JAMES DICKSON PHILLIPS, Circuit Judge:
This appeal presents fundamental issues respecting the source and scope of constitutional rights of protection against, and statutory remedies for, the ■ excessive use of force by state police in making arrests. Here the district court, expressing reservations about both the source and scope of any such right and of any consequent remedy, and professing inability to apply what it considered to be this court’s unclear precedents in the matter, summarily dismissed, on legal grounds, such an excessive force claim brought under 42 U.S.C. § 1983. We reverse, on the basis that the fourth and fourteenth amendments provide general protection against such conduct, and we remand for further proceedings to determine whether the protection is available here.
I
In a
pro se
complaint, Dennis Ray Kidd alleged that in attempting to arrest him on April 15, 1983, defendants Robert O’Neill and Mike Lomonaco, Fairfax County, Virginia, police officers “brutally” and “severely” beat, kicked, and maced him while he was handcuffed, and that this resulted in bruises, a head gash requiring stitches, and continued headaches, dizziness and blurred vision. By answer and then by summary judgment affidavits, the police officer defendants admitted that one struck Kidd with a nightstick and that both maced him, but they asserted that they used only the force needed to subdue him; that Kidd was violently (and in the end, successfully) resisting arrest, that only one of his hands was handcuffed, and that with the other he was attempting to take a gun from one of the officers. By a responsive counter-affidavit, Kidd added minor factual details to his pleaded version of events, repeated in variant form his conclusory pleading allegations of “brutality” and “excessiveness,” and directly denied, albeit conclusorily, the defendants’ assertions that they acted reasonably and “in self-defense.”
On this state of the record, the district court entertained defendants’ motion for summary judgment, and granted it. Essentially declining to consider whether on the record there existed any genuine issue of material fact respecting the claim of unconstitutionally excessive force, the district court ruled that as a matter of law no cognizable claim of constitutional violation had been advanced by Kidd. This ultimate conclusion was based on the following line of reasoning, which, because of its importance to our decision, we summarize in some detail.
Not every act by a state agent that would constitute a violation of state tort law constitutes a deprivation of constitutional right simply because the act is committed by a state official,
citing, inter alia, Baker v. McCollan,
443 U.S. 137, 140 [99 S.Ct. 2689, 2692, 61 L.Ed.2d 433] (1979). Deciding when a “state tort” of battery becomes a “constitutional tort” cannot properly be done simply by attempting to assess the degree of its severity in terms of the state agent’s motivation and the victim’s harm. This has now been demonstrated by the failure of the Fourth Circuit to provide a “workable guideline,” a “practical standard,” for differentiating between a physical striking that amounts to a mere “state tort” and one that involves “constitutional deprivation.” This is illustrated by the standards articulated by that court respectively in
King v. Blankenship,
636 F.2d 70, 73 (4th Cir.1980) (excessive force in subduing convicted prisoner) and in
Hall v. Tawney,
621 F.2d 607, 613 (4th Cir.1980) (excessive force in disciplining public school student).
This suggests that the underlying theory of
Hall
and
King
is flawed, that the degree of severity of batteries by state agents, in terms of agent-motivation and victim harm, is not the proper test of constitutional deprivation. Indeed, to make degree of severity the test in such cases leads to the result forbidden by the Supreme Court of using § 1983 to “create a font of tort law,” citing
Paul v.
Davis,
424 U.S. 693, 701 [96 S.Ct. 1155, 1160, 47 L.Ed.2d 405] (1976).
The proper test of constitutional deprivation in state agent battery cases is rather to be found in a line of district court decisions out of the Eastern District of Virginia that specifically reject the perceived Fourth Circuit precedents as legally flawed and practically unworkable,
e.g., Dandridge v. The Police Department of the City of Richmond,
566 F.Supp. 152 (E.D.Va.1983) (excessive force in arrest);
Sellers v. Roper,
554 F.Supp. 202 (E.D.Va.1982) (excessive force in disciplining prison inmate). Under these district court decisions, a battery by a state agent amounts to a constitutional deprivation cognizable under 42 U.S.C. § 1983 only if “it infringes a specific constitutional right ... [and] [a] battery by a state officer can only infringe a specific constitutional right if the officer intended to infringe that right through the battery or could reasonably have foreseen ... the result,” citing
Dandridge,
566 F.Supp. at 160.
An example of such an “intended deprivation” of “specific right” would be a blow, of whatever degree, consciously intended by the state agent to retaliate for the victim’s taking legal action against the agent; this would constitute a deprivation of the victim’s specific [first amendment] right to petition for “a redress of grievances.” On the other hand, a “guard’s beating of a prisoner, standing alone, does not [constitute] a constitutional [deprivation],” no matter the severity, [presumably because “standing alone” such a “beating” is unrelated in purpose to any known, specific constitutional right].
On the authority of
Sellers,
554 F.Supp. 202, this critical requirement of “specific intent to deprive of a known constitutional right” in § 1983 physical battery cases is traceable to
Screws v. United States,
325 U.S. 91 [65 S.Ct. 1031, 89 L.Ed. 1495] (1945), and is confirmed in later Supreme Court decisions,
e.g., Ingraham v. Wright,
430 U.S. 651 [97 S.Ct. 1401, 51 L.Ed.2d 711] (1977);
Baker v. McCollan,
443 U.S. 137 [99 S.Ct. 2689, 61 L.Ed.2d 433] (1979);
Parratt v. Taylor,
451 U.S. 527 [101 S.Ct. 1908, 68 L.Ed.2d 420] (1981). In the instant case, it mandates dismissal of Kidd’s § 1983 claim because “there are no facts indicating the violation of a constitutional right or that the defendants intended such an infringement by the allegedly excessive force used in the arrest.”
J.A. 33-39.
Having disposed of the basic claim of a substantive constitutional violation under the foregoing analysis, the district court then held additionally that no procedural due process claim could be proved because Virginia provides an adequate post-deprivation remedy, for “assault and battery,” citing
Henderson v. Counts,
544 F.Supp. 149 (E.D.Va.1982). The action was accordingly dismissed as to all defendants and this appeal followed.
On appeal, Kidd does not challenge the district court’s rejection of any claim of procedural due process that might have been pleaded, and that question is accordingly not before us. Challenge is confined to the summary dismissal of his claim of deprivation of substantive constitutional right.
II
The district court’s legal analysis as applied in this case misinterpreted and misapplied binding precedent, not only of this court but of the Supreme Court. The resulting legal error requires reversal and remand for further proceedings. The court’s error has two principal aspects.
The first is the failure to recognize the fourth amendment, through the fourteenth,
as a direct source of constitutional protection against uses of excessive physical force by state police officers in arresting suspects. If any doubt has existed that the fourth amendment’s prohibition against unreasonable seizures of the person protects against the use of excessive force in making arrests as well as against making arrests without probable cause, it has now been laid to rest by the Supreme Court’s decision in
Tennessee v. Garner,
— U.S. —, 105 S.Ct. 1694, 85 L.Ed.2d 1 (1985) (§ 1983 claim for “seizure” of person by deadly force).
The
Garner
Court flatly rejected the argument that the amendment’s protection only runs to the probable cause requirement, holding that “reasonableness depends on not only when a seizure is made, but also how it is carried out.” — U.S. at —, 105 S.Ct. at 1699. The Court pointed out that the contrary argument “ignores the many cases in which [the Court] ... has examined the reasonableness of the manner in which a search or seizure is conducted.” — U.S. at —, 105 S.Ct. at 1699.
Not only did the holding in
Garner
make explicit what has long been at least implicit in such earlier decisions of the Supreme Court as
Terry v. Ohio,
392 U.S. 1, 28-29, 88 S.Ct. 1868, 1888-84, 20 L.Ed.2d 889 (1968) (cited for proposition by
Garner
Court, — U.S. at —, 105 S.Ct. at 1699), it also confirmed long-standing authority to the same effect in this circuit.
See Jenkins v. Averett,
424 F.2d 1228, 1232 (4th Cir.1970) (Sobeloff, J.) (“shield” of the fourth amendment protects “individual physical integrity” against “wanton” infliction of injury by police in attempt to halt fleeing suspect).
To the extent therefore that the district court’s analysis involved the perception that excessive force, “standing alone,” in effecting an arrest implicates no specific constitutional right, the court’s decision is at odds not only with circuit but with Supreme Court precedent. Such police conduct may indeed constitute both a “mere” state law tort and á deprivation of fourth amendment rights; in such circumstances § 1983 provides a cumulative federal civil remedy unaffected by the availability of any state law remedy.
Monroe v. Pape,
365 U.S. 167, 183, 81 S.Ct. 473, 481, 5 L.Ed.2d 492 (1961).
The second aspect of the court’s error is the related perception that § 1983 only provides a remedy for excessive use of physical force by state agents (presumably whether or not in arrest situations) if the agents “intended to infringe a specific constitutional right through the battery or could reasonably have foreseen ... the result.” This state-of-mind requirement, whose origin was ascribed by the district court to
Screws v. United States,
325 U.S. 91, 65 S.Ct. 1031, 89 L.Ed. 1495 (1945), has long since been rejected by the Supreme Court as being inapplicable to civil actions under § 1983. The rejection indeed occurred at the very outset of the Supreme Court’s development of contemporary § 1983 jurisprudence, in
Monroe v. Pape,
365 U.S. 167, 81 S.Ct. 473, 5 L.Ed.2d 492 (1961). To the suggestion there advanced that, following
Screws,
§ 1983 should only reach unauthorized conduct of state officials that was “willful” in the sense that it was intended to deprive of a constitutional right the Court said:
[i]n the
Screws
case we dealt with a statute [18 U.S.C. § 242] that imposed criminal penalties for acts “wilfully” done. We construed that word in its setting to mean the doing of an act with “a specific intent to deprive a person of a federal right.” 325 U.S. at 103. We do not think that gloss should be placed on [§ 1983] which we have here. The word “wilfully” does not appear in [§ 1983]. Moreover, [§ 1983] provides a civil remedy, while in the
Screws
case we dealt with a criminal law challenged on the ground of vagueness. [Section 1983]
should be read against the background of tort liability that makes a man responsible for the natural consequences of his actions.
365 U.S. at 187, 81 S.Ct. at 484.
See also Parratt v. Taylor,
451 U.S. 527, 534-35, 101 S.Ct. 1908, 1912-13, 68 L.Ed.2d 420 (1981) (quoting
Monroe
as support for hold: ing that § 1983 “has never been found by [the Supreme] Court to contain a state-of-mind requirement”). The district court’s latter-day importation of the
Screws
scien-ter requirement into § 1983 jurisprudence was therefore legally erroneous and further flaws its dismissal of the claims against the police officers.
Ill
Having indicated specifically wherein we think the district court erred in its legal analysis of the constitutional source and statutory remedy for excessive-force arrests, we turn to the proper analysis for application by the court on remand.
That analysis is most appropriately found in
Tennessee v. Garner,
as the Supreme Court’s most recent and apposite authority on the fourth amendment as source of constitutional right and § 1983 as the means of remedy for violation of the right.
Garner,
dealing specifically only with the extreme example of deadly force, nevertheless makes clear that the use of any significant degree of excessive force in effecting otherwise constitutionally valid arrests may constitute an unreasonable seizure of the person in violation of fourth amendment rights. Whether a particular arrest is constitutionally unreasonable in this sense inescapably depends upon the factual circumstances.
Garner,
— U.S. at —, 105 S.Ct. at 1699 (presents “totality of circumstances” question). To determine the question requires that a court “balance the nature and quality of the intrusión on the individual’s Fourth Amendment interests against the importance of the governmental interests alleged to justify the intrusion.”
Garner,
— U.S. at —, 105 S.Ct. at 1699 (quoting
United States v. Place,
462 U.S. 696, 703, 103 S.Ct. 2637, 2642, 77 L.Ed.2d 110 (1983).
Garner
identified the state’s primary interest as being the apprehension of suspects so as to “set[] the criminal justice mechanism in motion.”
Garner,
— U.S. at —, 105 S.Ct. at 1700. Balancing this interest against the fourth amendment rights of suspects, the
Garner
Court recognized that under some circumstances even deadly force would not be constitutionally unreasonable.
Garner,
— U.S. at — - —, 105 S.Ct. at 1700-01. Implicit in this is the further proposition that varying degrees of force short of, but up to, deadly force would also not be constitutionally unreasonable.
Obviously, the legitimate interest of the state in apprehending suspects is threatened whenever the suspect forcibly resists arrest or, with or without force, attempts to avoid arrest by fleeing. To accommodate this reality,
Garner
recognizes, for example, that “if the suspect threatens the officer with a weapon ... deadly force may be used if necessary to prevent escape.”
Id.
On the other hand, the use of deadly force to overcome milder forms of resistance, or to prevent nonforcible flight where there was no probable cause to fear danger to others from the fleeing suspect, might be constitutionally unreasonable.
Id.
(Tennessee statute held unconstitutional as applied to authorize deadly force in latter circumstance.) By necessary implication from
Garner’s
analysis then, the use of any significant force, up to and including deadly force, not reasonably necessary to effect an arrest — as where the suspect neither resists nor flees
or where the force is used after a suspect’s resistance has been overcome or his flight thwarted—would be constitutionally unreasonable.
Whether the force used in this case in attempting to arrest Kidd constituted a constitutionally unreasonable seizure of his person must be decided under this general standard. Because that standard has not yet been applied, the case must be remanded for its application to the facts as they may be established in the district court.
IV
Though the dispositive essence of the district court’s error—failure to recognize the fourth amendment’s specific applicability and the imposition of a willfulness requirement—has been sufficiently identified, the error’s basis ran wider. Because of the resulting potential for correspondingly wider future applications of the error, we think it appropriate also to address its wider basis. In the process, we take the opportunity to put some of our decisions respecting § 1983 excessive-force claims in what may be a better, more integrated perspective than appeared to the district court.
As indicated, the district court, picking up on the
Sellers-Dandridge
analysis of those decisions that it considered the most apposite (though flawed) circuit authority respecting the claim here at issue, looked to
Hall v. Tawney,
621 F.2d 607 (4th Cir.1980) and
King v. Blankenship,
636 F.2d 70 (4th Cir.1980). As our discussion has now revealed, those two decisions were not the most apposite to the specific fourth amendment claim here in issue. Neither involved a claim of excessive force in an arrest situation:
Hall
dealt with an allegedly brutal beating of a public school child;
King,
with an allegedly excessive use of force in subduing a convicted prisoner. The most directly apposite circuit authority was, as noted, Judge Sobeloff’s panel opinion in
Jenkins v. Averett,
424 F.2d 1228 (4th Cir.1970), which pre-dated both
Hall
and
King
and dealt specifically, on fourth amendment grounds, with a claim of excessive force in arresting a fleeing suspect. For whatever reason, the district court did not consider
Jenkins’
most obvious relevance.
We point this up not gratuitously to chide the district court, but as a basis for attempting now to put those decisions and this one in proper perspective and relationship under developed 42 U.S.C. § 1983 jurisprudence. We start by observing that they all have this in common: . each involved a § 1983 claim that a state agent acting under color of state law used excessive force that inflicted bodily harm upon a claimant. They differ, however, and critically, in the factual contexts in which force was applied:
Jenkins
and the instant case involved the arrests of resisting or fleeing suspects;
Hall,
the corporal punishment of a public school student;
King,
the forcible subduing of a convicted prisoner. Within those different factual contexts, however, there is again a critical similarity: in each, the state agent’s use of force was arguably a privileged use in furtherance of a legitimate state interest.
Taking into account the critical differences and similarities in these cases in light of evolving § 1983 jurisprudence, we think the following legal synthesis emerges. The legal right asserted in each case is the right to be free of bodily harm, of physical abuse, at the hands of government’s agents. In terms of common law origins, the core right is that recognized in the tort and crime of battery. The vexing conceptual problem in the evolution of § 1983 “constitutional tort” jurisprudence has been whether, and if so where, the constitution provides specific protection against the infliction of bodily harm by state agents, i.e., whether there is any constitutional, as opposed to state common law, right to be free of such inflicted harms.
No provision of the bill of rights speaks literally in terms of bodily harm or personal security or the like. But two amendments, the fourth and the eighth, do so clearly prohibit conduct that may involve bodily harm — “unreasonable seizures” of the “person,” (fourth), “cruel ... punishments” (eighth) — that they have been held to embody specific protections, vindicable under § 1983, against the infliction of bodily harm in those contexts.
Tennessee v. Garner, supra
(fourth amendment);
Estelle v. Gamble,
429 U.S. 97, 97 S.Ct. 285, 50 L.Ed.2d 251 (1976) (eighth amendment);
see also Jackson v. Bishop,
404 F.2d 571 (8th Cir.1968) (same). Because no other provision of the bill of rights speaks directly, as do the fourth and eighth amendments, to conduct that may cause bodily harm, it has been possible to question whether the constitution provides any other direct protection against the infliction of bodily harm “standing alone,” i.e., unrelated to the deprivation of other specifically secured rights.
This presumably was the view applied by the district court here, either overlooking that in the arrest situation the fourth amendment does provide direct protection, or considering that even its protection runs only to “willful” deprivations.
See
Part II,
supra.
An interpretation that the fourth and eighth amendments, but no other bill of rights provisions, directly proscribe the infliction of bodily harm by agents of the state of course leads to palpable practical and constitutional anomalies. Under such an interpretation, constitutional protections against this most egregious form of governmental intrusion exists only in those persons being taken into custody as criminal suspects and those already convicted of crime. On this view no one else in society has the protection, not even those persons in between arrest and conviction as pre-trial detainees, not to say all those sufficiently virtuous or lucky not to have created any probable cause for their arrest and prosecution in the first place. While, as indicated, it is obvious that persons in arrest and post-conviction settings may have the greatest statistical need for this particular protection, it simply defies belief that the framers of the bill of rights considered that only those persons were to have the constitution’s protection against physical brutality.
Reflecting this perception, there has been a conceptual struggle ever since revival of § 1983 to identify the source of a general constitutional right to bodily security comparable to that given special protection in the fourth and eighth amendments and which could be asserted under § 1983 by persons other than arrest suspects and convicts. Perhaps predictably, notions of “substantive due process” judicially devel
oped to void criminal convictions obtained by outrageously “inhumane” police conduct were early drawn upon to extend § 1983’s remedy to persons unable to invoke the specific protections of the fourth or eighth amendments. Judge Friendly’s influential opinion in
Johnson v. Glick,
481 F.2d 1028 (2d Cir.1973), did this, drawing on the “substantive due process” analysis in the “stomach-pump” case,
Rochin v. California,
342 U.S. 165, 72 S.Ct. 205, 96 L.Ed. 183 (1952), to find a claim of guard brutality against a pre-trial detainee cognizable under § 1983, and noting in the process that the claimed protection could not be found in either the fourth or eighth amendment or in any “ ‘specific’ of the Bill of Rights.”
Id.
at 1032.
Notwithstanding the suggestion in
Johnson v. Glick
that the bodily security right there recognized lay in no “specific” Bill of Rights source, later Supreme Court decisions have now identified a specific source in the fifth amendment’s fundamental guarantee of “liberty.”
In
Ingraham v. Wright,
430 U.S. 651, 97 S.Ct. 1401, 51 L.Ed.2d 711 (1977), a § 1983 action claiming,
inter alia,
a denial of procedural due process in the infliction of corporal punishment on a public school student, the Court had first to address the specific question whether a constitutionally protected liberty interest was implicated. It was, held the Court, finding the source in the “liberty preserved from deprivation” in the due process clause of the fifth amendment as made applicable to the states through the fourteenth.
Id.
at 672-73, 97 S.Ct. at 1413.
Holding specifically that “[a]mong the historic liberties so protected was a right to be free from ... unjustified intrusions on personal security,”
id.
at 673, 97 S.Ct. at 1413, the Court noted that the fourth amendment’s unclaimed special protection of the same interests was related, but that its “principal concern” was “with intrusions ... in the course of criminal investigations.”
Id.
at 673 n. 42, 97 S.Ct. at 1413 n. 42. In the context of corporal punishment of school children outside the course of criminal investigations, the Court defined the liberty interest at stake as that of being free of unreasonable punishment as measured by the teacher’s common law privilege.
Id.
at 674, 97 S.Ct. at 1414.
Later, in
Bell v. Wolfish,
441 U.S. 520, 99 S.Ct. 1861, 60 L.Ed.2d 447 (1979), the Court recognized that the same liberty interest in bodily security, grounded in the fifth amendment’s due process clause, existed in
federal pre-trial detainees.
Id.
at 535, 99 S.Ct. at 1871;
see also id.
at 580-81, 99 S.Ct. at 1895-96 (Stevens, J., dissenting) (agreeing that source of interest recognized is “the word ‘liberty’ itself as used in the Due Process Clause”). In respect of such persons, the specific interest at stake is that in being free of “punishment,”
id.
at 535, 99 S.Ct. at 1871. In the context of valid pre-trial detention this requires inquiry into whether a particular bodily intrusion was “imposed for the purpose of punishment or ... as an incident of some other legitimate governmental purpose.”
Id.
at 538, 99 S.Ct. at 1873.
Garner, Gamble, Ingraham
and
Wolfish,
in conjunction therefore have recognized cognate liberty interests in “personal security” grounded respectively in the specific guarantees of the fourth and eighth amendments and in the “liberty” given protection against deprivation without due process in the fifth amendment. While the interests are cognate, they are not perfectly congruent, for each is qualified by the existence of legitimate but different governmental interests that may justify forcible intrusions of varying degrees upon the interest by government officials. Under the fourth amendment, the intrusion may be “reasonable” in relation to the need to effect a particular arrest; under the eighth amendment, it may not amount to “cruel” or “unusual” “punishment.” Under the fundamental “liberty” interest recognized in the fifth amendment, it may not amount to “punishment” of pre-trial detainees or to “unreasonable punishment” of public school students.
This means of course that “excessive force” claims may lie under § 1983 to vindicate either fourth, eighth, or fifth amendment violations of the cognate right to “personal security” in any of the specific factual contexts dealt with in those decisions and certainly in others not touched upon.
Our decisions in
Jenkins
(fourth amendment arrest);
King
(physical subduing of convicted inmate);
Hall
(fifth amendment corporal punishment of student),
are therefore all consistent with
these Supreme Court authorities in their recognition of constitutionally protected interests in “personal security.” Because in each context the interest was qualified by the existence of countervailing governmental interests arguably justifying some degree of forcible intrusion, the issue of violation in each was inevitably one of “degree,” of permissible force under the circumstances.
There is, in consequence, no escaping the necessity lamented by the district court in the instant case to grapple with the "degree” of force applied in determining § 1983 excessive force claims in any context — such as those involved in
Hall,
in
King,
and in
Jenkins
and the instant case — where the force may be “privileged” as furthering a legitimate state interest. Our formulátion of constitutional standards in those cases, employing such terms as “inhumane,” “malicious,” “sadistic,” “shocking to the conscience,” etc., are properly understood therefore as descriptives of degrees of force that inevitably exceed those degrees “privileged” in the various contexts by the legitimate state interests being served.
Inquiries into degree and circumstance are not — as the district court saw it — mandated to differentiate between a “mere state tort” and a “constitutional deprivation,” but to determine whether a deprivation of constitutional right to “personal security” has occurred in a context where some degree of force may be justified. As the most apposite Supreme Court
decisions
—Garner,
Wolfish, Ingraham
—all illustrate, “personal security” claims in any of these contexts require the adjudication of constitutional rights on a case-by-case basis, with the factual circumstances a major component of the adjudication.
The district court therefore erred broadly in its stated perception that questions of degree are not relevant to proper constitutional inquiry in these related excessive force adjudications under § 1983. That broader error of analysis flawed its specific determination here that a factual inquiry into the degree of force in Kidd’s arrest was not required to determine his § 1983 claim.
REVERSED AND REMANDED.