On Petition for Rehearing
FERREN, Senior Judge:
After our initial hearing of this case, the court ruled in favor of the defendant-appellee, Lorenzo Ali Debruhl.1 We affirmed the trial court’s pre-trial ruling that the federal “exclusionary rule,” rather than the “good faith exception” to that [294]*294rule, should be applied retroactively to enforce the Supreme Court’s decision in Arizona v. Gant.2 Thus, we affirmed suppression at trial of the drugs and related contraband found by the police in a war-rantless search of Debruhl’s car. After our decision, the Supreme Court addressed the good-faith exception in Davis v. United States,3 and sustained its application in the very post-Gcmi context presented in Debruhl. We therefore granted rehearing to determine whether, in light of Davis, our decision in Debruhl can be reaffirmed. We now reverse and remand.
I.
On January 11, 2009, Metropolitan Police Department officers saw an Oldsmobile traveling with its lights off between 1:00 and 2:00 a.m. in the 900 block of Hamilton Street, N.E. They conducted a traffic stop, ran a license plate check, and discovered no listing for the tags. Upon checking the driver’s license and registration, the officers found that the registration matched the license plate but not the car. A check of the vehicle identification number revealed that the car was unregistered. One of the officers asked the sole occupant, the driver, Debruhl, to step out of the car, placed him under arrest, secured him with handcuffs, and placed him behind the car. The other officer searched the passenger compartment and found a brown paper bag under the driver’s seat containing a pair of gloves, a digital scale, razor blades, some currency, and a clear plastic bag with a white rock substance inside that field-tested positive for cocaine. A grand jury indicted Debruhl on one count each for possession of a controlled substance with intent to distribute and for possession of drug paraphernalia. Before trial, Debruhl filed a motion to suppress the drugs and drug paraphernalia, which the trial court granted in September 2009.
Meanwhile, three months or so after Debruhl’s arrest and the related search of his car, the Supreme Court in Gant clarified the reach of New York v. Belton.4 It declared unconstitutional the warrantless seizure of evidence from the passenger compartment of a vehicle from which the occupants — like Debruhl — had been removed, handcuffed, and sequestered.5 No one disputes that, by virtue of Griffith v. Kentucky,6 Gant’s interpretation of the Fourth Amendment would apply not only to Gant’s trial and future cases but also to all cases, such as Debruhl’s, that were “not yet final”7 when Gant was decided. In Debruhl I, however, we acknowledged that the exclusionary rule will not always accompany a retroactive application of the Fourth Amendment.8 Rather, the good-faith exception will apply, we said, when police officers who unlawfully seize evidence from a car, without a warrant, have [295]*295relied on settled appellate precedent that permitted their search under Belton before the Supreme Court added Gant’s clarification to the contrary.9 Our decision therefore turned on whether, at the time of arrest and seizure, there was settled precedent in this jurisdiction, under Bel-ton, that justified the warrantless search of Debruhl’s car. We held that there was not, because no case in this jurisdiction was “settled as to all the material facts” the officers faced.10
A material fact, in the context here, is any fact necessary to justify-or forbid-the warrantless search of an automobile compartment under Belton. After the Supreme Court in Gant added sequestration of the occupants as a material consideration for the scope of a search, the officers in Debruhl — for the good-faith exception to apply — must have relied on pre-Gant precedent holding that this now material fact was immaterial at the time of their search. If the officers searched without a warrant in the absence of such settled precedent, then Debruhl’s case itself, we noted, would contain a factual variant-sequestration of the occupants — that had not yet been addressed by this court, and thus was not part of our “settled” jurisprudence justifying warrantless car searches.11 With this understanding, we elaborated our decision as follows.
In Belton, the Supreme Court announced a “ ‘bright-line’ rule”12 permitting the warrantless search of an automobile passenger compartment. The Court held that “when a policeman has made a lawful custodial arrest of the occupant of an automobile, he may, as a contemporaneous incident of that arrest, search the passenger compartment of that automobile.” 13 This was intended as a practical rule, justified only because of its close approximation to the scope of a warrantless search of an individual permitted by Chimel v. California.14 The rationale justifying a Chimel search, and thus a search under Belton, was the dual need to protect the arresting officer against a “weapon” the suspect might have within his “immediate control,” as well as to preserve “destructible evidence” the suspect might otherwise dispose of.15 Eventually, however, Belton’s bright-line rule became problematic.16 As we demonstrated in Debruhl I,
bright-line rules do not easily remain radiant. Although Belton was ... intended for simple, clear cut application, [296]*296it is evident from Belton’s history that such rules are often likely to remain truly “bright line” only for a limited period of time as factual scenarios test their limits. Belton has become a classic case of a rule beclouded over time by exceptions generated by unique facts that pushed decisions beyond the “bright line” license-as evidenced by ... three federal circuits that anticipated Gant.[17]
The three federal precursors of Gant, and then Gant itself, focused upon the reality that once all occupants had been removed from a vehicle, handcuffed, and sequestered, Chimel’s dual rationale-protection of officer and preservation of evidence — had been satisfied; the reasons for a warrantless, protective search had been eliminated. Thus came Gant’s clarification limiting Belton searches to those in which one or more occupants were still in the vehicle or, if removed, were not yet sequestered in a manner that assuredly would protect officers and preserve evidence.18 Hence the question in Debruhl: was there an appellate decision in this jurisdiction,19 authorizing a Belton search after all occupants of the car had been handcuffed and sequestered, that would justify our application of the good-faith exception to permit introduction at trial of the contraband seized from Debruhl’s car despite the officers’ Fourth Amendment violation in seizing it?
The two decisions of this court approximating the facts in Debruhl, and on which we principally relied for analysis at the government’s urging, are Staten v. United States20 and United States v. Harris.21 Unlike Debruhl I, however, both dealt with protective searches of automobiles either before all occupants had been removed (Harris) or before all occupants had been sequestered (Staten). We summarily distinguished three other Belton cases cited by the government,22 none of which involved protective searches, on the grounds that one focused on the “contemporaneous incident” element of Belton23 another turned on the “inevitable discovery” doctrine,24 and the third held the motion to suppress untimely filed and thus addressed Belton only in dictum.25 Accordingly, because neither Staten nor Harris included a fact — sequestration before a Belton search — that was material, indeed determinative, in Gant, we found no relevant, “settled” precedent on which the police officers in Debruhl could have relied to support the good-faith exception. We [297]*297held, therefore, that the exclusionary rule applied in Debruhl I.
II.
Then came Davis.26 Like Gant and De-bruhl, Davis concerned situations in which the police — before the Gant decision came down — removed, handcuffed, and sequestered the occupants of a car and “then searched the passenger compartment,” finding the contraband at issue.27 According to Davis, the police
followed the Eleventh Circuit’s Gonzalez[28] precedent to the letter. Although the search turned out to be unconstitutional under Gant, all agree that the officers’ conduct was in strict compliance with then-binding Circuit law and was not culpable in anyway.[29]
Then, in affirming the circuit court’s ruling based on the good-faith exception, the Supreme Court held that “when the police conduct a search in objectively reasonable reliance on binding appellate precedent, the exclusionary rule does not apply.”30
By premising the good-faith exception on “strict compliance” with “binding appellate precedent,” and by noting that the search was justified by police reliance “to the letter” on precedent that would later be held unconstitutional under Gant, the Court was saying, we believe, that the precedent (Gonzalez) on which the police relied for the good-faith exception in Davis, embraced all the facts material to the decision in Davis, including handcuffed and sequestered occupants.31 Otherwise, as we emphasized in Debruhl I, there would have been no “explicit protection or ‘cover’ ” from the appellate court on which the officer could rely to block application of the exclusionary rule.32 There would, in other words, have been a gap in precedent leaving no reasonable, objective basis for a police officer to conclude that the appellate court would have been bound to uphold the search on Gant-type facts before Gant was decided.
The particular “binding appellate precedent” on which the Eleventh Circuit relied in Davis, namely its earlier decision in Gonzalez, was precedent in which the relevant settled material facts — handcuffed and sequestered occupants — were virtually identical to, and thus legally congruent with, the facts before the later, Davis court that was bound by that precedent. There remains, nonetheless, the question whether “binding appellate precedent,” as contemplated by Davis, could have a more comprehensive meaning than the “settled material facts” rule we applied in Debruhl 7.33 We turn to that inquiry, beginning [298]*298from the premise that for purposes of this case we are to focus on the appellate jurisprudence of this court.34
III.
For “binding” or “settled” precedent in this jurisdiction we look first to our decision in M.A.P. v. Ryan.35 There, we adopted the rule that “no division of this court will overrule a prior decision of this court”; only the en banc court may do so.36 Accordingly, we can reaffirm Debrulil I if — but only if — we correctly concluded that none of our prior Belton decisions bound us, as a division, to hold that the police objectively relied on precedent that justified the search of Debruhl’s passenger compartment. We therefore turn more specifically to what it means for a “prior decision of this court” to be “binding” under M.A.P. v. Ryan — and thus “binding appellate precedent” under Davis.
This court has equated binding precedent under M.A.P. with the rule of stare decisis.37 We have stressed, however, that “stare decisis is never properly invoked unless in the decision put forward as precedent the judicial mind has been applied to and passed upon the precise question.”38 Indeed, “ ‘[questions which merely lurk in the record, neither brought to the attention of the court nor ruled upon, are not to be considered as having been so decided as to constitute precedents.’ ”39
In Debrulil I, when applying the “settled material facts” rule to define binding precedent, we concluded that neither of this court’s factually relevant Belton decisions, Staten and Harris, satisfied the rule. Neither decision (in the words of our stare decisis precedent) had “passed upon the precise question”40 presented in Debrulil I, namely, whether Belton permitted a warrantless search of the passenger compartment after complete sequestration of the occupants. We therefore deemed that factual difference material, indeed disposi-tive, because it reflected Gant’s recognition that the very justification for a Belton protective search — Chimel’s concern about officer safety and preservation of evidence — had disappeared upon sequestration.41
[299]*299Is it possible, however, for Belton precedent — specifically, Staten and Hams — to provide “binding” support for the officer’s search in Debruhl even though not all the occupants of the car in either case were handcuffed and sequestered at the time of the search? We held in Debruhl I that the answer is “no” — that the settled material facts rule is the universe for binding preee-dent-because of our concern that factual variants from the precedent on which a police officer relies could lead this jurisdiction toward acceptance of the discredited “mistake of law” justification for Fourth Amendment violations.42 That said, however, even a settled material facts rule requires a process for determining what facts in a particular context are “material” or “immaterial” for purposes of comparing a pending matter with a prior decision; and that question about process requires a further inquiry than we pursued in De-bruhl I — an inquiry that expands our understanding of “settled” or “binding” appellate precedent.
IV.
We turn first to the facts material in Belton. We have said that Belton announced a “two-pronged test” and created factual boundaries with this “bright-line” rule.43 This bright-line rule emerged be[300]*300cause Chimel’s justifications for a warrant-less search of the area under an individual’s “immediate control”44 (police officer safety and retrieval of disposable evidence) 45 had become problematic for courts to apply to individuals whom the police had lawfully rousted from automobiles of different sizes and configurations. What precisely was the so-called “grab” area inside the vehicle that corresponded to the area of danger and disposal within a particular occupant’s “immediate control”? 46
Writing for the Court in Belton, Justice Stewart noted that after Chimel, in United States v. Robinson,47 the Court had “hewed to a straightforward rule, easily applied, and predictably enforced” authorizing a “full search of the person” incident to a “lawful custodial arrest.”48 Robinson, therefore, had rejected, in that context, the need for litigation of particular “reasons supporting the authority for a search.”49 The Belton Court concluded accordingly that a similar, “workable rule” unfettered by fact-parsing was necessary to govern auto compartment searches — especially to resolve the confusion encountered by police officers and judges in evaluating the legality of searches after a “recent occupant” had been removed from the car.50 Thus came Belton’s bright-line rule, limiting the facts that would be material to determining the constitutionality of a search.51
Belton did not end all legitimate inquiries concerning new material facts, however. This court, for example, held that an individual who had parked, locked, and walked fifteen to twenty feet away from his car was no longer an “occupant” within the scope of a lawful Belton search.52 On the other hand, we held that a search conducted after a wait for a showup identification of the driver was a sufficiently “contemporaneous incident” of the arrest to satisfy Belton.53 In both cases, however, we were construing the language of Belton’s express holding;54 we were not conducting a factual inquiry into application of Chimel’s intendments.
[301]*301As noted in Debruhl I, bright-line rules can easily fade as circumstances challenge their justification.55 And, as we elaborated, Belton itself began to erode as courts concluded, on particular facts — notably handcuffed and sequestered occupants— that Belton’s bright-line rule for searching an automobile passenger compartment was too sweeping for justification under Chi-mel’s rationale.56 Hence, eventually came Thornton,57 revealing the Supreme Court’s fracture over Belton,58 and then Gant, recalibrating Belton to clarify and impose a discernible Chimel limitation upon war-rantless automobile compartment searches.59
In Debruhl I, we were persuaded by pre-Gant rulings from other courts which had acknowledged the material change of fact from Belton — sequestration of the occupant — that justified a limitation on Bel-ton’s compartment search rule. And, we were satisfied that neither Belton nor Staten and Harris (both of which merely reflected the facts in Belton) could bind this court to ignore the same factual change in Debruhl. After rehearing, however, based on splendid briefs and oral arguments by the parties and amicus curiae, we have concluded that we must approve the government’s argument that the good-faith exception applies here, reject our ruling in Debruhl I, and reverse the trial court’s suppression order.
V.
In order to determine whether a fact is “material” to a precedent on which a police officer relies for the good faith exception, we look, first, at the defendant’s contention. In this case, Debruhl asserts that, for a precedent to be “settled” or “binding,” its holding must have rejected the defense, successful in Gant, that Bel-ton cannot be stretched to authorize the warrantless search of a car (here comes the material fact) after all the occupants have been securely sequestered. Second, we look at whether the government-proffered precedent incorporates that fact and, if not, whether there is, nonetheless, a sound basis for the government’s argument that the precedent is binding. In Staten and Harris, on which the government principally relies, all agree that the alleged material fact — handcuffed sequestration as to all occupants — is missing. So, what does the government say next?
The government replies that Staten and Harris foreclose Debruhl’s contention because both cases “explicitly held that Belton was not limited to its facts.” In support, the government notes Staten’s rejection of the appellant’s argument that the record lacked evidence justifying Chi-mel’s concerns that underlay Belton (officer safety and disposable evidence).60 It then stresses Harris’s rejection of the motion judge’s reliance, in granting suppression, on “material factual differences” from Belton, including the lack of “reasonable fear” for officer safety.61 In short, argues the government, this court twice has addressed and rejected De-bruhl’s contention that the decisional law of this jurisdiction, before Gant, allowed a court to narrow Belton’s reach by reference to facts germane to Chimel.
[302]*302These observations might have currency-more easily if the factual scenarios in Staten and Hams differed in any significant — call it material — way from the facts in Belton. But they did not. Despite the defense proffers in Staten and Hams that the searches were unlawful because the facts revealed no concern for officer safety, Staten and Hams, like Belton, involved one or more occupants of a car who were not securely sequestered at the time of the search. Accordingly, neither in Staten nor in Hams did the facts permit this court, in reaching a decision, to distinguish the situation from Belton itself; in neither case did the trial court, let alone this court, have leeway to rule for the defendants based on the facts. Thus, rejection of the factual arguments proffered in Staten and Hams (we have called them “pure” Belton cases)62 could not, as such, create a precedent that would foreclose a contention based on different facts in a later case— Debruhl — unless for some other reason Belton itself stood in the way.
VI.
Although this court’s rejection of the defendants’ factual contentions in Staten and Hams did not, in itself, preclude a factual inquiry in other cases, we nonetheless reverse now for two reasons. This Part VI explains the first. If this case had arisen before Gant, we believe that our en banc court, like the Arizona Supreme Court in Gant,63 would have had room to hold that Belton’s reach did not extend to warrantless searches of automobiles after all occupants had been sequestered and secured. However, in light of this court’s unequivocal affirmation in Staten and Harris that a Belton compartment search was limited by only two variables-probable cause and contemporaneity 64-we must agree with the government that this court’s precedent bound our division under M.A.P. v. Ryan65 not to stray into factual territory that was otherwise ripe for rethinking.
We repeat our observation in Part V: the facts of Staten and Hams reflected the facts of Belton; thus, Staten and Harris, in rejecting factual arguments against a Belton search, did not, merely by reason of that rejection, foreclose application of Belton to another factual scenario, as in Debruhl But the analysis cannot stop there. Staten and Hams offered interpretive pronouncements, broader than we initially credited, about Belton’s factual reach. Those decisions made clear (without expressly stating) that the bright-line rule allowed for no limiting material facts beyond those identified in the plain language of Belton’s holding; implicitly, therefore, sequestration was an immaterial fact.66 For that reason, we now believe that it would bend M.A.P. too far to conclude, as we did earlier, that these unequivocal pronouncements do not extend to the Gant-type facts before the court in Debruhl. In sum, the interpretation of Belton in Staten and Hams that was in effect at the time of Debruhl’s arrest constitutes “binding appellate precedent”67 on which the police officers who arrested De-bruhl could rely in objective good faith.
In recognizing the binding effect of a bright-line rule, without room for scrutiny [303]*303of the facts to which it is applied, we are doing so only for purposes of construing our obligation as a division of this court under M.A.P. v. Ryan, and thus for determining “binding appellate precedent” for purposes of the good-faith exception to the exclusionary rule under Davis. In no way are we saying that bright-line rules are exempt from judicial scrutiny for unlawful — often unconstitutional — application to facts not contemplated, or improvidently considered, when the rule was announced. Gant itself is a typical example of constitutional reevaluation of such a rule, just as this court’s recent en banc decision in Hed-gepeth v. Whitman-Walker Clinic68 is an example of common law reevaluation of a previously bright-line negligence rule. Normally, moreover, binding precedent is limited to case law controlled and distinguished by facts, not (as in Belton) by bright-line rules that subordinate the facts to an unbending pronouncement.69 Finally, even recognition of a bright-line rule does not derogate from the settled material facts rule70 we relied on in Debruhl I to determine binding appellate precedent. A bright-line rule is merely a generalization describing a fact-bound universe — in Bel-ton’s case, all automobile compartment searches limited by probable cause,71 occupancy,72 and contemporaneity.73 Ultimately, all binding precedent is determined by reference to settled material facts that justify the court’s decision.74 Thus, we reaffirm that analysis in Debruhl I.
[304]*304VII.
In addition to our protective search jurisprudence under Belton, reflected in Staten and Harris, the government urges a second reason for concluding that the facts in Debruhl do not undermine reliance by police officers on Belton to justify the good-faith exception. In Staten, the court relied on Smith,75 our first decision to invoke Belton (which was decided during the course of Smith’s appeal). Unlike Staten, Harris, and now Debruhl — all protective search cases — Smith was a showup case in which the occupant had been sequestered awaiting arrival of an eyewitness who eventually identified him as the person who had committed an assault. This court cited Belton to justify the eventual search of the suspect’s car, rejecting the defense contention that the search had not been sufficiently contemporaneous with the arrest.76 Later, we cited Belton without discussion in Hicks,77 another showup case, to justify a search of the occupant’s car. This time the focus was almost entirely on our analysis concluding that an unlawful search could be cured by the inevitable discovery doctrine.78 Neither Smith nor Hicks discussed whether reliance on Bel-ton in the context of a showup, not a protective search, was valid — manifestly a context ripe for a Gcroi-type argument that was never made (either at division or en banc). Moreover, in both cases the auto[305]*305mobile exception would have permitted a warrantless search of the automobile;79 Belton was not an indispensable citation. And yet in Smith and Hicks, after our respective resolutions of the contemporaneity and inevitable discovery issues, this court needed authority to justify entry into the car and the ensuing search. For whatever reason, most likely the briefing,80 the court picked Belton.
Amicus curiae embraces our rejection in Debruhl I of Smith and Hicks as authority justifying the good-faith exception.81 Amicus argues that, in the way Smith and Hicks unfolded,82 the court’s “judicial mind” in citing Belton was not “applied to” a sequestration issue.83 That issue, amicus cautions, was “never brought to the attention of the court”; it “merely lurk[ed] in the record.”84 Accordingly, concludes amicus, the indispensability of Belton to the decisions in Smith and Hicks does not mean that those decisions should be given “precedential weight” on the question of sequestration.85
The cases relied on by amicus, primarily Murphy and Sierra Club, were decisions reflecting challenges to the court’s authority to resolve the parties’ disputes. In both cases, the opposing party argued that the jurisdictional issue had implicitly been decided in its favor in an earlier case as a necessary predicate to the decision on the merits. This court disagreed, however, because in both of the earlier cases the issue had not been raised and joined. On each occasion, therefore, we ruled on the issue de novo.86 In Sierra Club, for example, we stressed that a “‘point of law merely assumed in an opinion, not discussed, is not authoritative.’ ... This is especially true with respect to jurisdictional issues, for ‘when questions of jurisdiction have been passed on in prior decisions sub silentio, this court has never considered itself bound when a subsequent case finally brings the jurisdictional issue be[306]*306fore us.’ ”87
In Smith and Hicks, the facts the parties deemed material in addition to probable cause were those addressed, respectively, to contemporaneity and inevitable discovery, not to the occupants’ situations awaiting a showup. Without doubt, in upholding the searches in those two cases, the court’s “judicial mind” was not at all focused on whether sequestration of a car’s occupants affected the legitimacy of a Bel-ton search. Furthermore, although in Sierra Club (relying on Murphy) we were particularly concerned that jurisdictional issues not be decided implicitly, we have never limited our concern about sub silen-tio decision-making to issues of judicial power.88
On the other hand, in each of the cases on which Amicus relies — Murphy, Sierra Club, and Umana — the earlier decision that we said had not implicitly resolved an issue was a decision that merely had assumed a legal proposition without challenge. In Smith and Hicks, to the contrary, this court did not merely assume that Belton applied to the search on the court’s way to decision; we employed Bel-ton affirmatively to refute a challenge to the search — to provide authority for the warrantless entry. We were saying that, on these facts (which included sequestration for a showup), Belton allows the police to enter the car.
Moreover, the argument that sequestration was not resolved in the showup cases is eclipsed by the reality that Smith itself is the source from this court for Staten’s “two-pronged test” describing Belton’s broad reach that we now recognize in Part VI as binding appellate precedent.89 By extending Belton’s bright-line rule beyond a protective search to a search after a showup, Smith irrefutably made the degree of sequestration a non-issue. Vis-a-vis Staten and Harris, our decision in Smith (later reflected in Hicks ) creates a fortiori the settled law-binding appellate precedent — that justified a Belton search on Gcmi-type facts, and thus permits the good-faith exception, in Debruhl.
[[Image here]]
We reverse and remand for further proceedings.
So ordered.