concurring and dissenting.
I join in Parts II and III of Judge Green-berg’s opinion and in Part I (Factual Background) insofar as it recounts the undisputed facts as revealed and supported by the record. For the reasons I discuss below, however, I do not join in Parts IV through VII of Judge Greenberg’s opinion.
I.
I begin by setting forth my agreement with Judge Scirica that no distinction should be drawn between the decision to. drop the incendiary device and the decision to let the fire burn. As Judge Scirica observes, both actions were undertaken to effect the same result, namely, to enable the police to insert tear gas into the house in order to force the MOVE occupants out. I also agree with Judge Scirica, for the reasons stated in his concurring and dissenting opinion, that the police effected a Fourth Amendment seizure in this case.
Finally, I join Judge Scirica in holding that the federal claims asserting an unconstitutional seizure in violation of the Fourth Amendment against the City of Philadelphia survive summary judgment and may proceed to trial. However, I do not agree with, either Judge Scirica or Judge Greenberg that Graham v. Connor, 490 U.S. 386, 109 S.Ct. 1865, 104 L.Ed.2d 443 (1989), provides the proper analytic framework for assessing the constitutionality of the use ,of force in this case. Instead, it is plain to me that Tennessee v. Garner, 471 U.S. 1, 105 S.Ct. 1694, 85 L.Ed.2d 1 (1985), provides the proper test.
II.
THE MAJORITY’S ERRONEOUS CONCLUSION AS TO THE APPLICABLE STANDARD
Judge Greenberg concludes that, the city defendants’, use of the incendiary device and ensuing fire to effect the seizure of the MOVE occupants did not constitute deadly force. While Judge Scirica acknowledges that deadly force may have been used against the MOVE members, he believes that Gamer is limited by its facts and distinguishable from the confrontation here. See Judge Scirica’s concurring and dissenting opinion at 973 n. 1. Thus, the majority concludes that the district court erred in applying Gamer, a case involving the use of deadly force, to assess the constitutionality of the use of force against MOVE. Instead, the majority holds that the district court should have applied the reasonableness test described in Graham, a case involving the use of non-deadly force, to test the constitutionality of the use of force in this ease. In so holding, I believe Judge Greenberg misapprehends the nature of the force contained in the incendiary device- and the ensuing roof-top fire, and I believe that Judge Scirica overly limits Garner ’s reach.
A.
The force embodied in the incendiary device and in the fire was, by any sensible *977standard, “deadly” force. It should not escape our attention that the destructive device in question was a bomb capable of blasting a hole in the roof of a city building. This lethal device could and did accomplish a degree of destruction well beyond that which even bullets fired from a high-powered weapon could achieve. To me, this constitutes deadly force, for I cannot imagine how anyone Can conclude that a device such as this was not capable of taking human life — which, in fact, it ultimately did. Contrary to the conclusion drawn by the majority, I believe the city defendants, who were professional law enforcement and fire-fighting personnel, had every reason to know that their actions created a substantial risk ■ of death or serious bodily injury to the MOVE members residing at 6221 Osage Avenue. See Model Penal Code § 3.11(2) (defining “deadly force” as “force which the actor uses with the purpose of causing or which he [or she] knows to create a substantial risk of causing death or serious bodily harm”) (emphasis supplied).
Moreover, I would conclude that the dropping of the explosive device would have constituted the use of deadly force even had no fire ensued as a result of this act. The intention of the defendant officials in dropping the explosive device was to blast the bunker from atop the roof. This intention must be considered in light'of the fact that the defendant officials had no way of knowing whether the bunker was occupied at the time the explosive device was deployed. The only viable conclusion, given the bomb’s purpose and the reasonable (and unexplored) possibility that the bunker was occupied at the time the bomb was dropped, is that in dropping the explosive device, the defendant officials knew that they were creating a substantial risk of causing death or serious bodily harm. Had the defendant officials’ objective to blast the bunker from the roof been fulfilled, and had the bunker been occupied at the moment of impact, the bunker’s occupant or occupants would have been forcibly thrown from the roof of 6221 Osage to the street below. The substantiality of the risk of death or serious bodily injury from such a fall is self-evident and beyond debate.
In light of reason, experience and the applicable standard, I could not disagree more strenuously with Judge Greenberg’s conclusion that in dropping the explosive device onto the roof of an occupied building and in letting a fire spread upon that occupied building’s roof, the city defendants did not use deadly force. In addition, and putting aside the fact that a fire ensued, I cannot accept Judge Greenberg’s conclusion that in dropping the device to blast the bunker off of the roof, city officials did not know they were creating a substantial risk of death or serious bodily harm.
B.
The majority concludes that Garner is inapplicable and that “the appropriate inquiry is the reasonableness of the city defendants’ acts.” Maj.Op. at 965. I am perplexed by this conclusion, for in reaching its holding regarding the constitutionality of deadly force used in the course of an arrest, the Supreme Court in Garner applied and relied upon the reasonableness test embodied in the Fourth Amendment. See Garner, 471 U.S. at 7-12, 105 S.Ct. at 1699-1702 (“there can be no question that apprehension by the use of deadly force is a seizure subject to the reasonableness requirement of the Fourth Amendment”).
The majority implies that Garner and Graham are incompatible cases. In fact, Garner and Graham are complementary cases. As the district court noted, the Supreme Court in Gamer for the first time analyzed within the reasonableness framework of the Fourth Amendment an excessive force claim arising in the context of an arrest. See Africa v. City of Philadelphia, 809 F.Supp. 375, 379 (E.D.Pa.1992) (explaining that prior to Garner, excessive force claims were analyzed under the “shocks the conscience” standard). In doing so, the Court squarely addressed the very issue we confront here, namely, the constitutionality of the use of deadly force by law enforcement officials in the course of an arrest. In Graham, the Court merely “ma[de] explicit what was implicit in Garner’ s analysis,” by holding that “all claims that law enforcement officers have used excessive force — deadly or not — in the course *978of an arrest ... should be analyzed under the Fourth Amendment and its ‘reasonableness’ standard.... ” Graham, 490 , U.S. at 395, 109 S.Ct. at 1871 (emphasis in original). The Court then proceeded to apply the reasonableness test to the use of non-deadly force, as it had already done with regard to the use of deadly force in Garner.
Taken together, then, Garner and Graham establish the following set of complementary principles: (1) the reasonableness test under the Fourth Amendment is to be used when assessing the constitutionality of police employment of force in the context of an arrest; (2) when the force employed constitutes “deadly force,” the constitutionality of its use is to be'determined according to the reasonableness test set forth in Garner, and such force is deemed unreasonable- unless it is “necessary to prevent the [suspect’s] escape and the officer has probable cause to believe that the suspect poses a significant threat of death or serious physical injury to the officer or others[,]” Garner, 471 U.S. at 3, 105 S.Ct. at 1697;, and (3) when the force employed is non-deadly, the constitutionality of its use is to be assessed according to the factors enumerated in Graham. Thus, while Garner and Graham do establish a complementary set of principles, Garner is appropriately applied to situations, such as this one, involving the use of deadly force, whereas Graham speaks to and controls situations involving the use of non-deadly force. They share a focus upon the use of excessive force, but it is the nature and degree of that excessive force which dictates their respective applications.
Accordingly, while I agree with Judge Sci-riea that as a matter of law, a reasonable jury could, under Graham, conclude that the decision to employ the incendiary device was an excessive use of force, because I believe that this case involves the use of deadly force, and in light of my understanding of the relationship between Garner and Graham, I think the proper test for determining the constitutionality of the use of force in this, ease is found in Garner, and not in Graham.1
Our disagreement as to applicable test is not without considerable significance. As should be clear, if Graham is the appropriate standard by which to determine the constitutionality of the use of force here, then general reasonableness factors would guide a jury’s determination as to whether excessive force was used. By stark contrast, Garner would not invite a jury to be guided by the more flexible general reasonableness standard. Gamer imposes a stricter standard governing police conduct and the use of excessive force — and with good reason, since the intrusiveness of deadly force is qualitatively distinct from all other forms of excessive force. Accordingly, Garner defines and explains the reasonableness of the excessive force to which it is addressed — deadly force — in narrower terms. In other words, while the reasonableness inquiry is a common component with regard to both standards, that inquiry is, and should be, more precise and exacting when deadly force has been used. Thus, if Garner were applied, a jury would be asked more pointedly to determine whether the deadly force employed was reasonable because it was necessary to prevent the escape of suspects believed to pose a significant threat of death or serious physical injury to the police or others. It is this test, and not the more lenient Graham standard, by which the propriety of the law enforcement officers’ decisions in this case should be gauged. And, as I have already indicated, I believe that under Garner only one reasonable conclusion can be reached here: the city defendants used excessive force.
III.
THE MAJORITY’S ERRONEOUS CONCLUSION AS TO THE APPLICABILITY OF QUALIFIED IMMUNITY
The majority states that it has no difficulty concluding that the individual defendants are entitled to qualified immunity because they reasonably could have considered their conduct to be lawful. Maj.Op. at 969-70. I .disagree. I would hold that the individual *979defendants are not entitled to qualified immunity. Further, I believe that the source of the majority’s error here flows directly from its misunderstanding and misapplication of our construction of the term “clearly established law,” in the wake of the Supreme Court’s decision in Harlow v. Fitzgerald, 457 U.S. 800, 102 S.Ct. 2727, 73 L.Ed.2d 396 (1982), which first articulated that standard.
As the majority correctly observes, the relevant question here is whether a reasonable officer, possessing the same information as did the defendants, could have believed that the use of the incendiary device and fire to destroy the bunker was reasonable under the Fourth Amendment, that is, “lawful, in light of clearly established law” as of May 13, 1985. See Anderson v. Creighton, 483 U.S. 635, 640, 107 S.Ct. 3034, 3039, 97 L.Ed.2d 523 (1987). Accordingly, our task is to determine what the “clearly established law” governing the defendants’ actions was on that date. The majority argues that there was no clearly-established law on the date in question which would have required the officials to choose an alternative approach to executing the warrants. Maj.Op. at 971-72. In failing to grasp the breadth of the “clearly established law” standard, and in thus failing, once again, to recognize the applicability of Tennessee v. Garner to the events in question, the majority has erred.
As I have already noted, in Garner, which admittedly was decided a mere 48 days prior to the events in question,2 the Supreme Court held that where an arresting officer has probable cause to believe that the suspect poses a threat of death or serious physical harm, either to the officer or to others, it is not constitutionally unreasonable to prevent escape by using deadly force. Garner, 471 U.S. at 11, 105 S.Ct. at 1701.
, -
Thus, if the suspect threatens the officer with a weapon or there is probable cause to believe that he- [or she] has committed a crime involving the infliction or threatened infliction of serious physical harm, deadly force may be used if necessary to prevent escape, and if, where feasible, some warning has been given.
Garner, 471 U.S. at 11-12, 105 S.Ct. at 1701.
In concluding that Gamer bears an insufficient factual correspondence to be deemed the “clearly established law” applicable to this case, the majority runs counter to our traditional broad application of Harlow v. Fitzgerald's “clearly established law” standard. See, e.g., People of Three Mile Island v. Nuclear Reg. Com’rs, 747 F.2d 139, 144-45 *980(3d Cir.1984); Stoneking v. Bradford Area School District, 882 F.2d 720, 726 (3d Cir.1989). Although we have recognized that we cannot expect executive officials to anticipate the evolution of constitutional law, we do demand that they apply general, well-developed legal principles in analogous factual settings, i.e., in settings bearing some, but not necessarily precise, factual correlation to the applicable precedent. See People of Three Mile Island, 747 F.2d at 144.
Yet the situation we face here is identical to the situation the Supreme Court confronted in Garner: a ease of police use of deadly force in carrying out a seizure of a person or persons actively resisting arrest. Even in their particulars, Garner and our case are compellingly analogous. Obviously, this case involves the use of deadly force against a group of armed and dangerous adults barricaded inside a house, while Garner involved the use of deadly force against an unarmed teenager who was running from the police. However, it bears emphasizing that in our case several hours had elapsed between the cessation of gunfire from within the MOVE dwelling and the dropping of the incendiary device by the police. Like Garner, the police simply were not facing a “shoot or be shot” (or á “drop bomb or be shot”) situation at the moment the bomb was dropped. Indeed, this case is easier than Garner. In Garner, the Court determined that seizure by deadly force of a fleeing suspect was not worth the costs: it was better to let the suspect escape. Garner, 471 U.S. at 11,105 S.Ct. at 1701. In our case, the suspects were not even fleeing. At the time the destructive device was deployed, the MOVE contingent (including several children) was not only barricaded inside a house, but was also surrounded and greatly outnumbered by well-equipped law enforcement personnel. Thus, unlike Garner, escape here was not even a reasonable or realistic possibility. As the Court in Garner recognized, “[i]f subsequent arrest [is] assured, no one would argue the use of deadly force was justified.” Garner, 471 U.S. at 9 n. 8, 105 S.Ct. at 1700 n. 8.
Based upon these considerations, I would hold that reasonable officials in the defendants’ position at the relevant time could not have believed, in light of Garner, that their conduct would be lawful.
Even where the officials in question should have been aware of the “clearly established” governing law, we have explained that executive officials are still entitled to qualified immunity “if based on the information available to them they could have believed their conduct would be consistent” with that clearly established law. See Good, 891 F.2d at 1092. However, I believe it would have been readily apparent to any reasonable official who possessed the same information as did the individual defendants on May 13, 1985, that at the time the explosive device was dropped, deadly force was not necessary to prevent the escape of the occupants of 6221 Osage Avenue.
Support for the counter-argument, so far as I can gather, is summed up by the majority in the following factors: (1) nightfall would render securing the neighborhood too difficult; (2) fatigue was setting in among the police personnel who had been on duty since the night before; and (3) certain members of the police appear to have believed that tunnels had been dug from the basement of the MOVE house. See Maj.Op. at 950. With respect to the nightfall factor, the entire block, including 6221 Osage Avenue, could have been illuminated artificially as are highway projects regularly. One can counter that these lights could then have been shot out by MOVE members. But I think a modicum of police ingenuity could have seen to it that the lights were placed at angles rendering them unsusceptible to gunfire. With respect to police fatigue, this, one presumes, could reasonably have been counteracted by the replacement of tired officers with fresh personnel.
And finally, with regard to the rumored tunnels, support in the record for a reasonable belief in the existence of escape tunnels is flimsy at best. The majority suggests that the defendants believed that MOVE members might have eluded capture through escape tunnels rumored to have been dug under the neighborhood. I would agree that had the defendants reasonably believed that MOVE had constructed escape tunnels underneath the .neighborhood, the dropping of *981an explosive device might reasonably have been perceived as necessary to prevent the escape of the MOVE members under cover of the coming night. However, the record as I read it does not support the view that the defendants actually believed that MOVE had constructed escape tunnels under the neighborhood; at most, it merely suggests a belief, based upon a statement that fifteen bushels of dirt had been deposited .by MOVE members on a curb for disposal, that the MOVE members might have dug a tunnel to a neighboring house. ' First of all, digging an escape tunnel to a neighboring house is different from digging an escape tunnel “under the neighborhood” to some unknown outlet. But more importantly, even if the record supported the existence of such a belief, I would have no difficulty in concluding that it was unreasonable, indeed, far-fetched, in light of the evidence before us. It is fantasy to believe that fifteen bushels of dirt, even to a neighboring house, an escape tunnel makes.
Thus, I would hold that the individual defendants are not entitled to qualified immunity even when the factual information available to them on May 13, 1985, is taken into consideration.
IV.
Since I would hold that none of the individual defendants are entitled to qualified immunity in relation to the section 1983 claims against them, I would reverse the grant of summary judgment in favor of Goode, Powell and Klein as to the state law claims against these defendants.
V.
Judge Greenberg observes that in reaching my conclusion that the individual defendants used excessive force, I offer “no explanation of what alternatives were available,” and am “Apparently ... willing to have mandated an indefinite standoff.” See Maj.Op. at 967 n. 23. This is in part true and largely irrelevant. It is certainly true that I am willing to “mandate” that which is reasonable and lawful. An “indefinite standoff,” among other options, qualifies. But it is not our primary purpose here to provide alternatives to unlawful police action. Instead, we are to evaluate the lawfulness of what did occur on May 13, 1985. We should only countenance the lawful resolution of that unfortunate incident. Thus,- the larger point here is that Judge Greenberg, and to a lesser extent, Judge Scirica and I simply disagree, based upon our understanding of the record and the law, as to whether the dropping of the incendiary device was lawful.
In concluding that the defendant officials in this case used excessive force, I do not mean to imply that situations law enforcement officers confront daily are not fraught with difficulty and do not frequently require split-second decisions involving matters of life and death. To the contrary, law enforcement officers are forced to digest a steady diet of some of the most impossible choices under some of the most pressing circumstances and unforgiving conditions. We should, of course, be concerned for and sensitive to their fundamental mission, which is to ensure the protection and preservation of society.
But this, too, must be said: it is hardly debatable in a civilized constitutional democracy that in the final analysis, the goal of the enforcement of law, and thus, the maintenance of public order, is not promoted through the use of deadly force in effecting seizures. As the Supreme Court has observed, the use of deadly force, in fact, frustrates the interest in enforcement of law through fair and objective judicial determination of guilt and punishment. Garner, 471 U.S. at 9, 105 S.Ct. at 1700. It is partly because the use of .deadly force “is a self-defeating way of apprehending a suspect and so setting the criminal justice mechanism in motion” that the Constitution, as interpreted by the Supreme Court, so narrowly circumscribes its use. See Garner, 471 U.S. at 10, 105 S.Ct. at 1700.
The majority in Part VII states that “courts must recognize that in certain instances the civil authorities are required to take strong steps to enforce the law and maintain public order[,]” and that “[t]he Constitution does not preclude these steps.” Maj.Op. at 972.. The danger inherent in this statement, which the majority appears to *982■ adopt as a precept of constitutional jurisprudence, lies in its breadth. I am deeply troubled by the necessary implication of the majority’s statement, which is that the Constitution does not preclude that which is perceived as necessary or expedient to enforce the law and- maintain public order. I hold a different view. I believe it to be beyond dispute that the Constitution precludes many acts which might, even reasonably, be deemed “necessary” for the enforcement of law and the maintenance of public order. In determining what it does and does not preclude, we cannot engraft upon the Constitution our own predilections as to what that document, as a matter of perceived social necessity, ought or ought not permit law enforcement officials to do in the name of law and order.
CONCLUSION
For the above reasons, I respectfully concur and dissent.
. I agree with Judge Scirica that under Graham, a jury question exists as to whether excessive force was used in this case. However, as I have stated, I do not believe that Graham controls this case. I believe that Garner controls, and under Garner, it is clear to me that the deadly force used here was excessive as a matter of law and, therefore, unlawful.
. While the majority’s analysis does not depend upon or even address this point, I do not believe the close temporal proximily between the decision in Gamer and the events in question to be relevant or controlling. In a recent opinion, the Court of Appeals for the Sixth Circuit observed that while the circuits "have struggled to decide how long after a decision state officials have to become familiar with 'the law,' ... no rule has emerged.” Lintz v. Skipski, 25 F.3d 304, 306 (6th Cir.1994). That court went on to hold that a "rule of reason” should apply in each case with respect to official compliance with new decisions. Id. I believe it to be not only reasonable, but legally correct, that once the Supreme Court has announced a decision which governs the behavior of government or society, unless the Court states otherwise, it's holding becomes.the law of the land when the judgment is entered and is binding upon those affected by it. We have, in effect, acknowledged this in Good v. Dauphin County Social Services, 891 F.2d 1087, 1092 (3d Cir.1989) (stating that the ultimate issue in the "clearly established law” inquiry is whether, despite the absence of a case applying established principles to the same facts, "reasonable officials in the defendants' position at the relevant time could have believed, in light of what was in the decided case law, that their conduct would be lawful”) (emphasis supplied). Moreover, and perhaps more important, as the district, court noted, the decision in Tennessee v. Garner in significant respects mirrored, and in fact relied in part upon, section 508 of the Pennsylvania Criminal Code with respect to a peace officer’s use of force in making an arrest. See In re City of Philadelphia Litigation, 849 F.Supp. 331, 336 (E.D.Pa.1994). Thus, Garner hardly represented a novel twist in the law governing the defendants’ actions: that law was already on the books. In fact, it would appear that section 508 squarely covered the actions taken by the defendants here. See 18 Pa.C.S. § 508 (providing that a peace officer is justified in using deadly force "only when he [or she] believes that such force is necessary to prevent death or serious bodily injury to himself or ... other[s] ... or when he [or she] believes both that: (1) such force is necessary to prevent the arrest from being defeated by resistance or escape; and (2) the person to be arrested has committed or attempted to commit a forcible felony or is attempting to escape and possesses a deadly weapon, or otherwise indicates that he [or she] will endanger human life or inflict serious bodily injury unless arrested without delay”).