2018). Other jurisdictions’ precedent can- not clearly establish the law in our Circuit. See Marsh , 268 F.3d at 1032 n.10; Thomas , 323 F.3d at 955.

To be sure, the Court today holds that we may consider a “robust consensus of cases of persuasive authority” to determine whether the challenged conduct falls within the substantive obvi- ous-clarity category. But most respectfully, it’s hard to see why a “robust consensus of cases of persuasive authority” would ever be necessary in determining whether “the conduct at issue so obvi- ously violated the Constitution that prior case law is unnecessary .” JW , 904 F.3d at 1260 (emphasis added). And so, to me, our holding today feels performative, not substantive.

II. We should overrule in part Marsh and Thomas so that “a robust consensus of cases of persuasive authority” may clearly establish law for purposes of abrogating an officer’s qualified immunity.

This Part reviews the Supreme Court’s precedent on “a ro- bust consensus of cases of persuasive authority,” shows how our precedent fails to follow that line of cases, and explains that we are a true outlier among the circuits. J., Concurring in part

I begin with Wilson v. Layne , 526 U.S. 603 (1999). That is the first case when the Court said that a “consensus of cases of persua- sive authority” can clearly establish the law.

There, the Court considered whether police officers violated the Fourth Amendment when, in executing an arrest warrant in a private home, they invited representatives of the media to accom- pany them. Id. at 605. Although the Court concluded that the me- dia ride-along violated the Fourth Amendment, it held that the de- fendant officers were entitled to qualified immunity. Id. at 606. At the time of the media ride-along—which had occurred in the Fourth Circuit—only the Sixth Circuit had addressed a materially similar question. Id. at 616 (citing Bills v. Aseltine , 958 F.2d 697 (6th Cir. 1992)). The Sixth Circuit had reached essentially the same con- clusion in Bills as did the Court in Wilson . That is, the Sixth Circuit had held that material issues of fact precluded summary judgment on whether the police exceeded the scope of a search warrant by allowing a private security guard to participate in the search to identify stolen property other than that described in the warrant. Id. at 616–17.

But the Court held that the single case from the Sixth Circuit could not clearly establish the constitutional violation at issue in Wilson in the Fourth Circuit. As the Supreme Court explained, the Sixth Circuit’s opinion was not “controlling authority in [the Fourth Circuit’s] jurisdiction.” Id. at 617. And although the case was of persuasive value, a lone case is not “a consensus of cases of persuasive authority such that a reasonable officer could not have J., Concurring in part believed that his actions were lawful.” Id. Because the law was “undeveloped,” the officers could not “have been ‘expected to pre- dict the future course of constitutional law.’” Id. at 618 (quoting Procunier v. Navarette , 434 U.S. 555, 562 (1978)). So the Court held that the lower courts appropriately found qualified immunity.

Since then, the Supreme Court has reaffirmed and further refined the “consensus of cases of persuasive authority” principle.

In Aschroft v. al-Kidd , the Court considered “whether a for- mer Attorney General enjoys immunity from suit for allegedly au- thorizing federal prosecutors to obtain valid material-witness war- rants for detention of terrorism suspects whom they would other- wise lack probable cause to arrest.” 563 U.S. 731, 733 (2011). The Court concluded he did. As it explained, “not a single judicial opin- ion had held that pretext could render an objectively reasonable ar- rest pursuant to a material-witness warrant unconstitutional.” Id. at 741.

Although a district-court decision supported the plaintiff’s position, the Court rejected that case as sufficient to clearly estab- lish law. As Justice Scalia put it, “a district judge’s ipse dixit of a holding is not ‘controlling authority’ in any jurisdiction, much less in the entire United States; and his ipse dixit of a footnoted dictum falls far short of what is necessary absent controlling authority: a robust ‘consensus of cases of persuasive authority .’” al-Kidd , 563 U.S. at 741– 42 (emphasis added) (quoting Wilson , 526 U.S. at 617).

Up next, we have Plumhoff v. Rickard . There, the Court ex- plained that to “defeat immunity” the plaintiff had to show “at a J., Concurring in part minimum” that “either controlling authority or a robust consensus of cases of persuasive authority ” established the asserted constitutional violation. 572 U.S. 765, 780 (2014) (cleaned up) (emphasis added).

To be sure, in two 2015 cases, the Court appeared to qualify its prior decisions. In City and County of San Francisco v. Sheehan , for instance, the Court quipped that “to the extent that robust consen- sus of cases of persuasive authority could itself clearly establish the federal right respondent alleges, no such consensus exists here.” 575 U.S. 600, 617 (2015) (cleaned up). And it repeated its “to the extent” qualification in Taylor v. Barkes . 575 U.S. 822, 826 (2015) (per curiam) (quoting Sheehan , 575 U.S. at 826).

But the Court has since unqualifiedly confirmed that, by it- self, a “robust consensus of cases of persuasive authority” can clearly establish the law. In District of Columbia v. Wesby , Justice Thomas, writing for the Court, explained that, to abrogate quali- fied immunity, the relevant legal rule “must be settled law, which means it is dictated by controlling authority or a robust consensus of cases of persuasive authority .” 583 U.S. 48, 63 (2018) (cleaned up) (emphasis added). And in criticizing the reasoning that led the lower court to abrogate qualified immunity, the Court said that “neither the panel majority nor the [plaintiffs] have identified a sin- gle precedent—much less a controlling case or robust consensus of J., Concurring in part cases—finding a Fourth Amendment violation under similar cir- cumstances.” Id. at 65 (cleaned up) (alteration added). ^[1] J., Concurring in part In short, over the last several years, the Supreme Court has established and repeatedly confirmed that a “robust consensus of cases of persuasive authority” can clearly establish the law.

The State of Georgia resists this fact. It characterizes the Supreme Court’s repeated statements of the “robust consensus of cases of persuasive authority” as “unexplained dicta.” The Tjoflat Concurrence chimes in, too, asserting that the Supreme Court “has never held that persuasive precedent can clearly establish the law.” Tjoflat Conc. Op. at 18.

He and the State misunderstand. Of course, we are not bound by every thought or musing in the United States Reports. Only the Supreme Court’s “holdings” bind us. ^[2] Andrew v. White , “persuasive authority” to mean only binding precedent from the immediate appellate court, not out-of-circuit authority. Id. But that musing runs head- long into Wilson , which considered the Sixth Circuit case the plaintiff proffered to be persuasive authority, even though it was not a Fourth Circuit case. See 526 U.S. at 616–17 (explaining the proffered Sixth Circuit authority was neither “controlling authority [in the plaintiffs’] jurisdiction” nor part of “a consensus of cases of persuasive authority” (emphasis added)). Plus, even if we assumed in-circuit precedent could never qualify as “controlling authority”—a position contrary to every circuit’s current application of Supreme Court precedent— the notion that a “robust consensus of cases” of in-circuit precedent would be necessary to clearly establish the law within a single circuit is nonsensical, given the prior-panel-precedent rule, which binds every panel to a prior panel’s holding unless and until that the Supreme Court or our en banc Court abrogates it. Smith v. GTE Corp. , 236 F.3d 1292, 1300 n.8 (11th Cir. 2001). ^[2] Still, we have emphasized the importance of Supreme Court dicta. See , e.g. , Schwab v. Crosby , 451 F.3d 1308, 1325 (11th Cir. 2006) (“[T]here is dicta and then there is dicta, and then there is Supreme Court dicta. . . . We have R J., Concurring in part 145 S. Ct. 75, 82 (2025); Royal Canin U.S.A., Inc. v. Wullschleger , 604 U.S. 22, 39 (2025).

But the Court has unambiguously told us that when it uses “a legal rule or principle to decide a case, that principle is a ‘holding’ of the Court.” Andrew , 145 S. Ct. at 81. ^[3] That’s so, even if the legal principle it announces isn’t “strictly necessary to a court’s judg- ment,” like when we choose between “two competing legal ‘tests.’” United States v. Files , 63 F.4th 920, 927–28 (11th Cir.), cert. denied , 144 S. Ct. 419 (2023). We consider that choice binding “even when it’s not clear that the case would have turned out differently under the other” test. Id. at 928.

After all, “no one thinks that when we do state a governing rule—as we typically do—we do so gratuitously and unneces- sarily.” Id. at 928 & n.5 (citing Michael Abramowicz & Maxwell Stearns, Defining Dicta , 57 S TAN . L. EV . 953, 984–86 (2005)). In previously recognized that ‘dicta from the Supreme Court is not something to be lightly cast aside.’” (citations omitted)).

[3] The Tjoflat Concurrence argues that the Supreme Court limited Andrew ’s rule about what constitutes a holding to the AEDPA context. Tjoflat Conc. Op. at 18 n.1. But its argument conflicts with our own precedent, which ap- plies Andrew ’s understanding of a holding to all contexts. See Files , 63 F.4th at 928 (confirming “statements of a legal rule” are holdings). Plus, even if we weren’t bound by our precedent, Andrew itself cited qualified-immunity cases to explain why the appellate court erred in refusing to apply the applicable legal principle, confirming that its holding-dicta distinction applies with equal force here. See Andrew , 145 S. Ct. at 82 (citing Taylor v. Riojas , 592 U.S. 7, 8 (2020), and Hope v. Pelzer , 536 U.S. 730 (2002), in a discussion about when gen- eral legal principles may clearly establish the asserted violation). J., Concurring in part

other words, declaring the minimum threshold that a plaintiff must meet to prevail on a claim does not somehow turn the Court into a “roving commission” attempting to “publicly opine on every legal question.” TransUnion LLC v. Ramirez , 594 U.S. 413, 423 (2021). Ra- ther, it establishes a governing principle to resolve the “question actually before the Court.” Cohens v. Virginia , 19 U.S. (6 Wheat.) 264, 399 (1821) (Marshall, C.J.).

The “robust consensus” standard is exactly that: a legal rule that the Supreme Court has announced in setting the minimum threshold for a plaintiff to succeed in a Section 1983 claim. Indeed, the Court keeps invoking it in resolving Section 1983 cases precisely because it is a governing principle. Plaintiffs may rely on a “robust consensus of cases of persuasive authority” to “defeat immunity,” Plumhoff , 572 U.S. at 780, because it is “what is necessary absent” other case law that is sufficient to defeat immunity—“controlling authority,” al-Kidd , 563 U.S. at 741–42.

As every other circuits’ adoption of the “robust consensus” standard shows, the Tjoflat Concurrence’s attempt to avoid the binding nature of the Supreme Court’s announcement of the “ro- bust consensus” standard is unpersuasive. See Tjoflat Conc. Op. at 9–20.

Under the Tjoflat Concurrence’s logic, a legal rule can be- come a holding only when a plaintiff prevails under it. The Tjoflat Concurrence would conclude that judgments for defendants can’t establish legal principles because a court could always alter the J., Concurring in part relevant threshold in a future case. But that’s not how precedent works.

An example proves the point. No court or litigant would say Strickland v. Washington did not announce the legal rule governing ineffective-assistance-of-counsel claims, even though, in that case, Strickland failed “to make the required showing of either deficient performance or sufficient prejudice.” 466 U.S. 668, 700 (1984). In fact, both the Supreme Court and we have described Strickland ’s standard as a holding and as binding precedent. See, e.g. , Woodford v. Visciotti , 537 U.S. 19, 22 (2002) (describing Strickland ’s holding); Calhoun v. Warden, Baldwin State Prison , 92 F.4th 1338, 1347 (11th Cir.) (same), cert. denied , 145 S. Ct. 443 (2024).

The Supreme Court’s articulation of the “robust consensus” standard in Wilson is a holding in the same way Strickland ’s test is. Wilson held that a single out-of-circuit case can’t clearly establish the law. 526 U.S. at 617. Had one, persuasive, and on-point case been sufficient, the Wilson plaintiff could have succeeded. So by clarifying the evidentiary bar that a plaintiff must meet to succeed on Section 1983 claims—that is, at a minimum, “a consensus of cases of persuasive authority,” id. —the Court issued a holding that binds us. That’s so despite the Court’s additional rationale that a split of authority among circuits had developed during the litiga- tion of the case. After all, “our precedent treats alternative hold- ings ‘as binding as solitary holdings.’” Files , 63 F.4th at 927 (quoting Bravo v. United States , 532 F.3d 1154, 1162 (11th Cir. 2008)). J., Concurring in part

The Tjoflat Concurrence can’t escape the Supreme Court’s repeated recognition of the “robust consensus” standard by trying to reimagine what the Supreme Court has said. The Tjoflat Con- currence claims, for instance, that in al-Kidd , the Supreme Court used the phase about a “robust consensus of cases of persuasive authority” merely “descriptively—to underscore how far the plain- tiff’s argument fell short.” Tjoflat Conc. Op. at 13.

The Supreme Court did no such thing. Here’s what the Court said: “Even a district judge’s ipse dixit of a holding is not ‘con- trolling authority in any jurisdiction, much less in the entire United States; and his ipse dixit of a footnoted dictum falls far short of what is necessary absent controlling authority: a robust ‘consensus of cases of persuasive authority .’” al-Kidd , 563 U.S. at 74142 (emphasis added) (quoting Wilson , 526 U.S. at 617). There is simply no accurate way to read these statements—whether in isolation or in the context of the rest of the opinion—to mean anything other than that a plain- tiff can precedentially show the law is clearly established in either of two ways: with “controlling authority” or with “a robust con- sensus of cases of persuasive authority.” A “robust consensus of cases of persuasive authority” is no more “descriptive” than is “con- trolling authority.” They’re both governing legal standards. And no amount of reimagining or naysaying changes that.

So Judge Tjoflat’s conjecture that the Supreme Court an- nounced the “robust consensus” standard as something that “ might ” equate to settled law, Tjoflat Conc. Op. at 16, finds support only in Sheehan ’s and Barkes ’s qualifiers. But Wesby dropped that J., Concurring in part qualifier, and Sheehan and Barkes cannot bear their weight when we look at the Court’s qualified-immunity precedents in full view. In- deed, the Court long ago explained that the “disparate decisions in various Circuits” can offer a “warning” to officials that “is fair enough” to impose liability. United States v. Lanier , 520 U.S. 259, 269 (1997) (rejecting the Sixth Circuit’s rule that only Supreme Court decisions can provide fair notice to officials of constitutional viola- tions).

That’s why every single one of our eleven geographically based sister circuits—the First, Second, Third, Fourth, Fifth, Sixth, Seventh, Eighth, Ninth, Tenth, and District of Columbia Circuits— uniformly abides by the principle that a “robust consensus of cases of persuasive authority” can clearly establish the law in qualified- immunity cases. ^[4] See, e.g. , Irish v. Fowler , 979 F.3d 65, 76 (1st Cir. 2020); Sloley v. VanBramer , 945 F.3d 30, 40 (2d Cir. 2019); James v. N.J. State Police , 957 F.3d 165, 170 (3d Cir. 2020); Johnson v. Robinette , 105 F.4th 99, 120–21 (4th Cir. 2024); Lincoln v. Scott , 887 F.3d 190, 197 (5th Cir. 2018); Akima v. Peca , 85 F.4th 416, 423 (6th Cir. 2023); Est. of Davis v. Ortiz , 987 F.3d 635, 638 (7th Cir. 2021); De La Rosa v. White , 852 F.3d 740, 745–46 (8th Cir. 2017); Hopson v. Alexander , 71 J., Concurring in part F.4th 692, 697, 707 (9th Cir. 2023); Lewis v. City of Edmond , 48 F.4th 1193, 1198 (10th Cir. 2022), cert. denied , 143 S. Ct. 1055 (2023); Vasquez v. District of Columbia , 110 F.4th 282, 288 (D.C. Cir. 2024).

Only we do not. In Marsh , we expressly rejected Wilson ’s standard, saying only that we did “not understand Wilson . . . to have held that a ‘consensus of cases of persuasive authority’ from other courts would be able to establish the law clearly.” Marsh , 268 F.3d at 1032 n.10 (quoting Wilson , 526 U.S. at 617). Marsh offered the rationale that, because “splits between jurisdictions on matters of law are not uncommon,” officials should not have “to sort out the law of every jurisdiction in the country.” Id .

But for three reasons, that can’t excuse our failure to abide by Supreme Court precedent.

First, it’s simply not accurate to describe the Court’s Wil- son / al-Kidd / Plumhoff / Wesby “robust consensus of cases of persua- sive authority” standard as requiring officials “to sort out the law of every jurisdiction in the country.” As I explain more in the next section, a “robust consensus” means there’s strong agreement, and the Supreme Court has already told us that district-court decisions can’t clearly establish the law. See al-Kidd , 563 U.S. at 741–42; Camreta , 563 U.S. at 709 n.7 (“[D]istrict court decisions—unlike those from the courts of appeals—do not necessarily settle consti- tutional standards or prevent repeated claims of qualified immun- ity.”).

Second, complying with the “robust consensus” standard hasn’t been a problem for law enforcement in any of our eleven J., Concurring in part sister circuits. And it’s not clear to me why the Eleventh Circuit will experience some unique trouble in applying that same princi- ple. In fact, the Supreme Court rejected these workability concerns decades ago. As the Court has explained, although “disparate de- cisions in various Circuits might leave the law insufficiently certain even on a point widely considered,” it is just a “circumstance” to “be taken into account in deciding whether” the law provides a “warning” to officers that “is fair enough.” Lanier , 520 U.S. at 269.

Plus, even our Circuit’s formulation—that only decisions from the Supreme Court, the Eleventh Circuit, or the highest court of the state can clearly establish the law—has always functioned under a legal fiction. No one really thinks that officers have the time to read, understand the significance of, and be up to date on all governing case law, while still performing their law-enforcement functions. But we indulge the legal fiction that precedent gives of- ficers notice, anyway. We do so because we know that law-enforce- ment departments have legal counsel whose job it is to stay on top of constitutional-law developments that affect officers’ responsibil- ities and to educate those officers about the law. And when a “ro- bust consensus of cases of persuasive authority” coalesces on an issue—no matter whether the cases are inside or outside a law-en- forcement department’s jurisdiction—you can be sure that counsel tracks that. After all, that’s counsel’s job. Indeed, conferences and seminars are regularly held on national developments in the law.

And third, we’re talking about binding Supreme Court prec- edent. Compliance isn’t optional and we shouldn’t treat it that way. J., Concurring in part

So Marsh and its reasoning fail. Thomas is no better. It just doubled down on Marsh ’s errors by invoking our prior-panel prec- edent rule: Thomas stated simply that Marsh “implicitly reaffirmed” our pre- Wilson position so that any “argument based upon deci- sions in other circuits [was] foreclosed by our precedent.” 323 F.3d at 955 (citing 268 F.3d at 1032 n.10). Marsh and Thomas offer no reason to avoid adopting binding Supreme Court precedent. And I would abrogate them and hold, as Supreme Court precedent re- quires, that a “robust consensus of cases of persuasive authority” can clearly establish the law.

III. A “robust consensus of cases of persuasive authority” clearly established that Plaintiff-Appellant Clarissa Gilmore’s Fourth and Fourteenth Amendment rights were violated.

So what does a “robust consensus of cases of persuasive au- thority” mean? This Part first defines the standard through the Su- preme Court’s guidance, the practice of our sister circuits, and the principle of fair notice that guides every qualified-immunity in- quiry. Then this Part applies the “robust consensus” standard to Gilmore’s claims. Gilmore’s claim meets any reasonable definition of a robust consensus of cases of persuasive authority, so Defend- ants do not enjoy qualified immunity.

A. If several unanimous decisions from federal appellate courts agree on a legal rule, they make up a “robust consensus of cases of persuasive authority.” J., Concurring in part

Luckily, we don’t need to write on a blank slate when we determine the meaning of a “robust consensus of cases of persua- sive authority.” Contrary to the Tjoflat Concurrence’s suggestion, this isn’t an area of law where “nobody knows” how to apply the controlling principle. Tjoflat Conc. Op. at 20. ^[5] We’re not J., Concurring in part attempting to discern the meaning of life or design a spaceship to take us to other galaxies. We’re talking about applying a legal standard—“robust consensus of cases of persuasive authority”— that our sister circuits have applied for years and that the decisions of the Supreme Court help explain. Indeed, the Supreme Court Court’s decisions and those of our sister circuits offer a workable framework for courts and litigants to employ in applying the “ro- bust consensus” standard.

I begin with Supreme Court precedent. Two firm rules fol- low from Supreme Court holdings.

First, out-of-circuit federal appellate decisions can clearly es- tablish the law. The Court told us as much in Wilson . See 526 U.S. at 617. It’s just that a single out-of-circuit decision standing alone can’t do so. See id. at 616–17. And that makes sense. After all, how could a single case standing alone ever satisfy the definition of a “robust consensus”? ^[6]

Second, the persuasive authorities generally must be unani- mous in their view. Take Wilson , for instance. There, the Court found the law wasn’t clearly established. In reaching this conclu- sion, the Court found significant that “[b]etween the time of the events of th[e] case and [the Court’s] decision, a split among the , J., Concurring in part Federal Circuits in fact developed on the question whether media ride-alongs that enter homes” constitute a violation of the Fourth Amendment. Id. at 618. “If judges thus disagree on a constitu- tional question,” the Court explained, “it is unfair to subject police to money damages for picking the losing side of the controversy.” Id. ; see also Pearson 555 U.S. at 244–45 (holding officers were entitled to rely on a doctrine accepted by “three Federal Courts of Appeals and two State Supreme Courts,” even though “their own Federal Circuit had not yet ruled on” it). So unless no one could reasonably believe that we would follow the minority approach, we should not consider an alleged constitutional violation to be clearly established when the federal courts of appeals and state supreme courts split in their views. See Fowler , 979 F.3d at 78.

But beyond these core holdings, the Supreme Court has not fleshed out the scope of the “robust consensus” standard. Nor has the Court identified the minimum amount of persuasive authority that a plaintiff must muster to show that reasonable state officials had fair notice of the alleged constitutional violation.

That’s where our sister circuits come in. Over the past sev- eral decades, our sister circuits have developed some case law on what amounts to a “robust consensus of cases of persuasive au- thority.”

To start, all our sister circuits agree that “[a] robust consen- sus does not require the express agreement of every circuit.” Id. at 76. That is, the “robust consensus” standard does not benefit only the plaintiff who is last in line. , J., Concurring in part

As for the minimum number of cases necessary to amount to a “robust consensus,” our sister circuits have offered a narrow range of answers. At the lower end of the spectrum, the Seventh and Eighth Circuits have determined that precedential decisions from as few as two sister circuits can clearly establish the law. See, e.g. Fis v. Dawson , 829 F.3d 895, 906 (7th Cir. 2016) (discussing three combined decisions from two circuits); Hayes v. Long , 72 F.3d 70, 74 (8th Cir. 1995) (relying on decisions from two circuits and one district court).

But two are not enough in other circuits. At least four cir- cuits have found a “robust consensus” when three other circuits have unanimously agreed in precedential decisions. See , e.g. , Mal- donado v. Fontanes , 568 F.3d 263, 270–71 (1st Cir. 2009); Varrone v. Bilotti , 123 F.3d 75, 79 (2d Cir. 1997); Williams v. Bitner , 455 F.3d 186, 193 (3d Cir. 2006); Daugherty v. Campbell , 935 F.2d 780, 787 (6th Cir. 1991). A fifth—the Ninth Circuit—has described the agreement of three other circuits’ precedential decisions as a “robust consensus” in a case where it also concluded that the Ninth Circuit’s own prec- edent clearly established the law. Tuuamelemalo v. Greene , 946 F.3d 471, 477–78 (9th Cir. 2019).

Three other circuits that have applied the “robust consen- sus” test have done so when more circuits have unanimously agreed on the law. But two of these circuits have not opined on whether the agreement of fewer circuits in precedential decisions could create a “robust consensus.” J., Concurring in part

The Fourth Circuit, for instance, found a “clear consensus of persuasive authority” when five sister circuits had precedential de- cisions on point. Williamson v. Stirling , 912 F.3d 154, 188 (4th Cir. 2018). But the court didn’t hold that a “robust consensus” couldn’t exist with fewer than five. Similarly, the Fifth Circuit determined a “robust consensus of persuasive authority existed” when four cir- cuits had precedential decisions on the issue and two other circuits had non-precedential decisions. Johnson v. Halstead , 916 F.3d 410, 420 (5th Cir. 2019). Like the Fourth Circuit, the Fifth Circuit did not opine on whether fewer decisions could comprise a “robust consensus.”

As for the Tenth Circuit, it has said that “the weight of au- thority from other circuits may clearly establish the law when at least six other circuits have recognized the right at issue.” Irizarry v. Yehia , 38 F.4th 1282, 1284 (10th Cir. 2022). Even so, the Tenth Circuit’s holding is somewhat nuanced. Four of the six other cir- cuit decisions established the relevant right on “facts materially similar” to those at issue in the Tenth Circuit’s case, and the remain- ing two opinions expressed a relevant broad principle. See id. at 1294–95.

To sum up, then, ten circuits have applied the “robust con- sensus” standard. Even when we view those decisions to apply the standard most stringently, every one of those circuits has concluded that the precedential decisions of fewer than seven other circuits can make up a “robust consensus.” In other words, a robust—in- deed super-“robust”—consensus of circuits agrees that, at the very J., Concurring in part most, six precedential decisions are necessary for a “robust consen- sus.”

That said, some very good reasons support a lower threshold than six. Given the seven-circuit weight of authority that abides by a three- (or two-) circuit consensus, I focus my comments on that number.

On the one hand, a three-circuit precedential-decision re- quirement ensures plaintiffs may successfully vindicate their con- stitutional rights. Requiring too many appellate courts to address an issue before concluding that a “robust consensus” exists would fail to “hold public officials accountable when they exercise power irresponsibly.” Pearson , 555 U.S. at 231. After all, it is “rare[]” to find an “overwhelming consensus of authority” from “nearly every court of appeals” that recognizes the challenged “conduct is viola- tive of a constitutional right.” Booker v. S.C. Dep’t of Corr. , 855 F.3d 533, 545 (4th Cir. 2017). A high threshold, then, allows an official to escape “liability for unlawful conduct due to the fortuity that a court in a particular jurisdiction had not yet had the opportunity to address the issue.” Cleveland-Perdue v. Brutsche , 881 F.2d 427, 431 (7th Cir. 1989). Congress had no such intention when it enacted Section 1983, and Harlow did not plan to “provide . . . license to lawless conduct.” 457 U.S. at 819.

On the other hand, concluding a “robust consensus of cases of persuasive authority” exists when at least three federal appellate courts align on an issue in precedential decisions appropriately “shield[s] officials from harassment, distraction, and liability when , J., Concurring in part they perform their duties reasonably.” Pearson , 555 U.S. at 231. Three decisions can constitute “a clear trend in the caselaw” that gives “fair assurance that the recognition of the right by a control- ling precedent was merely a question of time.” Brutsche , 881 F.2d at 431 (relying on three federal appellate decisions to conclude de- fendants may have violated clearly established law); cf. I AN F LEMING G OLDFINGER 166 (Penguin Books 2002) (“Mr. Bond, they have a saying in Chicago: Once is happenstance. Twice is coinci- dence. The third time it’s enemy action.”).

Plus, as I’ve noted, even if several courts perfectly align on a federal issue, contrary persuasive authority may suggest the law is still sufficiently “undeveloped” that it would be “unfair to subject police to money damages for picking the losing side of the contro- versy.” Wilson , 526 U.S at 617–18; Pearson , 555 U.S. at 244–45.

For instance, defendant officials may benefit from divisions in the case law that develop after the relevant conduct in a dispute. See Wilson , 526 U.S at 617–18. Or something unique about our own precedent may make it reasonable for an officer to believe we would not follow the existing “robust consensus.” Compare Terebesi v. Torreso , 764 F.3d 217, 233 (2d Cir. 2014) (discounting persuasive authority from two other circuits because an in-circuit unpublished opinion supported the officer’s position), with Fowler , 979 F.3d at 78 (holding, based on in-circuit precedent, that “defendants could not reasonably have believed that we would” follow the minority rule). So the Supreme Court’s stringent clarity and consensus J., Concurring in part requirements create a backstop against liability for reasonable state officers.

Put simply, when three federal appellate holdings clearly “apply . . . to the specific conduct in question,” Lanier , 520 U.S. at 271, it won’t be “a fortuitous coincidence” that we follow suit, Mar- tin v. Heckler , 773 F.2d 1145, 1153 (11th Cir. 1985) (en banc) (adding that “[h]armony among circuits should be a goal” and that “the law is well served by a court’s attempt to achieve uniformity of deci- sion”), disapproved of on other grounds by Texas State Tchrs. Ass’n v. Garland Indep. Sch. Dist. , 489 U.S. 782 (1989). Persuasive case law will have offered a warning that is “fair enough” to expose defend- ants to monetary remedies for their violation of federal rights. La- nier , 520 U.S. at 269. ^[7]

The Tjoflat Concurrence complains that the “robust consen- sus” standard “severs” qualified immunity “from what the law ac- tually is” by clearly establishing law before we address the pertinent J., Concurring in part legal issue and by causing officers to prophylactically conform to decisions from our sister circuits (the “prophylactic-conformance argument”). Tjoflat Conc. Op. at 25. The William Pryor Concur- rence repeats a similar concern, suggesting officials are “obliged” and “misled” to follow the “robust consensus,” even if we ulti- mately decide it’s wrong. Pryor Conc. Op. at 2. Three, independ- ent points doom this argument.

First, to the extent that the argument assumes officers would respect non-existent constitutional rights, the argument misunder- stands the point of qualified immunity. The doctrine protects indi- viduals who “make reasonable but mistaken judgments.” al-Kidd , 563 U.S. at 743. It is not a license for state officials to push consti- tutional boundaries with impunity. “Where an official could be ex- pected to know that certain conduct would violate statutory or con- stitutional rights, he should be made to hesitate . . . .” Harlow , 457 U.S. at 819 (emphasis added). And that hesitation (that is, prophy- lactic conformance) is warranted when a “robust consensus of cases of persuasive authority” informs officers that they are push- ing constitutional boundaries. See al-Kidd , 563 U.S. at 741–42; Wil- son , 526 U.S. at 617 (explaining a “consensus of cases of persuasive authority” can cause “a reasonable officer” not to “believe[] that his actions were lawful”). To be sure, we may ultimately rule that of- ficial conduct falls on the permissible side of the constitutional line. But if we decide otherwise, the state official had warning that was “fair enough” to impose liability. Lanier , 520 U.S. at 269.

32 , J., Concurring in part

Second, to the extent the argument assumes we would pre- fer to adopt stronger constitutional rules than those of our sister circuits, see Pryor Conc. Op. at 2, the “robust consensus” standard would not affect how we approach qualified immunity in those cases. We would always be free to reject a “robust consensus” of our sister circuits and adopt a stronger constitutional rule or pro- tect additional federal rights.

And doing so would change exactly nothing from how we operate currently, without the “robust consensus” standard. In both situations—either with or without the “robust consensus” standard—if we adopted a stronger constitutional rule than our sis- ter circuits, the officer whose conduct matched the other circuits’ rule would still be entitled to qualified immunity in that case of first impression. After all, even if we’re more protective of a plaintiff’s rights than are our sister circuits, if several of our sister circuits unanimously thought that the officers’ conduct didn’t violate the Constitution, a plaintiff couldn’t establish that no reasonable officer would have engaged in the officer’s conduct there. See Wilson , U.S. at 618; Pearson 555 U.S. at 244–45.

And third, even on the argument’s own logic, it proves too much. The prophylactic-conformance argument’s logic would re- quire us to overturn our current precedent that decisions of the relevant state supreme court can clearly establish the law. If it truly “makes no sense” to consider the law clearly established when it’s possible we may decide otherwise on the merits, Tjoflat Conc. Op. at 25, then it equally “makes no sense” for our qualified-immunity J., Concurring in part precedent to credit state-court decisions. Those don’t bind us when it comes to federal questions. Gallardo ex rel. Vassallo v. Dudek , 963 F.3d 1167, 1180 (11th Cir. 2020), aff’d sub nom. Gallardo ex rel. Vassallo v. Marstiller , 596 U.S. 420 (2022). So under our current governing standard, which allows rulings of the highest state court to clearly establish the law, precisely the same potential for the alleged prophylactic-conformance situation exists.

The highest court of a state could clearly establish a consti- tutional violation before we have reason to consider the issue. So officers in that state would have to prophylactically comply. But we might later disagree with that state supreme court’s view of federal law. And if we did, those officers would have prophylactically re- spected nonexistent constitutional rights. Yet we’ve always consid- ered a state supreme court decision sufficient to clearly establish the law. See Marsh , 268 F.3d at 1032 n.10.

The Tjoflat Concurrence tries to escape this problem by as- serting that I “misunderstand[] both our precedent and how the notice inquiry works.” Tjoflat Conc. Op. at 26. But deflecting doesn’t cure the Tjoflat Concurrence’s problem.

In particular, the Tjoflat Concurrence argues that “officials must conform their conduct to [a state supreme] court’s rulings or risk liability [for federal constitutional violations] in their own state courts.” Id. at 26. And to be sure, a state supreme court decision could bind a state officer that is a state-court defendant. But in the context of Section 1983, that’s not a likely—or even a plausible— risk. A Section 1983 defendant can always remove to a federal J., Concurring in part forum that may not agree with the state court’s reading of federal law. See, e.g. , 28 U.S.C. § 1441(a). And he or she always will in that circumstance. So in reality—not the Tjoflat Concurrence’s hypo- thetical—a state supreme court decision that gets ahead of our precedent is no different than a “robust consensus of cases of per- suasive authority” that also gets ahead of our precedent.

Plus, even as to the potential liability of state officials under state law in state court or administrative proceedings, the Tjoflat Concurrence’s point is illusory. The Tjoflat Concurrence posits that state officers would conform to state law, even if it differs from federal law, because they would face “state tort liability, adminis- trative discipline, or evidentiary exclusion in criminal proceedings” if they didn’t. Tjoflat Conc. Op. at 26–27. But that analysis rests on a faulty premise.

Under state law in all three states in the Eleventh Circuit, official immunity protects state officials’ acts unless they are “per- formed with actual malice or with intent to cause injury.” Dukes v. Deaton , 852 F.3d 1035, 1044 (11th Cir. 2017) (Georgia law); Grider v. City of Auburn , 618 F.3d 1240, 1254–55 (11th Cir. 2010) (noting that, under Alabama law, a state agent is not immune if the plaintiff “show[s] ‘bad intent’”); Baxter v. Roberts , 54 F.4th 1241, 1270 (noting that, under Florida law, an officer “may not be held personally lia- ble . . . unless such officer . . . acted in bad faith or with malicious purpose or in a manner exhibiting wanton and willful disregard of human rights, safety, or property” (emphases omitted) (quoting Fla. Stat. § 768.28(9)(a))). To be clear, “[a]ctual malice means ‘a J., Concurring in part deliberate intention to do wrong, and does not include implied malice, i.e., the reckless disregard for the rights or safety of others’”; an officer must intend “to cause the harm suffered by the plain- tiffs.” Dukes , 852 F.3d at 1045 (quoting Murphy v. Bajjani , 282 Ga. 197, 203 (2007)).

That’s a much higher standard than qualified immunity im- poses. As we’ve explained, “[q]ualified immunity invokes an ob- jective standard; that is, if a reasonable person in the defendant’s place could have acted the same way, the defendant’s subjective intent is irrelevant.” Post v. City of Fort Lauderdale , 7 F.3d 1552, 1558 n.6 (11th Cir. 1993).

The upshot is that, under state law, if an officer doesn’t act with “a deliberate intention to do wrong,” she doesn’t risk liability under state law, even if she violates state rights she should have known about and even if she would be liable under federal law if it were the same as state law. So contrary to the Tjoflat Concur- rence’s suggestion that officers must follow state law even if it dif- fers from federal law at the risk of personal liability, that’s simply not so, as long as the officer doesn’t act with “actual malice.” And to the extent that state law conflicts with a robust consensus of per- suasive authority under federal law, it’s hard to imagine how an officer complying in good faith with federal law could act with “ac- tual malice” under state law.

Apparently aware of this disconnect, the Tjoflat Concur- rence tries to muddy the waters by arguing that it’s simply unfair to hold officers in this Circuit to a “robust consensus of cases of J., Concurring in part persuasive authority” because the standard of “fair notice demands more than persuasive reasoning.” Tjoflat Conc. Op. at 27. But if a “robust consensus of cases of persuasive authority” were the law here like it is in every other circuit , officers would have both legal and institutional reasons to be aware of decisions from other circuits. As I’ve explained, see supra at 20–21, the notion that officers them- selves keep on top of all legal developments without any guidance is a legal fiction. Instead, attorneys regularly provide officers with legal information so they will understand and follow the law. Around the country, lawyers update law-enforcement officers on legal developments not only from their own circuits but also from a “robust consensus of cases of persuasive authority.” It’s just silly to suggest that Eleventh Circuit attorneys would be incapable of similarly educating law-enforcement officers with whom they work and that officers here can’t get fair notice from a “robust con- sensus of cases of persuasive authority” when every other officer in the country can. Simply, if we adopted the “robust consensus” standard, then officers will receive “fair notice.”

The Tjoflat Concurrence’s prophylactic-nonconformance misfire is even more apparent when we consider our panel opin- ions. Panel opinions may clearly establish the law, but an en banc court may always disagree later. No matter whether “[a] published panel decision is binding when issued,” Tjoflat Conc. Op. at 27, if the en banc court vacates it and changes the rule, officers will have prophylactically conformed to the panel decision in the intervening period. And that’s precisely the alleged harm the Tjoflat Concur- rence complains of. J., Concurring in part

Perhaps sensing this problem, the Tjoflat Concurrence tries to move the goal posts from its original stated concern of prophy- lactic conformance. It responds that “[p]anel decisions and state supreme court rulings provide fair notice to officials because they carry legal force at the time of the conduct.” Id. Okay. But that doesn’t address the original “problem” the Tjoflat Concurrence raises of prophylactic conformance—that is, respecting nonexistent rights—when we later disagree with a state supreme court or panel opinion. After all, regardless of notice concerns—which we all agree drive liability under qualified immunity—when we interpret federal law, we necessarily decide what the law “ always meant.” Rivers v. Roadway Exp., Inc. , 511 U.S. 298, 313 n.12 (1994); see id. at 312–13 (“A judicial construction of a statute is an authoritative statement of what the statute meant before as well as after the de- cision of the case giving rise to that construction.”). As a result, when officers follow panel opinions that our en banc Court later overturns, it means they respected rights that never existed in the first place.

So the Tjoflat Concurrence’s arguments are not unique to the “robust consensus” standard. And our precedent already im- plicitly rejects their logic. Prophylactic adherence isn’t qualified immunity’s lodestar; fair notice is. So the law can be clearly estab- lished at the time of the alleged misconduct even if it is eventually disputed.

In short, persuasive case law can and does afford state offic- ers fair notice of violations of federal rights. Still, this dispute J., Concurring in part requires us to consider only whether unanimous decisions from seven of our sister circuits can clearly establish an alleged statutory or constitutional violation. That’s so because, as I discuss in the next section, by 2017, at the time of the relevant conduct, at least seven federal appellate courts had held that prison officials must have reasonable suspicion to strip search a prison visitor. So any reasonable definition of a “robust consensus of cases of persuasive authority” means Gilmore defeats qualified immunity. Cf. Rucho v. Common Cause , 588 U.S. 684, 744 (2019) (Kagan, J., dissenting) (“How about the following for a first-cut answer: This much is too much.”).

B. At the time of Defendants’ alleged misconduct, seven unani- mous decisions from federal appellate courts clearly established that prison officials need reasonable suspicion to strip search a visitor.

Our precedent requires “reasonable suspicion for strip searches of arrestees, students, and border entrants.” Gilmore , 111 F.4th at 1131; see Evans v. Stephens , 407 F.3d 1272, 1279–80 (11th Cir. 2005) (en banc) (arrestees); Cuesta v. Sch. Bd. of Miami-Dade Cnty. , 285 F.3d 962, 969 (11th Cir. 2002) (same); Safford Unified Sch. Dist. No. 1 v. Redding , 557 U.S. 364, 377 (2009) (student); T.R. ex rel. Brock v. Lamar Cnty. Bd. of Educ. , 25 F.4th 877, 882, 855, 888 (11th Cir. 2022) (same); United States v. Alfaro-Moncada , 607 F.3d 720, 729 (11th Cir. 2010) (border entrant); Brent v. Ashley , 247 F.3d 1294, 1302 (11th Cir. 2001) (same). J., Concurring in part

In our (now-vacated) panel opinion, we extended “that rea- sonable-suspicion requirement to searches of prison visitors.” Gil- more , 111 F.4th at 1131. We explained that, as of that time, nine of our sister circuits had already so held. See Wood v. Clemons , 89 F.3d 922, 928–29 (1st Cir. 1996); Varrone , 123 F.3d at 79; Calloway v. Lokey , 948 F.3d 194, 202 (4th Cir. 2020); Thorne v. Jones , 765 F.2d 1270, 1276 (5th Cir. 1985); Daugherty v. Campbell , 935 F.2d 780, 787 (6th Cir. 1991); Burgess v. Lowery , 201 F.3d 942, 945 (7th Cir. 2000); Hunter v. Auger , 672 F.2d 668, 674 (8th Cir. 1982); Cates v. Stroud , 976 F.3d 972, 985 (9th Cir. 2020); Romo v. Champion , 46 F.3d 1013, 1020 (10th Cir. 1995). And crucially, by 2017, when Gilmore’s search occurred, “seven circuits had required reasonable suspicion for a strip search of a prison visitor.” Gilmore , 111 F.4th at 1131.

In short, there’s no reasonable dispute that a “robust consen- sus of cases of persuasive authority” clearly established the legal rule Gilmore now seeks to apply. Each of our sister circuits would reach the same conclusion. In fact, many already have. Take the Sixth Circuit. Thirty years ago, it determined that “the law was clearly established” that “the Fourth Amendment required reason- able suspicion before authorizing a body cavity search,” as “[t]hree circuits had reached this conclusion” already. Spear v. Sowders , 71 F.3d 626, 630 (6th Cir. 1995); see also Varrone , 123 F.3d at 78–79 (con- cluding the law clearly established that officers need reasonable sus- picion to strip search a prison visitor because “three other circuits had established” such a standard); Burgess , 201 F.3d at 945 (“In a long and unbroken series of decisions by our sister circuits . . . it had be- come well established . . . that strip searches of prison visitors were , J., Concurring in part unconstitutional in the absence of reasonable suspicion that the visitor was carrying contraband.”).

So should Gilmore prove that Defendants lacked reasonable suspicion to strip search her, it would not be “unfair to subject” them “to money damages.” Wilson , 526 U.S. at 618.

Nor do Defendants’ arguments that no “robust consensus of cases of persuasive authority” exists here have merit.

First, Defendants suggest that two of the seven pre-2017 cases Gilmore cites cannot provide fair warning to officials because they say the reasonable-suspicion standard those cases announced was dicta. In support of this assertion, Defendants note that the officers in those cases had reasonable suspicion. See, e.g. Varrone , 123 F.3d at 79; Romo , 46 F.3d at 1020. But Defendants are mistaken that these courts’ rulings amount to only dicta.

As I’ve explained, a governing legal rule is a holding, not dicta, even if it is “technically unnecessary to a case’s resolution.” Files , 63 F.4th at 928; see Andrew , 145 S. Ct. at 81. Think again about Strickland ; we treat Strickland as announcing a binding legal rule even though Strickland failed to meet either prong of the Court’s test there. See, e.g. , Woodford , 537 U.S. at 22; Calhoun , 92 F.4th at 1347.

And in Varrone and Romo , the courts had to determine the governing law before they could apply it. That is, to assess whether the defendants’ reasonable suspicion in those cases was enough to make their actions constitutional, those courts had to first establish that reasonable suspicion complies with the Fourth Amendment , J., Concurring in part under the circumstances. So Varrone and Romo announce a binding legal rule that officers must heed. As a result, it’s incorrect to de- scribe their holdings as dicta.

Second, Defendants argue that no consensus exists because the Hawaii Supreme Court purportedly twice upheld strip searches of prison visitors without reasonable suspicion. See, e.g. , State v. Custodio , 62 Haw. 1 (1980); State v. Martinez , 59 Haw. 366 (1978). But no reasonable officer (or officer’s attorney) could rely on Cus- todio or Martinez to show that the reasonable-suspicion rule Gil- more seeks to apply was not clearly established by 2017. See Fowler , 979 F.3d at 78. Both cases relied on the consent of the searched party. Martinez , 59 Haw. at 371; Custodio , 62 Haw. at 5. So even if Custodio and Martinez apply here, Defendants could prevail only if they proved at trial that Gilmore consented. But Gilmore has at- tested that Defendants strip searched her without consent and without reasonable suspicion. See Lee , 284 F.3d at 1190 (construing facts in the non-movant’s favor at summary judgment).

Third, Defendants argue that no “robust consensus” exists because the decisions that Gilmore relies on predate our decision in Powell v. Barrett and the Supreme Court’s decision in Florence v. Board of Chosen Freeholders of County of Burlington . 541 F.3d 1298 (11th Cir. 2008) (en banc); 566 U.S. 318 (2012). In those cases, we, and then the Supreme Court, upheld a policy of strip searching de- tainees—without reasonable suspicion—before admitting them to a jail’s general population. Powell , 541 F.3d at 1300, 1314; Florence 566 U.S. at 339. Defendants suggest a reasonable officer could rely J., Concurring in part on those decisions to conclude prison visitors are in a similar posi- tion to those entering a jail’s general population. And they stress that Florence came after the out-of-circuit precedent Gilmore cites. So, they argue, a reasonable official in 2017 could think that those courts would or should re-evaluate their earlier decisions on visi- tors.

I disagree. To start, no reasonable officer could read Florence to undermine the “robust consensus of cases of persuasive author- ity” requiring reasonable suspicion to strip search a prison visitor. The Supreme Court emphasized that its ruling extended no further than the limited situation before it. Chief Justice Roberts high- lighted how “important” it was that the “Court does not foreclose the possibility of an exception to the rule it announces.” Id. at 340 (Roberts, C.J., concurring). Justice Alito also cautioned that the Court did not bless routine strip searches of arrestees. Id. at 341 (Alito, J., concurring). And Justice Breyer added that “[t]he case is limited to strip searches of those arrestees entering a jail’s general population.” Id. at 342 (Breyer, J., dissenting).

But perhaps most importantly, Justice Kennedy, writing the majority opinion, offered two limitations. He first explained that the plaintiff’s case there did “not require the Court to rule on the types of searches that would be reasonable in instances where . . . situations may diminish the need to conduct some aspects of the searches at issue.” 556 U.S. at 338–39. So the Court refused to ex- tend its ruling to arrestees who may not be entering a jail’s general population. See id. And given that the Court refused to rule that J., Concurring in part jails may always strip search arrestees without reasonable suspi- cion, it’s unreasonable to read Florence as abrogating rulings that officers need reasonable suspicion to strip search visitors . After all, arrestees are jailed upon at least probable cause to believe they committed a crime. But visitors come into a prison with no legal finding against them.

Besides that, the Court also declined to declare constitu- tional invasive “searches that involve the touching of detainees.” Id. at 339. So at a minimum, Florence can’t control Gilmore’s case because Gilmore asserts Defendants manually searched her naked body. See Gilmore , 111 F.4th at 1124–25. As our precedent already emphasized by the time the officers searched Gilmore, searches in- volving physical contact implicate far greater Fourth Amendment concerns than do visual inspections. See United States v. Touset , 890 F.3d 1227, 1234 (11th Cir. 2018) (citing United States v. Vega-Barvo , 729 F.2d 1341, 1345 (11th Cir. 1984)). Defendants were on notice that their alleged conduct violated Gilmore’s constitutional rights because the Supreme Court in Florence “staked out” the same “bright line” as we previously did. Post , 7 F.3d at 1557.

Defendants next argue that some decisions Gilmore cites re- lied on cases that Florence unsettled. This argument reaches for straws. Florence abrogated only some decisions that the “robust consensus of cases of persuasive authority” cited in string citations; it didn’t abrogate any authority that the “robust consensus” deci- sions relied on in any material way. See Blackburn , 771 F.2d at 565; Varrone , 123 F.3d at 79; Thorne , 765 F.2d at 1276. Instead, many of J., Concurring in part those decisions relied on general Fourth Amendment principles that apply equally to Gilmore’s case.

Blackburn , for instance, weighed the “official interest in maintaining security against the intrusion entailed by a strip search.” 771 F.2d at 564. It explained that “the Constitution nor- mally requires a more particularized level of suspicion before indi- viduals wishing to visit a jail may permissibly be subject to a grossly invasive body search.” Id. Indeed, it described this conclusion as a “basic . . . constitutional norm.” Id. And the court highlighted that it could find no “published federal case—other than those involving incarcerated individuals —in which a court has approved body cavity searches of individual visitors about whom no particular suspicion is harbored.” Id.

Florence did not unsettle these principles; it adheres to them. As the Court explained, the relevant question was “whether un- doubted security imperatives involved in jail supervision override the assertion that some detainees must be exempt from the more invasive search procedures at issue absent reasonable suspicion of a concealed weapon or other contraband.” 566 U.S. at 330. And Florence expounded on the risk that detainees posed “for facility staff, for the existing detainee population, and for a new detainee himself or herself.” Id. The opinion expressly reserved the exten- sion of its decision to other circumstances when the need to con- duct invasive searches is “diminish[ed].” Id. In short, nothing in Florence undermines the logic that led seven (and now nine) of our J., Concurring in part sister circuits to unanimously conclude that prisons may not strip search visitors without reasonable suspicion.

A “robust consensus of cases of persuasive authority” clearly established that prison officials may not strip search a non-consent- ing visitor in the absence of reasonable suspicion. So a genuine dispute of material fact exists as to whether Defendants violated Gilmore’s clearly established constitutional rights.

IV. Conclusion

I would answer both the questions that we asked the parties to brief and that they presented on at oral argument. And to re- spect Supreme Court precedent and bring our precedent into line with every other circuit, I would answer them in the affirmative:

(1) We should overrule in part Marsh v. Butler County , 268 F.3d 1014 (11th Cir. 2001) (en banc), and Thomas ex rel. Thomas v. Roberts , 323 F.3d 950 (11th Cir. 2003), so that a robust consensus of cases of persuasive au- thority may clearly establish law for purposes of ab- rogating an officer’s qualified immunity.

(2) A robust consensus of cases of persuasive author- ity clearly establishes that Plaintiff-Appellant Clarissa Gilmore’s Fourth and Fourteenth Amendment rights were violated.

I respectfully suggest that the Court should have reached these conclusions today. T JOFLAT , J., Concurring

T JOFLAT , Circuit Judge, joined by L AGOA and B RASHER , Circuit Judges, and by L UCK , Circuit Judge, as to Parts III, IV, and V, con- curring:

“Qualified immunity gives government officials breathing room to make reasonable but mistaken judgments.” Ashcroft v. al- Kidd , 563 U.S. 731, 743, 131 S. Ct. 2074, 2085 (2011). But it does not shield “the plainly incompetent or those who knowingly violate the law.” Malley v. Bris , 475 U.S. 335, 341, 106 S. Ct. 1092, 1096 (1986). The line between the two rests on one thing: fair notice.

“[T]he salient question . . . is whether the state of the law” at the time of the violation gave officials “fair warning” that their conduct was unconstitutional. Hope v. Pelzer , 536 U.S. 730, 741, 122 S. Ct. 2508, 2516 (2002). The notice must be clear to “every reason- able official,” and it must place the law “beyond debate.” al-Kidd , 563 U.S. at 741, 131 S. Ct. at 2083 (citation and internal quotation marks omitted).

That fair notice often comes from precedent. But not al- ways. In some cases, the conduct is so extreme and so obviously inconsistent with the Constitution’s protections that it provides its own warning. When the constitutional violation is that clear, the conduct itself supplies the notice that qualified immunity demands.

This is such a case.

I agree with the Court that the officers violated Ms. Gil- more’s Fourth Amendment rights and are not entitled to qualified immunity. On these facts—taken in the light most favorable to Ms. Gilmore—no reasonable officer would think it lawful to coerce a T JOFLAT , J., Concurring civilian into a strip search, manipulate her breasts and buttocks, and inspect her genitals, all without even reasonable suspicion. It was obviously unconstitutional, and that obviousness is enough to de- feat qualified immunity.

I write separately to underscore two points.

First, the Court’s conclusion that the law was clearly estab- lished rests entirely on the obviousness of the constitutional viola- tion. No precedent was necessary to reach that conclusion, and the Court does not rely on any. The facts speak for themselves. The Court cites out-of-circuit cases only to show that the rule here is broadly accepted. But it could have reached the same conclusion without citing a single one. The officers needed no judicial roadmap; the Fourth Amendment already drew the line. See U.S. Const. amend. IV (protecting “[t]he right of the people to be secure in their persons . . . against unreasonable searches and seizures”).

Second, I address Judge Rosenbaum’s suggestion that a “ro- bust consensus” of nonbinding decisions can clearly establish the law. That is wrong. The Supreme Court has never adopted that view, and it has gone out of its way to avoid saying that even bind- ing circuit precedent is enough. Qualified immunity requires fair notice. A collection of nonbinding decisions—no matter how uni- form—cannot place the law “beyond debate.” See al-Kidd , 563 U.S. at 741, 131 S. Ct. at 2083.

My concurrence proceeds in four parts.

First, I outline the foundations of qualified immunity and the recognized sources of clearly established law: Supreme Court T JOFLAT , J., Concurring decisions, binding circuit precedent, and in rare cases, the obvious- ness of the violation itself.

Second, I explain why this case falls into that last category. The officers’ conduct was so obviously unlawful that no precedent was needed.

Third, I respond to Judge Rosenbaum’s argument that the Supreme Court has required courts to treat persuasive authority as sufficient to clearly establish the law. It has not.

And fourth, I explain why our Circuit is right to reject the “robust consensus” standard. It is vague in theory, messy in prac- tice, and inconsistent with both the fair-notice principle at the heart of qualified immunity and the role of courts of appeals as inde- pendent judicial actors.

I. A Primer on Qualified Immunity Qualified immunity, as it is known today, has its roots in common law good-faith immunities. In Pierson v. Ray , 386 U.S. 547, 87 S. Ct. 1213 (1967), the Supreme Court held that police officers sued under 42 U.S.C. § 1983 could invoke a defense akin to the good-faith and probable-cause immunity available to officers at common law. Id. at 555–57, 87 S. Ct. at 1218–19. A few years later, in Scheuer v. Rhodes , 416 U.S. 232, 94 S. Ct. 1683 (1974), the Court reaffirmed that public officials should not be held liable for consti- tutional violations if they acted in good faith and reasonably be- lieved their conduct was lawful. Id. at 247–48, 94 S. Ct. at 1692. T JOFLAT , J., Concurring

The early formulation of qualified immunity contained both subjective and objective components—officials would lose immun- ity if they knew they were violating clearly established rights (sub- jective bad faith) or if they acted in a plainly incompetent manner without reasonable grounds (objective unreasonableness). See Wood v. Strickland , 420 U.S. 308, 321–22, 95 S. Ct. 992, 1000–01 (1975).

That dual standard did not last. In Harlow v. Fitzgerald , 457 U.S. 800, 102 S. Ct. 2727 (1982), the Court jettisoned the subjective prong and reformulated the test in purely objective terms. To spare courts the burden of probing officials’ states of mind, Harlow held “that government officials performing discretionary functions gen- erally are shielded from liability for civil damages insofar as their conduct does not violate clearly established statutory or constitu- tional rights of which a reasonable person would have known.” Id. at 818, 102 S. Ct. at 2738.

Harlow ’s objective standard “provides ample protection to all but the plainly incompetent or those who knowingly violate the law.” Malley , 475 U.S. at 341, 106 S. Ct. at 1096. The focus on “clearly established” law ensures fair notice to officials so that they can con- form their conduct to settled legal norms. See id. But Harlow left open a key question: What counts as “clearly established”? The Court explicitly declined to resolve whether lower-court precedent suffices or whether only Supreme Court decisions qualify. Harlow , 457 U.S. at 818, 102 S. Ct. at 2738 n.32. T JOFLAT , J., Concurring

In the decades since, the Court has refined that standard. Our Circuit has distilled the precedent into three paths for showing that a right was clearly established. See Vinyard v. Wilson , 311 F.3d 1340, 1350–51 (11th Cir. 2002). I discuss each in turn.

A. Obvious Violation Without Case Law First, some conduct is so egregious that any reasonable offi- cial would know it violates the Constitution with no precedent re- quired. “[T]he words of the pertinent federal statute or federal con- stitutional provision in some cases will be specific enough to estab- lish clearly the law applicable to particular conduct and circum- stances and to overcome qualified immunity, even in the total ab- sence of case law . ” Id. at 1350 . We have referred to this as the “ob- vious clarity” case. Id. (internal quotation marks omitted).

The Supreme Court illustrated this approach in Hope v. Pelzer . There, prison guards twice handcuffed an inmate to a hitch- ing post for hours in the sun. Hope , 536 U.S . at 733–35, 122 S. Ct. at 2512–13. The Court found that to be a clearly established violation of the Eighth Amendment because “[t]he obvious cruelty inherent in this practice should have provided respondents with some notice that their alleged conduct violated Hope’s constitutional protection against cruel and unusual punishment. Hope was treated in a way antithetical to human dignity.” Id. at 745, 122 S. Ct. at 2518.

B. General Principles That Clearly Apply When the conduct is not obviously unlawful on its face, courts turn to precedent. Sometimes “some broad statements of principle in case law are not tied to particularized facts and can T JOFLAT , J., Concurring clearly establish law applicable in the future to different sets of de- tailed facts.” Wilson , 311 F.3d at 1351 . That is, a court may hold that “X Conduct” is unconstitutional without confining its holding to a specific factual scenario. Id. In such cases, the principle itself may give officials fair warning. Id.

Again, Hope is instructive. A prior panel of this Court held that no case clearly forbade handcuffing an inmate to a hitching post. Hope v. Pelzer , 240 F.3d 975 (11th Cir. 2001), rev’d 536 U.S. 730, 122 S. Ct. 2508 (2002). True, a prior Fifth Circuit case— Gates v. Col- lier , 501 F.2d 1291 (5th Cir. 1974)—had condemned handcuffing in- mates to fences for extended periods. And another case, Ort v. White , 813 F.2d 318 (11th Cir. 1987), held that “physical abuse di- rected at [a] prisoner after he terminated his resistance to authority would constitute an actionable [E]ighth [A]mendment violation.” Id. at 324. Still, even though our panel found that the prison officials violated the Eighth Amendment, it found that the right was not clearly established because neither Gates nor Ort involved a hitching post. Hope , 240 F.3d at 981–82.

The Supreme Court rejected that hair-splitting. Hope , 536 U.S. at 742–43, 122 S. Ct. at 2516–17. It found that the broader prin- ciples from our precedent “put a reasonable officer on notice that the use of the hitching post under [these] circumstances . . . was unlawful.” Id. at 745–46, 122 S. Ct. at 2518. The Court went further and said, “[t]he fair and clear warning that [ Ort and Gates ] provided was sufficient to preclude the defense of qualified immunity at the T JOFLAT , J., Concurring summary judgment stage.” Id. at 746, 122 S. Ct. at 2518 (citation and internal quotation marks omitted).

C. Case Law With Indistinguishable Facts Finally, the most conventional route: binding precedent with materially similar facts. Wilson , 311 F.3d at 1351–52. “That is, we look for cases in which the Supreme Court or we, or the pertinent state supreme court has said that ‘Y Conduct’ is unconstitutional in ‘Z Circumstances.’” Id.

This is the narrowest of the three paths. It requires near con- gruence between the precedent and the case at hand. Id. at 1351– 52. If such a precedent exists, the law is clearly established, and qualified immunity falls. Id. The “contours [of the right] must be sufficiently clear,” Hope , 536 U.S. at 739, 122 S. Ct. at 2515, and the facts of both cases must be “materially similar,” Wilson , 311 F.3d at 1352.

II. Gilmore’s Case Some constitutional violations fall in gray areas. The search here does not.

Ms. Gilmore was a civilian visiting her husband at a state jail. Before entering the visitation room, she passed through an initial security screening that involved (1) a pat-down, (2) a metal detector wand search, and (3) an electromagnetic radiation body-scan search. No alarm sounded. No contraband was found. And no cor- rectional officer thought anything was suspicious. Officers then es- corted Ms. Gilmore to a second building containing the visitation T JOFLAT , J., Concurring room. After Ms. Gilmore spent about 30 minutes in the visitation room with her husband, officers removed her and took her to an empty bathroom. They did not tell her why. An officer then handed Ms. Gilmore a blank strip-search approval form and told her to sign or else she would go to jail and be unable to visit her husband. The officers even told Ms. Gilmore that if she did not sign the form they would “search [her] anyway.”

And search her they did. Officers ordered Ms. Gilmore to disrobe entirely. One officer lifted Ms. Gilmore’s breasts. Another touched between her buttocks. Ms. Gilmore was commanded to spread her vagina for inspection. No probable cause justified this intrusion. No reasonable suspicion supported it. No option to de- cline existed. Ms. Gilmore’s acquiescence to the search was pro- cured by threat. And the scope of the search was exorbitant.

No reasonable officer could think this conduct was lawful. The Fourth Amendment’s protection against unreasonable searches is neither arcane nor obscure. At its core lies a principle that the State may not invade the body of a presumptively innocent person without justification. See U.S. Const. amend. IV (“The right of the people to be secure in their persons . . . against unreasonable searches and seizures, shall not be violated.”). Regardless of whether that justification needed to be probable cause or reasona- ble suspicion, that principle was obliterated here.

The Court rightly concludes that this is an “obvious clarity” case. The facts speak for themselves. The coercive, suspicionless na- ture of this strip search, coupled with its deeply invasive execution, T JOFLAT , J., Concurring rendered it plainly unconstitutional. No reasonable officer could have believed otherwise. And no precedent was needed to say so.

The Court also cites several out-of-circuit decisions involv- ing similar conduct to confirm that the constitutional principles at issue are well established. Those citations should not be under- stood as necessary—or even as a contributing factor—to the con- clusion reached here. This is not a case that depends on precedent to supply fair notice. The officers’ conduct was so extreme, so lack- ing in justification, and so invasive of personal dignity that it fell outside the bounds of any reasonable understanding of lawful au- thority. The violation was obvious. That alone is sufficient to defeat qualified immunity. The presence of reinforcing authority may lend confirmation, but it plays no role in the fair notice inquiry that compels today’s result.

III. The So-Called “Robust Consensus” Standard In her concurring opinion, Judge Rosenbaum asserts that the Supreme Court requires our Circuit to recognize that a “robust consensus of cases of persuasive authority” can clearly establish the law. See Rosenbaum Concurrence at 1. That is wrong.

She relies on six Supreme Court cases which, in her view, hold that such a consensus can suffice. They do not. I address each in turn.

One thing to keep in mind when reading this analysis is that the lodestar of qualified immunity is notice. Notice is qualified im- munity’s first principle. And the kind of notice that defeats immun- ity is legal, not judicial. Courts must ask not whether they could T JOFLAT , J., Concurring reason their way to a conclusion of unconstitutionality, but whether the official had clear and fair warning before the fact.

This is the core distinction that much of the “robust consen- sus” debate overlooks. Courts can resolve legal ambiguity; officers cannot. Judges are trained to parse legal distinctions, synthesize precedent, and reconcile seemingly conflicting lines of authority. Officers are not. They are expected to know what the law clearly forbids, not to anticipate how appellate courts will later interpret contested rules. That is impossible with the so-called “robust con- sensus” standard. Cf. Brown v. Giles , 95 F.4th 436, 439 n.1 (6th Cir. 2024) (Thapar, J.) (“Police officers protect the public in uncertain, dangerous, and rapidly evolving situations—not in the cold cruci- ble of the courtroom. Asking [officers] to divine ‘clearly estab- lished’ law from [a] smattering of [nonbinding] cases . . . would de- mand more than the Supreme Court requires.”).

As the Supreme Court has put it, clearly established law must place the right “beyond debate.” al-Kidd , 563 U.S. at 741, 131 S. Ct. at 2083.

A. Wilson v. Layne In Wilson v. Layne , federal marshals brought reporters into a home during an arrest as part of a police ride-along program. 526 U.S. 603, 605, 119 S. Ct. 1692, 1695 (1999). The Court held that do- ing so violated the Fourth Amendment—but still granted qualified immunity because the law was not clearly established. Id. at 605– 06, 119 S. Ct. at 1695. Why? Because “[t]he constitutional question presented by this case is by no means open and shut,” id. at 615, 119 T JOFLAT , J., Concurring S. Ct. at 1700, and “the state of the law . . . was at best undevel- oped,” id. at 617, 119 S. Ct. at 1701. The Court concluded that the officers “[could ]not have been expected to predict the future course of constitutional law.” Id. (citations and internal quotation marks omitted) .

The Rosenbaum Concurrence seizes on this sentence: Petitioners have not brought to our attention any cases of controlling authority in their jurisdiction at the time of the incident which clearly established the rule on which they seek to rely, nor have they identi- fied a consensus of cases of persuasive authority such that a reasonable officer could not have believed that his actions were lawful.

Id. at 617, 119 S. Ct. at 1700. But this language does not endorse a path to clearly establishing law through nonbinding circuit deci- sions. The Court cited the absence of controlling authority and the division among persuasive authorities to explain why the rule was not clearly established. The fractured state of the relevant Fourth Amendment law was the very reason immunity applied.

The Court’s broader point confirms this reading: even the judiciary was divided on whether media ride-alongs violated the Fourth Amendment. The Court went out of it way to highlight that “[b]etween the time of the events of this case and today’s de- cision, a split among the Federal Circuits in fact developed on the question whether media ride-alongs that enter homes subject the police to money damages.” Id. at 618, 119 S. Ct. at 1701 (citations T JOFLAT , J., Concurring omitted). And the Court emphasized that “[i]f judges thus disagree on a constitutional question, it is unfair to subject police to money damages for picking the losing side of the controversy.” Id.

B. Ashcroft v. al-Kidd In Ashcroft v. al-Kidd , the Supreme Court considered whether former Attorney General Ashcroft was entitled to qualified im- munity for his alleged role in the detention of Abdullah al-Kidd un- der the federal material witness statute. 563 U.S. at 734, 131 S. Ct. at 2080. The Court held that Ashcroft was entitled to qualified im- munity because there was “no Fourth Amendment violation,” and even if there was, it was not clearly established. Id. at 740–41, 131 S. Ct. 2083.

On the second point, the Court was unequivocal: “At the time of al-Kidd’s arrest, not a single judicial opinion had held that pretext could render an objectively reasonable arrest pursuant to a material-witness warrant unconstitutional.” Id. at 741, 131 S. Ct. at 2083. The best the plaintiff could muster was a stray line in a district court footnote, unsupported by citation, which the Court dis- missed as insufficient and as “an extraordinary proposition.” Id. at 741, 131 S. Ct. at 2083–84.

The Court ridiculed the idea that such a comment, which “boldly call[ed] [Ashcroft] out by name,” could amount to clearly established law. Id. at 741, 131 S. Ct. at 2084 (internal quotation marks omitted). It explained that “even a district judge’s ipse dixit of a holding is not ‘controlling authority’ in any jurisdiction,” much less the entire United States. Id . And it pointedly added: “his ipse T JOFLAT , J., Concurring dixit of a footnoted dictum falls far short of what is necessary ab- sent controlling authority: a robust ‘consensus of cases of persua- sive authority.’” Id. at 741–42, 131 S. Ct. at 2084 (quoting Wilson , 526 U.S. at 617, 119 S. Ct. at 1700).

The Rosenbaum Concurrence takes that phrase as a doctri- nal test. It is not. The Court used it descriptively—to underscore how far the plaintiff’s argument fell short. The point was not that a consensus of out-of-circuit decisions would have sufficed, but that the plaintiff had not even come close to meeting any standard. So, like Wilson , the law was too unsettled for officers to have known the right answer—particularly when courts themselves did not.

C. Plumhoff v. Rickard In Plumhoff v. Rickard , the Supreme Court determined that police officers who shot the driver of a fleeing vehicle and ended a dangerous car chase were entitled to qualified immunity. 572 U.S. 765, 768, 134 S. Ct. 2012, 2016–17 (2014). The Court found that “the officers did not violate the Fourth Amendment” and “[i]n the alter- native,” the law was not clearly established. Id. at 768, 134 S. Ct. at 2017.

Writing for the Court, Justice Alito said:

[I]t was not clearly established that it was unconstitu- tional to shoot a fleeing driver to protect those whom his flight might endanger. . . . To defeat immunity here, then, respondent must show at a minimum ei- ther (1) that the officers’ conduct in this case was ma- terially different from the conduct in Brosseau or (2) T JOFLAT , J., Concurring that between February 21, 1999, and July 18, 2004, there emerged either “‘controlling authority’” or a “robust ‘consensus of cases of persuasive authority,’” al-Kidd supra , at 2084 (quoting Wilson v. Layne , 526 U.S. 603, 617, 119 S. Ct. 1692, 143 L.Ed.2d 818 (1999); some internal quotation marks omitted), that would alter our analysis of the qualified immunity question. Respondent has made neither showing.

Id. at 779, 134 S. Ct. at 2023.

But again, the Court was not crafting a standalone rule. It was quoting al-Kidd and Wilson —where the absence of consensus confirmed the lack of clearly established law. Plumhoff restates the same point: when the legal terrain is unsettled, qualified immunity applies.

D. City and County of San Francisco v. Sheehan In City and County of San Francisco v. Sheehan , the Court ad- dressed whether officers were entitled to qualified immunity after pepper-spraying and shooting a mentally ill woman. 575 U.S. 600, 602–06, 135 S. Ct. 1765, 1769–71 (2015). The Court found that the law was not clearly established. Id. at 613, 135 S. Ct. at 1775.

Like the other cases, the Court based its holding on the lack of case law. First, it noted that “nothing in our [Supreme Court] cases suggests the constitutional rule applied by the Ninth Circuit.” Id. Second, it explained that the Ninth Circuit’s precedential cases were too factually dissimilar to put the matter beyond debate. Id. at 614–16, 135 S. Ct. at 1776–77. T JOFLAT , J., Concurring

Then came this statement: “to the extent that a ‘robust con- sensus of cases of persuasive authority’ could itself clearly establish the federal right respondent alleges, no such consensus exists here.” Id. at 617, 135 S. Ct. at 1778 (internal citation omitted) (emphasis added).

Those qualifying phrases—“to the extent that” and “could itself ”—are dispositive. It signals that the Court did not adopt, and was not relying on, a consensus-based rule. Rather, it expressly left the question open. And what it did say was clear: even assuming such a rule could exist, there was no consensus here. That fore- closes the idea that Sheehan somehow ratified a consensus standard. It did not. It dodged deciding the issue.

E. District of Columbia v. Wesby In District of Columbia v. Wesby , the Supreme Court held that officers who arrested partygoers inside a vacant home were entitled to qualified immunity. 583 U.S. 48, 65, 138 S. Ct. 577, 591 (2018). After concluding that the officers had probable cause, the Court exercised its discretion to address the “clearly established” prong and squarely rejected the argument that any precedent had given officers fair notice that their conduct was unconstitutional. Id. at 62, 64, 138 S. Ct. at 589, 590. The Court wrote:

To be clearly established, a legal principle must have a sufficiently clear foundation in then-existing prece- dent. The rule must be “settled law,” which means it is dictated by “controlling authority” or “a robust ‘consensus of cases of persuasive authority.’” T JOFLAT , J., Concurring Id. at 63, 138 S. Ct. at 589–90 (citations internal quotation marks omitted).

But, again, the Court was not adopting a new standard. It was describing the kind of legal clarity that might equate to settled law. And the Court included a disclaimer:

We have not yet decided what precedents—other than our own—qualify as controlling authority for purposes of qualified immunity. See, e.g. , Reichle v. Howards , 566 U.S. 658, 665–66, 132 S. Ct. 2088, 2094 (2012) (reserving the question whether court of ap- peals decisions can be “a dispositive source of clearly established law”). We express no view on that ques- tion here.

Id. at 66, 138 S. Ct. at 591 n.8. This footnote cuts against the reading that the Rosenbaum Concurrence offers. Far from mandating that non-binding circuit decisions could clearly establish the law, Wesby expressly declined to reach whether any circuit decisions were suf- ficient.

Judge Rosenbaum resists this point by leaning on Wesby ’s phrasing: that clearly established law must be “dictated by control- ling authority or a robust consensus of cases of persuasive author- ity.” Id. at 63, 138 S. Ct. at 589–90 (emphasis added). From that dis- junctive phrasing, she infers that “persuasive authority” alone— which she defines as nonbinding, out-of-circuit precedent—can be enough to clearly establish the law. But that inference extends be- yond anything the Court has ever endorsed. T JOFLAT , J., Concurring

To the Supreme Court, everything outside its own prece- dent—including our published decisions—is merely persuasive. See, e.g. , CBOCS West, Inc. v. Humphries , 553 U.S. 442, 472, 128 S. Ct. 1951, 1970 (2008) (Thomas, J., dissenting) (“Unlike decisions of this Court, decisions of the courts of appeals, even when unanimous, do not carry stare decisis weight.”). In other words, the boundary between “binding” and “persuasive” authority looks very different from the Supreme Court’s perch than from ours. When the Su- preme Court refers to persuasive authority, it may well mean bind- ing circuit precedent—what we already recognize as sufficient. Or it may refer to decisions from district courts, state courts, or other circuits. Judge Rosenbaum offers no reason to think the Court in- tended to place those nonbinding sources on equal footing with binding circuit precedent.

Until the Supreme Court clarifies what qualifies as persua- sive authority and how it fits into the analysis, we should not treat “robust consensus” as a rule of decision. The Court has invoked the phrase only to confirm that the law was not clearly established. That context speaks for itself.

F. Summary The cases the Rosenbaum Concurrence invokes share a theme: each uses “robust consensus” language not to impose a rule, but to illustrate the absence of one. The Supreme Court has T JOFLAT , J., Concurring never held that persuasive precedent can clearly establish the law. ^[1] But it has repeatedly said that a lack of consensus can preclude such a finding.

That is no accident. Qualified immunity protects “all but the plainly incompetent or those who knowingly violate the law.” Mal- ley , 475 U.S. at 341, 106 S. Ct. at 1096. It demands that a legal rule be so clearly settled that “every reasonable official” would under- stand it. Mullenix v. Luna , 577 U.S. 7, 11–12, 136 S. Ct. 305, 308 (2015) (citation omitted). Persuasive authority may signal uncertainty, shielding an officer from liability. But the Supreme Court has never used it to strip immunity away. See also City of Escondido v. Emmons , 586 U.S. 38, 42–43, 139 S. Ct. 500, 503 (2019) (per curiam) (reserving the question of what precedent can clearly establish the law); Rivas- T JOFLAT , J., Concurring Villegas v. Cortesluna , 595 U.S. 1, 5, 6, 142 S. Ct. 4, 7, 8 (2021) (per curiam) (same).

The Supreme Court’s consistent hesitation to endorse even binding circuit precedent as sufficient evidences the flaw in the Ros- enbaum Concurrence’s theory: If the Court has repeatedly de- clined to say whether binding circuit precedent is sufficient, it cer- tainly has not said that non-binding circuit decisions, standing alone, are sufficient. Indeed, the Supreme Court has never held that persuasive precedent can clearly establish the law. Rather, the Court has only ever used persuasive precedent to illustrate that the law was not clearly established.

All told, the cases paint a consistent picture: persuasive au- thority can inform the analysis, but it cannot resolve it. See also, e.g. Reichle , 566 U.S. at 666, 669–70, 132 S. Ct. at 2095, 2096 (“Decisions from other Federal Courts of Appeals [support that the right was not clearly established].”); Stanton v. Sims , 571 U.S. 3, 6, 10–11, 134 S. Ct. 3, 6–7, 10 (2013) (per curiam) (reversing the Ninth Circuit’s decision that a right was clearly established partly because “federal and state courts nationwide [we]re sharply divided on the ques- tion”); Pearson v. Callahan , 555 U.S. 223, 244–45, 129 S. Ct. 808, 822– 23 (2009) (reversing the Tenth Circuit’s decision that a right was clearly established because other courts of appeals ruled to the con- trary and so it was not clearly established). Every time the Supreme Court has used a “robust consensus of persuasive authority,” it has done so descriptively, not prescriptively, and always in the negative. T JOFLAT , J., Concurring

And if there were any doubt, the Court’s own language dis- pels it. In case after case, the Supreme Court has made clear that it has not yet decided whether even binding circuit precedent quali- fies as controlling authority for purposes of qualified immunity. Yet the Rosenbaum Concurrence reads those same opinions as though the issue had already been resolved—and resolved in its favor.

That reading cannot be squared with the Court’s own words. The Supreme Court has reserved the question repeatedly and expressly. And until it decides otherwise, lower courts are not bound to treat persuasive authority as a sufficient basis for denying qualified immunity.

IV. The “Robust Consensus” Standard in the Eleventh Circuit

Having concluded that the Supreme Court has not man- dated that we treat a “robust consensus” of out-of-circuit cases as controlling, the question remains whether we should adopt that standard in the Eleventh Circuit. We should not.

A. Ambiguity The first problem with the so-called “robust consensus” standard is simple: nobody knows what it means. Not even the Ros- enbaum Concurrence, which advocates for the phrase, can define it. Many seem to assume it refers to agreement across geographic circuits. But the phrase— “robust consensus of persuasive authority” — does not say that. It speaks of quantity and persuasion, not geogra- phy. T JOFLAT , J., Concurring

So what counts? Fifteen opinions all from the Fifth Circuit, all pointing in the same direction—robust or not? They are surely persuasive, and plenty in number. But apparently not the right number, from the right places. If we do limit it to a “robust consen- sus of federal circuit court decisions from different circuits,” how many make a consensus “robust”? Three? Four? Must the decisions be published? Must they be unanimous? Must they rely on the same reasoning? What if some contain hedged language, contrary dicta, or vigorous dissents?

No one applying this standard has given consistent answers. Yet we are told this vague phrase can produce clearly established law—law that governs whether public officials are personally liable for money damages. All of this in the face of the Supreme Court’s insistence that clearly established law be “beyond debate.” al-Kidd , 563 U.S. at 741, 131 S. Ct. at 2083. It is hard to see how a phrase that even learned judges cannot coherently define establishes a frame- work ensuring fair notice to officials on the ground—let alone the kind of unmistakable clarity that satisfies the “beyond debate” standard.

Judge Rosenbaum insists that the ambiguity in the “robust consensus” standard is insignificant because some of our sister cir- cuits have tried to define the phrase. See Rosenbaum Concurrence at 25–28. But that claim falters, both in premise and in practice.

Start with the premise. The Supreme Court has never de- fined what constitutes a “robust consensus of persuasive author- ity.” I do not understand Judge Rosenbaum to dispute that. As T JOFLAT , J., Concurring discussed above, what counts as binding or persuasive authority dif- fers dramatically depending on the court. For the Supreme Court, persuasive authority might include our published opinions. For cir- cuit courts, those same opinions are binding. So when sister circuits invoke the “robust consensus” formulation, they are not applying a standard set by the Supreme Court—they are creating one of their own.

That lack of Supreme Court guidance has produced confu- sion in application: our sister circuits have not agreed on what the standard means. The only consensus is that there is no consensus. Some circuits count three aligned decisions as enough; others re- quire six or more. Some consider district court or state court opin- ions; others exclude them. Some allow unpublished decisions; oth- ers reject them. Far from reflecting uniformity, the circuits’ scat- tered methods underscore how vague and unreliable the “robust consensus” standard has become.

That uncertainty has real consequences. It undermines qual- ified immunity’s core function: giving officials fair notice of what the law prohibits. See Wilson , 526 U.S. at 618, 119 S. Ct. at 1701 (“If judges thus disagree on a constitutional question, it is unfair to sub- ject [officials] to money damages for picking the losing side of the controversy.”).

B. Structure Even if a “robust consensus” could somehow be consistently defined, it would still be incompatible with the structure of the fed- eral judiciary. T JOFLAT , J., Concurring

The federal courts of appeals are not a single, nationwide panel that drafts opinions by committee and votes by circuit-wide rollcall. They are independent, coequal courts, each vested with au- thority over its own geographic jurisdiction and charged with say- ing what the law is for the citizens within it. As we have explained:

By design the federal court system allows courts to reach multiple answers to the same legal question. . . . There are 94 federal district courts around the coun- try and 12 regional circuit courts of appeals. The de- cision of any one of those courts typically has little effect on the other courts of its type: one circuit’s de- cisions are not binding on the others.

Georgia v. President of the United States , 46 F.4th 1283, 1304 (11th Cir. 2022).

In the Eleventh Circuit, only this Court, the Supreme Court, and the relevant state supreme courts have the power to clearly es- tablish law. Other circuits have no power to do so. See id. at 1304 (citation omitted).

The so-called “robust consensus” standard ignores this struc- ture. The standard instructs officials in our jurisdiction to heed de- cisions from courts with no authority over them. Worse still, it tells us that if enough circuits agree on an issue—though we cannot say how many or in what form—officials in our Circuit must treat their agreement as effectively binding. That turns our Court into a spec- tator—our judgment displaced by a headcount of out-of-circuit T JOFLAT , J., Concurring panels. Qualified immunity, in turn, becomes not a doctrine of fair notice, but a tally of judicial votes from other circuits.

But our job is not to count heads. Our duty is to get the law right—even when others disagree. Cf. Loper Bright Enters. v. Rai- mondo , 603 U.S. 369, 403, 144 S. Ct. 2244, 2267 (2024) (“[T]here is little value in imposing a uniform interpretation of a statute if that interpretation is wrong.”); see also Georgia , 46 F.4th at 1304 (“Differ- ing opinions aid the development of important questions of law.” (citation and internal quotation marks omitted)). Consensus does not relieve us of the responsibility to exercise independent judg- ment.

Some may respond that the “robust consensus” standard is not deference on the merits—it is only about whether notice ex- isted that a law was clearly established. But that distinction cannot bear weight. If the other Circuits say that conduct violates the law, then either the conduct does or it does not. Circuit splits exist for a reason: judges disagree. See Georgia , 46 F.4th at 1304 (“Conflicts [be- tween circuits] are inevitable.”); see also Mullenix , 577 U.S. at 17, 136 S. Ct. at 312 (finding that the Fifth Circuit’s holding of whether a law was clearly established conflicted with a case from our Circuit). If our Circuit holds that some conduct was lawful in the face the disagreement of other courts of appeals, it makes no sense to say the law was “clearly established” to the contrary.

Qualified immunity demands notice, not numerosity. And Article III demands independence, not conformity. We should de- cline to adopt a rule that undermines both. T JOFLAT , J., Concurring

Judge Rosenbaum responds by reframing my concern: that I object to a “robust consensus standard” because it “would force us to impose liability even if we think officials’ conduct is lawful or when other circuits disagree with us.” Rosenbaum Concurrence at 30 n.7. But that misstates my argument. I have not claimed that the standard forces us to impose liability despite our own views. My objection is more basic: the so-called standard severs “clearly estab- lished” law from what the law actually is.

Consider this hypothetical: six circuits hold that a particular type of conduct is unconstitutional. We have not yet addressed the issue. Under Judge Rosenbaum’s view, that consensus would clearly establish the law in this Circuit. See id. at 28–30 (suggesting that agreement among just three circuits should amount to a “ro- bust consensus” sufficient to clearly establish the law). But suppose we later confront the issue and hold that the conduct was lawful all along. Her approach would treat the law as “clearly established” before we had spoken—only for us to then declare that no violation ever occurred. That makes no sense. See al-Kidd , 563 U.S. at 741, 131 S. Ct. at 2083 (“[C]learly established” means that “existing prece- dent must have placed the statutory or constitutional question be- yond debate” (citations omitted)).

Judge Rosenbaum resists this point by saying the logic “proves too much.” Rosenbaum Concurrence at 32. She argues that it “would require us to overturn our current precedent that deci- sions of the relevant state supreme court can clearly establish the law,” because “[t]hose decisions don’t bind us when it comes to T JOFLAT , J., Concurring federal questions.” Id. at 32–33. But that misunderstands both our precedent and how the notice inquiry works.

A state supreme court’s interpretation of federal law may not bind us, but it does bind the state officials who are subject to the court’s authority. Those officials must conform their conduct to that court’s rulings or risk liability in their own state courts. That is what makes such decisions a legitimate source of notice—because they carry legal consequence.

Judge Rosenbaum does not dispute that state officials are bound by their own supreme courts. Rather, she notes that a state supreme court’s interpretation of federal law does not bind federal courts in a § 1983 action. Id. at 33–35. That is correct—but beside the point. Again, the question is not whether state court decisions bind us; it is whether they bind the state officials subject to that court’s authority. And they do. State officials are unquestionably governed by their own state’s highest court. When that court an- nounces a rule, those officials must conform their conduct to it or face consequences—whether through state tort liability, adminis- trative discipline, or evidentiary exclusion in criminal proceedings, among other things. That binding law provides exactly what quali- fied immunity requires: fair notice that the conduct is, in fact, un- lawful.

By contrast, out-of-circuit precedent creates no such obliga- tions. It binds no one in this Circuit—not officers, not District Courts, and not this Court. Treating decisions from other circuits as if they bore the same weight as binding law collapses the T JOFLAT , J., Concurring distinction between guidance and command. But fair notice de- mands more than persuasive reasoning; it requires a source of law that governs the official’s conduct. And decisions from other cir- cuits, however thoughtful, do not qualify.

The same confusion underlies Judge Rosenbaum’s analogy to panel opinions. She suggests that, under this logic, panel opin- ions cannot clearly establish the law either, since they may later be overturned en banc. See id. at 36–37. But that argument also misses the point. A published panel decision is binding when issued. See Martin v. Singletary , 965 F.2d 944, 945 n.1 (11th Cir. 1992). It governs the conduct of officials in this Circuit unless and until we say oth- erwise. See, e.g. , id. That is what qualified immunity requires. The notice inquiry is backward-looking: it asks whether the law was clearly established when the official acted . See al-Kidd , 563 U.S. at 741, 131 S. Ct. at 2083. Officers are not expected to anticipate future re- versals. They are expected to follow the law in effect at the time.

That is the distinction Judge Rosenbaum overlooks. Panel decisions and state supreme court rulings provide fair notice to of- ficials because they carry legal force at the time of the conduct. Out-of-circuit precedent carry none. The former create obliga- tions; the latter do not. Qualified immunity does not ask whether an officer could have guessed that a panel of judges elsewhere would disapprove of his conduct. It asks whether any authority with jurisdiction over him already said it was unlawful.

In the end, qualified immunity turns on two questions: whether a legal violation occurred and whether the law clearly T JOFLAT , J., Concurring established it. The “robust consensus” standard allows the second to be satisfied even when the first is not. That disconnect puts offi- cials in an impossible position. They must choose between follow- ing binding law as it exists or insulating themselves by conforming to nonbinding decisions we may ultimately reject. That is not fair notice. It is guesswork. And it turns qualified immunity from a shield into a gamble.

C. Application Even assuming, arguendo, that a “robust consensus of per- suasive authority” could suffice to clearly establish the law—a prop- osition neither the Supreme Court nor this Circuit has endorsed— such a standard would not alter the outcome of this case. The Ros- enbaum Concurrence asserts that, had we recognized such a the- ory today, the officers’ conduct would have violated clearly estab- lished law under the “robust consensus” standard. See Rosenbaum Concurrence at 3, 45 (answering in the affirmative whether “a ro- bust consensus of cases of persuasive authority clearly establish[ed] that Plaintiff-Appellant Clarissa Gilmore’s Fourth and Fourteenth Amendment rights were violated”). That contention cannot be cor- rect.

The relevant inquiry for qualified immunity is not what the law says today, nor what a court might wish it had said, but whether the legal rule was clearly established at the time the challenged con- duct occurred. al-Kidd , 563 U.S. at 741, 131 S. Ct. at 2083. Time and again the Supreme Court has made clear that the analysis “must be assessed in light of the legal rules that were clearly established at T JOFLAT , J., Concurring the time [the action] was taken.” Ziglar v. Abbasi , 582 U.S. 120, 151, 137 S. Ct. 1843, 1866 (2017) (citation and internal quotation marks omitted).

Here, no party disputes that the conduct occurred while Thomas and Marsh were binding. And both decisions expressly re- ject that a consensus of persuasive authority—no matter how ro- bust—could clearly establish law in this Circuit. So even if this Court were to adopt the “robust consensus” standard today, no such doctrinal shift can retroactively strip officials of the immunity they possessed under then-existing law.

To hold otherwise would defy the most basic premise of qualified immunity: that officials are entitled to fair notice before they can be held personally liable. See Nathan S. Chapman, Fair No- tice, the Rule of Law, and Reforming Qualified Immunity , 75 Fla. L. Rev. 1, 6 n.18 (2023) (collecting Supreme Court cases emphasizing the fair notice rationale). A legal standard announced today cannot supply the clarity required to strip immunity from officials who acted yesterday.

V. Conclusion Qualified immunity turns on fair notice. And here, the offic- ers had all the notice they needed. The conduct itself—without more—put the constitutional line “beyond debate.” See al-Kidd , 563 U.S. at 741. No precedent was necessary.

Judge Rosenbaum suggests that the notice could have in- stead come from a robust consensus of nonbinding authority. That is wrong. Qualified immunity is not a numbers game. A smattering T JOFLAT , J., Concurring of nonbinding decisions, however aligned, cannot place the law “beyond debate.” See id. They may confirm what is already obvi- ously unconstitutional, but they cannot make it so.

Notes

[1] The regulations of the Georgia Department of Corrections provide that “[n]o strip search shall be conducted until the Strip Search Approval Form . . . is signed by one of the following designees: . . . Warden or Deputy Warden . . . Administrative Duty Officer or the Officer in Charge with verbal approval of the Administrative Duty Officer.”

[2] We recognize that Ms. Gilmore also asserted a Fourth Amendment claim against Officer Lupo on a § 1983 conspiracy theory. We do not address that claim and leave it for the panel.

[3] Because, as explained later, any consent by Ms. Gilmore was coerced, we express no view on whether Martinez was correctly decided. We note, how- ever, that the Fifth Circuit has rejected the Martinez consent rationale in the prison visitor/strip search context. See Thorne , 765 F.2d at 1276 (“LSP next argues that the trial court erred in finding Mr. Thorne’s [strip] search unrea- sonable under the [F]ourth [A]mendment, either because Mr. Thorne con- sented to his search or because he waived his [F]ourth [A]mendment rights when he entered the prison. LSP locates this consent or waiver in the visitor form signed by Mr. Thorne and in the warning notices posted at the prison gates. If accepted, this argument would render reasonable a strip search of any such prison visitor; as discussed above, such at-will, random searches are not reasonable under the Fourth Amendment. The argument must therefore fail.”).

[4] As noted, the sign posted outside the sallyport did not provide visitors any warning that they could be (or would be) subject to suspicionless strip searches. As a result, there can be no argument that Ms. Gilmore’s entry con- stituted implied consent to a strip search. Cf. United States v. Sihler , 562 F.2d 349, 350–51 (5th Cir. 1977) (holding that a prison employee consented to a search of his lunch bag upon entering the institution for work because a posted sign warned that “all persons entering upon these confines are subject to rou- tine searches of their person, property or packages”).

[5] Mercado and Terrell are not the only Eleventh Circuit decisions after Marsh and Thomas to this effect. See, e.g., Crocker v. Beatty , 995 F.3d 1232, 1240 (11th Cir. 2021) (“[O]nly decisions of the United States Supreme Court, this Court, or the highest court in a state can ‘clearly establish’ the law.”); Loftus v. Clark- Moore , 690 F.3d 1200, 1206 (11th Cir. 2012) (“Clark-Moore, as a Florida official performing a discretionary duty, cannot be held to a standard of conduct which is unsettled by the Supreme Court[,] . . . this Circuit[,] or the highest state court in Florida.”) (internal quotation marks and citation omitted); Kelly v. Curtis , 21 F.3d 1544, 1551 n.6 (11th Cir. 1994) (“By distinguishing those two out-of-circuit decisions that Kelly has cited, we do not mean to imply that the

[6] Again, we leave for another day what constitutes a “robust consensus of per- suasive authority” and whether such a consensus can by itself create clearly established law under the other two methods in the absence of Supreme Court or Eleventh Circuit precedent.

[1] The Tjoflat Concurrence invokes Wesby ’s eighth footnote to argue that the Supreme Court has not ruled that a “robust consensus of cases of persuasive authority” can clearly establish the law. See Tjoflat Conc. Op. at 15–17. That’s a swing and a miss. In fact, Wesby ’s footnote eight neither draws nor supports any conclusion of the kind. Wesby ’s footnote eight says, “We have not yet decided what precedents—other than our own—qualify as controlling authority for purposes of qualified immunity.” 583 U.S. at 66 n.8 (emphasis added). And that footnote’s reliance on Reichle v Howards , 566 U.S. 658, 665–66 (2012), illus- trates that same thing: the Supreme Court hasn’t yet defined the meaning of “controlling authority” in the qualified-immunity context. But the Supreme Court’s recognition that it hasn’t said what qualifies as “controlling authority” doesn’t renounce the Court’s earlier holdings that “controlling authority” (whatever may so qualify) and “persuasive authority” can both clearly estab- lish the law. Indeed, nothing in the footnote the Tjoflat Concurrence points to undermines the Court’s previous holding that “decisions in various Cir- cuits” can provide a “warning [that] is fair enough” for officers to be held lia- ble. United States v. Lanier , 520 U.S. 259, 269 (1997). Just the opposite. Wesby repeats that a “robust consensus of cases of persuasive authority” can clearly establish (or “settle[]”) the law. Wesby , 583 U.S. at 63 (internal quotation marks omitted). And if Wesby left open the possibility that decisions of the courts of appeals are not controlling authority , then a “robust consensus of cases of per- suasive authority” can abrogate qualified immunity all the more. Otherwise, only Supreme Court precedent could clearly establish the law—and no deci- sion in the qualified-immunity context has accepted (or even suggested) that rule. See Lanier , 520 U.S. at 269 (explaining “that in applying the rule of quali- fied immunity,” the Court has “referred to decisions of the Courts of Appeals when enquiring whether a right was ‘clearly established’”). Attempting to thread Wesby ’s footnote through Lanier ’s holding, the Tjoflat Concurrence posits that the Supreme Court’s understanding of “controlling authority” and “persuasive authority” may differ from ours. Tjoflat Conc. Op. at 16–17. From that, it conjects that the Supreme Court could have defined

[4] The Federal Circuit addressed a separate issue where it employed a clearly- established-law framework. It acknowledged in a string citation that “clearly established law in the qualified immunity context” includes “cases from the Supreme Court and the U.S. Court of Appeals” for the relevant circuit, as well as “cases from other courts exhibiting a consensus view.” U.S. Capitol Police v. Off. of Compliance , 916 F.3d 1023, 1027 (Fed. Cir. 2019) (cleaned up) (quoting Bame v. Dillard , 637 F.3d 380, 384 (D.C. Cir. 2011)).

[5] The Tjoflat Concurrence suggests that because no uniform rule about what constitutes a “robust consensus” exists, the standard undermines qualified im- munity’s core purpose of fair notice. Tjoflat Conc. Op. at 21–22. That logic misunderstands the difference between persuasive and controlling authority. Yes, any definition of a “robust consensus of cases of persuasive authority” would necessarily refer to the concept of persuasive authority. But if we used this opportunity to define what a “robust consensus” is in the Eleventh Circuit, then we would create controlling precedent on that definition. And officers would have fair notice of when a constitutional violation is clearly established under that standard in this Circuit. It makes no difference that we may define the “robust consensus” standard differently than other circuits do because our definition would control in this Circuit under our prior-panel-precedent rule. See GTE Corp. , 236 F.3d at 1300 n.8. After all, circuits occasionally split on sub- stantive law in Section 1983 cases. But no one would suggest a circuit split deprives officers in one circuit of fair notice if their circuit has already held that certain conduct violates clearly established rights: the officers know to which side of the split their circuit adheres. See, e.g. , Laskar v. Hurd , 972 F.3d 1278, 1294, 1298 (11th Cir. 2020) (acknowledging “that our conclusion departs from the consensus of our sister circuits” but confirming the plaintiff alleged that the defendants “violated clearly established law” under our precedent). The same is true of the meaning of “robust consensus.” Controlling precedent governs in both situations. Plus, the implication of the Tjoflat Concurrence’s suggestion that we need not follow Supreme Court precedent if we don’t agree with it—that is, we need not accept the “robust consensus” standard even though the Supreme Court has repeatedly identified it as a governing standard, just because the Tjoflat Concurrence finds it wanting—is simply un- tenable.

[6] We don’t parse the text of court opinions like we do statutes. Reiter v. Sono- tone Corp. , 442 U.S. 330, 341 (1979). Still, we are bound by Supreme Court holdings and want to be faithful to them. “Consensus” means “[a] general agreement; collective opinion.” B LACK ’ S L AW D ICTIONARY 382 (12th ed. 2024). So by definition, more than one decisionmaker must be involved.

[7] To the extent that the Tjoflat Concurrence suggests that the “robust consen- sus” standard somehow makes the opinions of our sister circuits binding and turns us into a “spectator,” Tjoflat Conc. Op. at 24, that’s just wrong. We’re always free to differ on the merits from the views of other circuits, no matter how many other courts may unanimously agree with the opposite merits po- sition. For that same reason, the Tjoflat Concurrence is wrong again in assert- ing that the “robust consensus” standard would force us to impose liability even if we think officials’ conduct is lawful or when other circuits disagree with us. Id. at 23–24. My point is a simple one: if we agree with several other federal appellate courts on a legal principle that clearly applies “to the specific conduct in question,” Lanier , 520 U.S. at 271, then officials will have had notice that is “fair enough” to impose damages liability, id. at 269.

[1] Judge Rosenbaum cites the Supreme Court’s recent decision in Andrew v. White , 145 S. Ct. 75 (2025), for the proposition that: “[T]he [Supreme] Court has unambiguously told us that when it uses ‘a legal rule or principle to decide a case, that principle is a ‘holding’ of the Court.’” Rosenbaum Concurrence at 15 (quoting Andrew , 145 S. Ct. at 81). But that is not what the Court said. The actual language is this: When this Court relies on a legal rule or principle to decide a case, that principle is a “holding” of the Court for purposes of the AEDPA . Andrew , 145 S. Ct. at 81 (citation omitted) (emphasis added). Judge Rosenbaum omits the limiting clause. But that clause confines the statement to habeas re- view. The Supreme Court does not include such qualifiers by accident, and we are not free to ignore them. Contra Rosenbaum Concurrence at 13 n.3.

Case Details

Case Name: Clarissa Gilmore v. Georgia Department of Corrections, et al

Court Name: Court of Appeals for the Eleventh Circuit

Date Published: Jul 11, 2025

Citations: 144 F.4th 1246; 23-10343

Docket Number: 23-10343

Court Abbreviation: 11th Cir.