holding that employees of private prison were not subject to claims under Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics, 403 U.S. 388Summary of this case from Murray v. Keller
Argued September 19, 2005.
Decided January 12, 2006.
Appeal from the United States District Court for the Eastern District of North Carolina, Louise W. Flanagan, J.
Mark Allen Davis, Womble, Carlyle, Sandridge Rice, Raleigh, North Carolina, for Appellants. James Phillip Griffin, Jr., North Carolina Prisoner Legal Services, Inc., Raleigh, North Carolina, for Appellee.
Before WILKINSON and MOTZ, Circuit Judges, and R. BRYAN HARWELL, United States District Judge for the District of South Carolina, sitting by designation.
Reversed by published opinion. Judge WILKINSON wrote the opinion, in which Judge HARWELL joined. Judge MOTZ wrote a separate opinion concurring in the judgment.
We granted interlocutory review in this case to decide whether individual employees of a privately operated prison face Eighth Amendment liability under Bivens v. Six Unknown Named Agents of Federal Bureau of Narcotics, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971), and its progeny for allegedly providing inadequate medical care to a federal inmate. We decline to extend the Bivens cause of action to these circumstances, both because the actions of the private prison employees are not fairly attributable to the federal government and because the inmate has adequate remedies under state law for his alleged injuries. We therefore reverse the district court's order denying defendants' motion to dismiss.
Plaintiff Ricky Lee Holly is a federal inmate at Rivers Correctional Institution, a privately run facility in North Carolina operated by the GEO Group, Inc. under contract with the federal Bureau of Prisons. Defendant Willie Scott serves as Rivers' warden, and defendant Gaddy Lassiter works as a physician there. Scott and Lassiter are both employed directly by GEO, and thus the only link between their employment and the federal government is GEO's contract with the BOP.
Holly is a diabetic. He contends that since arriving at Rivers in August 2002, defendants have failed to provide him with adequate medical care for his condition. According to Holly, the medical staff at Rivers has ignored his complaints that his insulin dosage was insufficient, which resulted in frequent blackouts. He suggests that the lack of attention to his medical needs stems, at least in part, from failure to request his medical records from another facility where he had been incarcerated previously. He further claims that in retaliation for a written complaint regarding the medical department, Lassiter ordered that he be locked in the medical unit for twenty-four days and threatened to keep him there for the remainder of his sentence.
After unsuccessfully seeking relief through an administrative scheme provided by Rivers, Holly filed a pro se complaint against Scott and Lassiter in federal district court. The district court read Holly's complaint as alleging a violation of his Eighth Amendment rights and stating a Bivens cause of action for damages.
Defendants filed a motion to dismiss under Federal Rules of Civil Procedure 12(b)(1) and 12(b)(6), arguing that Bivens and its progeny do not support a cause of action against individual employees of a private correctional facility. The district court denied the motion, finding that Holly's claim satisfied the three preconditions for a Bivens remedy as specified in Hall v. Clinton, 235 F.3d 202, 204 (4th Cir. 2000). Congress had neither created an exclusive statutory remedy nor expressly precluded money damages, and the district court found no "special factors counselling hesitation in the absence of affirmative action by Congress." Hall, 235 F.3d at 204 (internal quotation marks omitted). We subsequently granted defendants' petition for an interlocutory appeal.
We review de novo a district court's denial of a motion to dismiss under Rule 12(b)(6). Anita's New Mexico Style Mexican Food, Inc. v. Anita's Mexican Foods Corp., 201 F.3d 314, 319 (4th Cir. 2000).
There is no need for us to review the denial of defendants' 12(b)(1) motion, because our jurisdiction over Holly's complaint is readily apparent. Whether a remedy exists for a plaintiff alleging a violation of his constitutional rights is itself a question of federal law sufficient to confer federal jurisdiction. See Bell v. Hood, 327 U.S. 678, 683-85, 66 S.Ct. 773, 90 L.Ed. 939 (1946).
In Bivens, the Supreme Court held that "violation of [the Fourth Amendment] by a federal agent acting under color of his authority gives rise to a cause of action for damages," despite the absence of any federal statute creating liability. 403 U.S. at 389, 91 S.Ct. 1999. Carlson v. Green, 446 U.S. 14, 100 S.Ct. 1468, 64 L.Ed.2d 15 (1980), extended Bivens to recognize an implied damages action against federal prison officials for violation of the Eighth Amendment. The question in this case is whether Carlson should itself be extended to allow a similar remedy against employees of a private corporation operating a prison.
Holly contends that a judicially implied cause of action for damages in his case follows logically from Bivens and Carlson. We disagree. The Bivens cause of action is not amenable to casual extension. Indeed, quite the opposite is true.
The Supreme Court has "responded cautiously to suggestions that Bivens remedies be extended into new contexts." Schweiker v. Chilicky, 487 U.S. 412, 421, 108 S.Ct. 2460, 101 L.Ed.2d 370 (1988). "In [over] 30 years of Bivens jurisprudence [the Court has] extended its holding only twice." Correctional Servs. Corp. v. Malesko, 534 U.S. 61, 70, 122 S.Ct. 515, 151 L.Ed.2d 456 (2001). The first time was in Davis v. Passman, 442 U.S. 228, 99 S.Ct. 2264, 60 L.Ed.2d 846 (1979), where the Court recognized that a female deputy administrative assistant could claim damages under the Fifth Amendment's Due Process Clause against a Congressman who had fired her on the basis of her gender. The second was Carlson's extension of Bivens to Eighth Amendment suits against federal prison officials. 446 U.S. 14, 100 S.Ct. 1468, 64 L.Ed.2d 15.
Since the Carlson decision in 1980, the Court has consistently declined to extend Bivens beyond these well-demarcated boundaries. See Bush v. Lucas, 462 U.S. 367, 103 S.Ct. 2404, 76 L.Ed.2d 648 (1983) (no Bivens action for employee who suffered adverse employment action allegedly in violation of First Amendment because Congress had provided remedial scheme); United States v. Stanley, 483 U.S. 669, 107 S.Ct. 3054, 97 L.Ed.2d 550 (1987) (no Bivens action arising out of military service); Schweiker, 487 U.S. 412, 108 S.Ct. 2460, 101 L.Ed.2d 370 (no Bivens action for alleged due process violation in denying Social Security benefits because Congress had set up a remedial scheme); Fed. Deposit Ins. Corp. v. Meyer, 510 U.S. 471, 114 S.Ct. 996, 127 L.Ed.2d 308 (1994) (no Bivens action against federal agency); Malesko, 534 U.S. 61, 122 S.Ct. 515, 151 L.Ed.2d 456 (no Bivens action against private corporation operating federal correctional facility).
The Court's repeated reluctance to extend Bivens is not without good reason. A Bivens cause of action is implied without any express congressional authority whatsoever. This is hardly the preferred course. The Supreme Court has "recently and repeatedly said that a decision to create a private right of action is one better left to legislative judgment in the great majority of cases." Sosa v. Alvarez-Machain, 542 U.S. 692, 124 S.Ct. 2739, 2762-63, 159 L.Ed.2d 718 (2004); see also Erie R.R. Co. v. Tompkins, 304 U.S. 64, 58 S.Ct. 817, 82 L.Ed. 1188 (1938) (abandoning the idea of a substantive federal common law). The Court has therefore on multiple occasions declined to extend Bivens because "Congress is in a better position to decide whether or not the public interest would be served" by the creation of "new substantive legal liability." Schweiker, 487 U.S. at 426-27, 108 S.Ct. 2460 (internal quotation marks omitted); Bush, 462 U.S. at 390, 103 S.Ct. 2404 (same).
Congress possesses a variety of structural advantages that render it better suited for remedial determinations in cases such as this. Unconstrained by the factual circumstances in a particular case or controversy, Congress has a greater ability to evaluate the broader ramifications of a remedial scheme by holding hearings and soliciting the views of all interested parties. See, e.g., Bush, 462 U.S. at 389, 103 S.Ct. 2404. And by debating policies and passing statutes rather than deciding individual cases, Congress has increased latitude to implement potential safeguards — e.g., procedural protections or limits on liability — that may not be at issue in a particular dispute.
As the last twenty-five years of Bivens jurisprudence demonstrate, so well-suited is Congress to determine the policies pertaining to a remedial scheme that neither the absence nor the incompleteness of such a scheme represents an invitation for a court to step in to correct what it may perceive as an injustice toward an individual litigant. See Malesko, 534 U.S. at 69, 122 S.Ct. 515 (citing Bush and Lucas). Here, for example, there are a variety of statutes authorizing the housing of federal inmates in privately operated facilities. See, e.g., 18 U.S.C. § 4013(b) (2000). Congress passed these statutes in the belief that private management would in some circumstances have comparative advantages in terms of cost, efficiency, and quality of service. To add a federal damages remedy to existing avenues of inmate relief might well frustrate a clearly expressed congressional policy.
In light of the governing precedents, we have stated that a plaintiff seeking a Bivens remedy must satisfy a three-part test. "[A] court must determine that (1) Congress has not already provided an exclusive statutory remedy; (2) there are no special factors counselling hesitation in the absence of affirmative action by Congress; and (3) there is no explicit congressional declaration that money damages not be awarded." Hall, 235 F.3d at 204 (internal quotation marks omitted). While the first and third prongs are satisfied by Congress's silence regarding remedies for plaintiffs in Holly's position, Holly cannot satisfy the second.
This case presents two "special factors counselling hesitation," each of which independently precludes the extension of Bivens. First, defendants are private individuals, not government actors. Second, Holly has an adequate remedy against defendants for his alleged injuries under state law.
It is in fact an understatement to consider these "special factors counselling hesitation." Governmental action and the lack of another legal remedy against individual defendants represent critical justifications for the very existence of the Bivens doctrine. To judicially infer a cause of action where these elements are absent would be to release that doctrine from its moorings and cast it adrift.
The first factor counseling hesitation is that defendants' actions are not "fairly attributable" to the federal government. Lugar v. Edmondson Oil Co., 457 U.S. 922, 937, 102 S.Ct. 2744, 73 L.Ed.2d 482 (1982). It is uncontested that defendants are employees of GEO, a wholly private corporation in which the federal government has no stake other than a contractual relationship. We decline to extend Bivens to impute liability in such circumstances.
Whatever our power to recognize new constitutional torts, we are not free to ignore the importance of a party's private status in our constitutional scheme. The Bill of Rights is a negative proscription on public action — to simply apply it to private action is to obliterate "a fundamental fact of our political order." Lugar, 457 U.S. at 937, 102 S.Ct. 2744. Statutory and common law, rather than the Constitution, traditionally govern relationships between private parties. See Edmonson v. Leesville Concrete Co., 500 U.S. 614, 619, 111 S.Ct. 2077, 114 L.Ed.2d 660 (1991). By placing limits upon the Constitution's application, we "preserve an area of individual freedom by limiting the reach of federal law and federal judicial power." Lugar, 457 U.S. at 936, 102 S.Ct. 2744. This area of personal freedom is one of the important structural features of our founding document. See Edmonson, 500 U.S. at 619, 111 S.Ct. 2077.
Neither we nor the Supreme Court has had occasion to consider whether and to what extent Bivens liability may apply to private citizens. As a threshold matter, we harbor some doubt as to whether such liability would ever be appropriate. Each of the defendants in Bivens, Davis, and Carlson were in the direct employ of the federal government. See Bivens, 403 U.S. at 389, 91 S.Ct. 1999 (liability for "agents of the Federal Bureau of Narcotics"); Davis, 442 U.S. at 230, 99 S.Ct. 2264 (liability for "United States Congressman"); Carlson, 446 U.S. at 16, 100 S.Ct. 1468 (liability for "federal prison officials"). As we have already discussed, there are ample reasons for the Court's reluctance to expand the boundaries of this judicially created cause of action beyond where those cases have placed them. Moreover, the Supreme Court has recently stated that "[t]he purpose of Bivens is to deter individual federal officers from committing constitutional violations." Malesko, 534 U.S. at 70, 122 S.Ct. 515 (emphasis added). It is unclear how permitting a lawsuit against an individual who is not a federal officer could serve this purpose.
The Supreme Court's most recent Bivens decision noted that the question of "whether a Bivens action might lie against a private individual" was not presented. Malesko, 534 U.S. at 65, 122 S.Ct. 515. In that case, the Court held that an inmate in a private correctional facility could not bring a Bivens suit against the corporation operating the facility. 534 U.S. at 63, 122 S.Ct. 515. While the Court's logic rested in part on its conclusion that Bivens creates individual liability rather than corporate liability, this does not answer the expressly withheld question whether Bivens creates liability for private individuals.
We need not decide this issue now, however. Even assuming that Bivens liability is sufficiently expansive to encompass at least some private individuals, our constitutional scheme necessarily constrains its scope. In the context of constitutional claims under 42 U.S.C. § 1983, which expressly applies to individuals acting "under color of" state law, courts have recognized the need to limit the liability of private persons through application of the "state action" doctrine. Lugar, 457 U.S. at 935, 102 S.Ct. 2744. Under this doctrine, we "insist " as a prerequisite to liability "that the conduct allegedly causing the deprivation of a federal right be fairly attributable to the State." Id. at 937, 102 S.Ct. 2744. By doing so, we maintain the Bill of Rights as a shield that protects private citizens from the excesses of government, rather than a sword that they may use to impose liability upon one another. Cf. Moose Lodge No. 107 v. Irvis, 407 U.S. 163, 172, 92 S.Ct. 1965, 32 L.Ed.2d 627 (1972). "Without a limit such as this," the constitutional order would be disrupted because "private parties could face constitutional litigation whenever they seek to rely on some state rule governing their interactions with the community surrounding them." Lugar, 457 U.S. at 937, 102 S.Ct. 2744.
There exists ample reason to be even more cautious about imputing liability to private actors under Bivens than under § 1983. Section 1983 is a congressional enactment that expressly creates liability for "[e]very person who, under color of any statute, ordinance, regulation, custom, or usage, of any State or Territory or the District of Columbia" deprives an individual of any "rights, privileges, or immunities secured by the Constitution" or other federal law.
Bivens, on the other hand, is a device of judicial creation. Although it shares some features with § 1983 — e.g., the qualified immunity analysis is the same under both, see Butz v. Economou, 438 U.S. 478, 504, 98 S.Ct. 2894, 57 L.Ed.2d 895 (1978) — the Supreme Court "ha[s] never expressly held that the contours of Bivens and § 1983 are identical." Malesko, 534 U.S. at 82, 122 S.Ct. 515 (Stevens, J., dissenting). Application of Bivens to private individuals simply does not find legislative sanction. Under such circumstances, the danger of federal courts failing "to respect the limits of their own power," Lugar, 457 U.S. at 936-37, 102 S.Ct. 2744, increases exponentially.
It is an open question in this circuit whether § 1983 imposes liability upon employees of a private prison facility under contract with a state. We need not decide that issue here.
We have recognized that there is "no specific formula" for determining whether state action is present. Hicks v. S. Md. Health Sys. Agency, 737 F.2d 399, 402 n. 3 (4th Cir. 1984) (internal quotation marks omitted). "What is fairly attributable [to the state] is a matter of normative judgment, and the criteria lack rigid simplicity." Brentwood Acad. v. Tenn. Secondary Sch. Athletic Ass'n, 531 U.S. 288, 295, 121 S.Ct. 924, 148 L.Ed.2d 807 (2001). While a variety of factors may bear upon the inquiry, none is individually dispositive; instead, they serve to inform an evaluation of the "totality of the circumstances." Goldstein v. Chestnut Ridge Volunteer Fire Co., 218 F.3d 337, 343 (4th Cir. 2000); see also Mentavlos v. Anderson, 249 F.3d 301, 311-12 (4th Cir. 2001) (cataloging various approaches to determining state action).
The alleged actions of these defendants were not of a sufficiently federal character to create constitutional liability. Defendants are not federal officials, federal employees, or even independent contractors in the service of the federal government. Instead, they are employed by GEO, a private corporation. There is no suggestion that the federal government has any stake, financial or otherwise, in GEO. See, e.g., DeBauche v. Trani, 191 F.3d 499, 507-08 (4th Cir. 1999) (recognizing the high degree of interdependence necessary for private enterprise to become state action). Nor is there any suggestion that federal policy played a part in defendants' alleged failure to provide adequate medical care, or that defendants colluded with federal officials in making the relevant decisions. See, e.g., Hicks, 737 F.2d at 403 (finding private entity was not a federal actor where government did not regulate relevant procedures, did not "control policy directives," and had "minimal influence over [defendant's] day-to-day functioning"). To be sure, GEO, like a great many private corporations, does business under contract with the government. But this is not by itself enough to subject it to constitutional liability, see Rendell-Baker v. Kohn, 457 U.S. 830, 840-41, 102 S.Ct. 2764, 73 L.Ed.2d 418 (1982), let alone to create such liability for its individual private employees.
Holly argues, however, that liability exists here by virtue of the fact that GEO's particular business involves prisons and is thus a "public function." The "`public function' theory" recognizes the existence of "narrow circumstances" where state action arises from the "`exercise by a private entity of powers traditionally exclusively reserved to the State.'" Andrews v. Fed. Home Loan Bank of Atlanta, 998 F.2d 214, 218 (4th Cir. 1993) (quoting Jackson v. Metro. Edison Co., 419 U.S. 345, 352, 95 S.Ct. 449, 42 L.Ed.2d 477 (1974)). Those circumstances are not present here, however. The Supreme Court's analysis in Richardson v. McKnight, 521 U.S. 399, 117 S.Ct. 2100, 138 L.Ed.2d 540 (1997), precludes argument that the operation of a prison is a traditionally exclusive state function.
Richardson involved a § 1983 suit against prison guards employed by a private prison management corporation under contract with the state of Tennessee. Id. at 401-02, 117 S.Ct. 2100. The Court did not decide whether § 1983 permitted such a suit and addressed only the narrow question whether the guards should receive qualified immunity. See id. at 413, 117 S.Ct. 2100. The Court decided that they should not, in part because "correctional functions have never been exclusively public." Id. at 405, 117 S.Ct. 2100. The Court's historical analysis revealed that the private operation of jails and prisons existed in the United States in the eighteenth and nineteenth centuries, id. at 405, 117 S.Ct. 2100, and in England the practice dated back to the Middle Ages, id. at 407, 117 S.Ct. 2100. Therefore, under the test articulated in Jackson, the operation of prisons is not a "public function."
Holly attempts to sidestep the analysis in Richardson in two ways. First, he urges that the "function" to which we should look is not the administration of a prison, but rather the power to keep prisoners under lock and key. This argument misapprehends the proper nature of our inquiry. In determining the presence of state action, we are not to conduct a far-flung investigation into all of a defendant's possible activities, but rather must focus on "`the specific conduct of which the plaintiff complains.'" Am. Mfrs. Mut. Ins. Co. v. Sullivan, 526 U.S. 40, 51, 119 S.Ct. 977, 143 L.Ed.2d 130 (1999) (quoting Blum v. Yaretsky, 457 U.S. 991, 1004, 102 S.Ct. 2777, 73 L.Ed.2d 534 (1982)); see also Blum, 457 U.S. at 1003, 102 S.Ct. 2777 ("Faithful adherence to the `state action' requirement . . . requires careful attention to the gravamen of the plaintiff's complaint."). The inadequate medical care alleged in this case unquestionably arises out of defendants' operation of the prison, not the fact of Holly's incarceration. If Holly's complaint were in fact that he is being held unlawfully, he would have filed a statutory habeas action under 28 U.S.C. § 2255, not a Bivens claim.
Second, Holly reads West v. Atkins, 487 U.S. 42, 108 S.Ct. 2250, 101 L.Ed.2d 40 (1988), to suggest that the provision of medical care to an inmate is always a public function, regardless of what entity operates the correctional facility where he is housed. West cannot bear the weight that Holly would place upon it. There, the Supreme Court held that "a physician employed by North Carolina to provide medical services to state prison inmates acted under color of state law for purposes of § 1983." 487 U.S. at 54, 108 S.Ct. 2250; see also Conner v. Donnelly, 42 F.3d 220 (4th Cir. 1994) (state action to treat prisoner housed in state-run facility even where physician had no contract with state and provided care outside prison). Even assuming we would consider liability under Bivens coextensive with § 1983 under such circumstances, West instructs that "the dispositive issue" in the state action inquiry "concerns the relationship among the State, the physician, and the prisoner." West, 487 U.S. at 56, 108 S.Ct. 2250. That relationship is very different in this case, where the correctional facility is privately run, than in West and Conner, where the state itself was directly responsible for managing the prison. See West, 487 U.S. at 43-44, 108 S.Ct. 2250 (noting that the "state-prison hospital" was "operated by the State"); Conner, 42 F.3d at 221 (noting that the question presented concerned "an inmate of a state-run prison"); see also West 487 U.S. at 56 n. 15, 108 S.Ct. 2250 (recognizing that physician's "delivery of medical care was not unaffected by the fact that the State controlled the circumstances and sources of a prisoner's medical treatment").
Richardson demonstrates that this distinction between public and private correctional facilities is critical. The state's responsibilities are necessarily greater when it undertakes direct authority over prisoners' day-to-day care. We cannot conclude that provision of medical care in a private prison is somehow a "public function" while maintaining fidelity to Richardson that the prison's general operation is not.
In addition to settling the issue whether defendants are liable for performing a public function, Richardson undercuts Holly's argument in another respect as well. Since qualified immunity under Bivens and § 1983 are identical, see Butz, 438 U.S. at 504, 98 S.Ct. 2894, were we to find defendants potentially liable, the holding of Richardson would preclude us from granting them a qualified immunity defense. This would put prisoners in private facilities in a more favorable position than their counterparts in federally operated prisons: they would be eligible to recover damages even where the unconstitutionality of the prison officials' conduct had not been clearly established by prior judicial decisions. In the absence of statutory authority, we are reluctant to create an anomaly whereby private defendants face greater constitutional liability than public officials. See Malesko, 534 U.S. at 71-72, 122 S.Ct. 515 (finding "no reason . . . to consider extending Bivens" in part because " no federal prisoners enjoy [plaintiff's] contemplated remedy").
In sum, even if there did exist circumstances where a private party could be subject to liability under Bivens, this case does not present them. Both judicial restraint in the creation of extra-statutory remedies and the doctrine of state action compel us to refrain from extending Bivens liability to the individual employees of a private corporation operating a correctional facility.
Our concurring colleague would create out of whole cloth constitutional liability for private employees of a private corporation. The concurrence sees no distinction between § 1983 and Bivens remedies, and intermingles the caselaw of both to argue that defendants are "government actors." But there is a world of distinction between § 1983 and Bivens remedies. Congress has authorized the former, and Congress has in no way authorized the latter. Contrary to the concurrence's intimation, therefore, it is impossible to create a "circuit split" with § 1983 cases which do not even address the issue before this court.
Ironically, the concurrence contends the majority itself intermingles § 1983 and Bivens, but the concurrence misses the critical point: the majority has used § 1983 to buttress the importance of Congress's role, not to undermine it. The majority has never relied on § 1983 or any other express cause of action created by Congress to extend liability for damages under a cause of action that Congress has not seen fit to create. And it certainly has not done so in the face of repeated Supreme Court admonitions that Bivens remedies are to be carefully circumscribed. If Congress wishes to adopt the concurrence's position and extend the Bivens cause of action to private employees of private entities, it has only to do so. The fact that it has not done so, however, gives the concurrence no license to use § 1983 or any other provision to create causes of action on its own.
This intermingling of § 1983 and Bivens cases leads the concurrence to misread Malesko. The concurrence acknowledges that Malesko "did not reach the precise question of whether a Bivens action would lie against individual `private correctional providers,'" but then asserts that the decision nevertheless "expressly recognize [d]" that "such individuals are government actors . . . subject to suit." Concurring Op. at 299. It relies for this contradictory assertion on a half-sentence in a footnote suggesting that private correctional employees could be liable under § 1983. See id. (citing 534 U.S. at 72 n. 5, 122 S.Ct. 515). This half-sentence did not purport to hold that private employees were liable under § 1983, because that issue was not before the Court. And it certainly cannot be read to imply the much more aggressive holding that a judicially implied Bivens remedy would extend to the private employees of a private company. See supra note 2 (noting that the Court did not intend to so hold).
Finally, we cannot agree with the concurrence's assertion that "defendants perform a public function." Concurring Op. at 299. As we have discussed, Richardson rather than West is controlling when a privately run correctional facility is at issue. With all due respect to the concurrence, we lack the power to displace Richardson's determination of historical fact that "correctional functions have never been exclusively public." 521 U.S. at 405, 117 S.Ct. 2100. And we likewise lack the power to imply private causes of action for damages against private individuals in private facilities without congressional input and to the utter exclusion of the legislative function.
In addition to the issue of government action, a second independent factor counseling hesitation is that Holly possesses alternative — and arguably superior — causes of action against defendants under the state law of negligence. The dangers of overreaching in the creation of judicial remedies are particularly acute where such remedies are unnecessary. We decline to invite such dangers by recognizing Bivens liability here.
The Supreme Court has extended Bivens in only two circumstances. Malesko, 534 U.S. at 70, 122 S.Ct. 515. One has been "to provide a cause of action for a plaintiff who lacked any alternative remedy for harms caused by an individual officer's unconstitutional conduct." Id. This was the rationale both for Bivens itself and for Davis, where the plaintiffs' injuries would have gone entirely unredressed without an implied constitutional remedy. See id. at 74, 122 S.Ct. 515 (noting that each case involved "a plaintiff in search of a remedy"); Davis, 442 U.S. at 245, 99 S.Ct. 2264 (recognizing damages remedy where "[f]or Davis, as for Bivens, it is damages or nothing") (internal quotation marks omitted); Bivens, 403 U.S. at 394, 91 S.Ct. 1999 (recognizing damages remedy where state law "may be inconsistent or even hostile" to interests protected by the Fourth Amendment).
The other rationale for extension of Bivens has been "to provide an otherwise nonexistent cause of action against individual officers alleged to have acted unconstitutionally." Malesko, 534 U.S. at 70, 122 S.Ct. 515. This was the rationale behind Carlson, where relief was available only against the government itself, and not against the individual tortfeasors. See id. at 74, 122 S.Ct. 515 (noting that in Carlson "a cause of action against an individual officer" was "otherwise lacking"); Carlson, 446 U.S. at 21, 100 S.Ct. 1468 (discussing the fact that the plaintiff's Federal Tort Claims Act remedy did not allow for recovery against the individual officers responsible for violating his rights).
But "where [these two] circumstances are not present," the Court has "consistently rejected invitations" to enlarge the scope of the judicially created Bivens remedy. Malesko, 534 U.S. at 70, 122 S.Ct. 515. Neither circumstance is present here. Holly does not contest that state law provides him with an "alternative remedy" and that this alternative remedy includes a "cause of action against individual officers." North Carolina law in fact supplies Holly with multiple potential claims against the individual defendants. See, e.g., Summey v. Barker, 154 N.C.App. 448, 573 S.E.2d 534, 536 (N.C.Ct.App. 2002) (suit by hemophiliac detainee for, inter alia, negligence, medical negligence, and cruel and unusual punishment, based upon authorities' and doctors' failure to properly treat his nosebleed). These claims include the possibility of recovering punitive damages if he can prove by clear and convincing evidence that defendants' conduct was willful, wanton, or malicious. N.C. Gen. Stat. § 1D-15 (2003). In addition to his causes of action against the individual defendants, he can also sue GEO under a respondeat superior theory. See Johnson v. Lamb, 273 N.C. 701, 161 S.E.2d 131, 137 (1968).
Holly's reliance upon Carlson as supporting his Bivens action here is therefore unpersuasive. The key feature of Carlson was that, unlike this case, it presented a situation where the plaintiff sought "a cause of action against an individual officer" that was "otherwise lacking." Malesko, 534 U.S. at 74, 122 S.Ct. 515. As opposed to the various state law causes of action against defendants and against GEO that are available to Holly, the Carlson plaintiff's "only alternative" to a Bivens claim "was a Federal Tort Claims Act (FTCA) claim against the United States." Id. at 68, 122 S.Ct. 515; see 446 U.S. at 16-17, 100 S.Ct. 1468. This placed Carlson squarely within the circumstance where extending Bivens "provide[d] an otherwise nonexistent cause of action against individual officers." Malesko, 534 U.S. at 70, 122 S.Ct. 515 (emphasis removed); see also id. at 71, 122 S.Ct. 515 ("core premise" of Bivens is the "deterrence of individual officers"); Meyer, 510 U.S. at 485, 114 S.Ct. 996 ("[T]he purpose of Bivens is to deter the officer."). Indeed, it is precisely because an inmate in a federally run facility has a cause of action against the government itself under the FTCA that he lacks the types of state law claims against individual defendants that are available to Holly. See 28 U.S.C. § 1346(b)(1) (2000) (creating a cause of action against the United States for the negligence of its employees); id. § 2679(b)(1) (precluding any other cause of action against the federal employee under such circumstances).
We agree therefore with the only other circuit to address this issue that an inmate in a privately run federal correctional facility does not require a Bivens cause of action where state law provides him with an effective remedy. See Peoples v. CCA Detention Ctrs., 422 F.3d 1090, 1103 (10th Cir. 2005). Holly already enjoys claims that an inmate in a government-run facility would not have. In requesting that we also grant him a Bivens claim — indeed, that we grant him a superior one in which qualified immunity is unavailable — Holly seeks much more than is necessary to remedy his alleged injuries. This is not a circumstance under which the extension of a judicially implied remedy is appropriate.
The judgment of the district court is therefore reversed and the case is remanded with directions that it be dismissed.
The majority's holding that private correctional employees are not governmental actors ignores or misreads controlling Supreme Court case law. Those cases, as well as numerous cases from other federal courts, establish that individual private correctional providers are government actors subject to liability as such. Accordingly, I cannot join the majority opinion. However, because Ricky Holly possesses an alternative remedy for his alleged injuries, no action under Bivens v. Six Unknown Named Agents of the Federal Bureau of Narcotics, 403 U.S. 388, 91 S.Ct. 1999, 29 L.Ed.2d 619 (1971), lies in this case. For that reason alone, I concur in the judgment.
The strictures of the Constitution generally apply only to public action. However, when private individuals take actions attributable to the government, they are liable, as government actors, for damages resulting from their constitutional violations. See, e.g., Evans v. Newton, 382 U.S. 296, 299, 86 S.Ct. 486, 15 L.Ed.2d 373 (1966) ("Conduct that is formally `private' may become so entwined with governmental policies or so impregnated with a governmental character as to become subject to the constitutional limitations placed upon state action."). In this case, the government has delegated its authority to the privately employed defendants, empowering them to incarcerate, to confine, to discipline, to feed, and to provide medical and other care to inmates who are imprisoned by order of the federal government. The defendants are acting as agents of the government; their actions are thus clearly attributable to the federal government, and a prisoner must be able to seek redress from the defendants if they cause him constitutional injury. Therefore, if Holly had no alternative remedy for the alleged deprivation of his constitutional rights, it seems to me that he could certainly bring a Bivens action against these defendants.
As the majority itself recognizes — nothwithstanding its response to this concurrence, ante at 294-95 n. 4, which I address post at 301-03 — the determination of whether the defendants are governmental actors for Bivens purposes rests on the "state action" principles developed by the Supreme Court in cases involving liability of private persons under 42 U.S.C. § 1983 (2000). See ante at 290-91, 292-93 (citing and relying on numerous § 1983 state action cases). But what the majority fails to recognize is that under this case law, the defendants here — individual private prison guards — are indeed governmental actors. In these cases, the Supreme Court has held that, in general, when a private person acts under authority given to him by the government, or is fulfilling an obligation or responsibility that is traditionally a public function, his conduct will be imputed to the government and thus considered government action for purposes of imposing liability. See, e.g., Edmonson v. Leesville Concrete Co., Inc., 500 U.S. 614, 620, 111 S.Ct. 2077, 114 L.Ed.2d 660 (1991); Rendell-Baker v. Kohn, 457 U.S. 830, 842, 102 S.Ct. 2764, 73 L.Ed.2d 418 (1982); Lugar v. Edmondson Oil Co., Inc., 457 U.S. 922, 937, 102 S.Ct. 2744, 73 L.Ed.2d 482 (1982) (all cited by the majority). Moreover, applying these principles, the Court has concluded that private correctional employees are governmental actors.
First, almost thirty years ago, the Court recognized an obligation on the part of the government's correctional employees "to provide medical care for those whom it is punishing by incarceration." Estelle v. Gamble, 429 U.S. 97, 103, 97 S.Ct. 285, 50 L.Ed.2d 251 (1976); see also Carlson v. Green, 446 U.S. 14, 18-19, 100 S.Ct. 1468, 64 L.Ed.2d 15 (1980) (holding that the federal government has the same obligation). The Estelle Court explained that necessity compelled this holding: "[a]n inmate must rely on prison authorities to treat his medical needs; if the authorities fail to do so, those needs will not be met." Id. Accordingly, the Court held that an inmate could bring a § 1983 action for Eighth Amendment violations against various state correctional employees, including a prison physician, for failing to provide adequate medical care. Then, almost twenty years ago, in West v. Atkins, 487 U.S. 42, 54, 108 S.Ct. 2250, 101 L.Ed.2d 40 (1988), the Court extended that obligation to private correctional employees under contract with the state, holding that a prisoner could bring a § 1983 action against a private doctor rendering medical care to inmates because the acts of the doctor were attributable to the government.
Tellingly, in West a majority of the Fourth Circuit had refused to so hold, reasoning that the private doctor was not a governmental actor. West v. Atkins, 815 F.2d 993, 994 (4th Cir. 1987) (en banc). After noting that every other circuit to consider the question had implicitly or explicitly rejected the Fourth Circuit's view, the Supreme Court unanimously reversed. West, 487 U.S. at 47-48 n. 7, 54, 108 S.Ct. 2250. Moreover, the Court held that the approach of the en banc Fourth Circuit majority misread and misunderstood prior Supreme Court precedent in its failure to recognize that a private doctor, "authorized and obliged" by the government to treat prison inmates, is a government actor subject to liability as such. Id. at 55, 108 S.Ct. 2250.
The mere fact that the government did not directly employ the private doctor in West did not preclude him from being a governmental actor. The Supreme Court explained that government action derives from the individual's "function within the state system, not the precise terms of his employment." Id. Under this standard, the Court found no relevant distinction between the publicly employed doctors in Estelle and the private doctor in West. Id. As with its public employees, the government was liable for the private doctor's Eighth Amendment violations because the government had invested the doctor with power "to provide medical services to state prison inmates." Id. at 54, 108 S.Ct. 2250. If the doctor misused that power, "the resultant deprivation was caused, in the sense relevant for state-action inquiry, by the State's exercise of its right to punish [the inmate] by incarceration and to deny him a venue independent of the State to obtain needed medical care." Id. at 55, 108 S.Ct. 2250. Because the private doctor was "fully vested with state authority to fulfill essential aspects" of his duties, the doctor must "be considered to be a state actor." Id. at 57, 108 S.Ct. 2250.
So it is in this case. The defendants have been "fully vested" with the governmental authority to "fulfill essential aspects" of their duties, e.g., confinement and care of prisoners incarcerated by the government. The defendants perform a public function delegated to them by the federal government, and they assume the necessary obligations inherent in that function. The assumption and performance of these critical governmental duties and the role these individuals play in the penal system surely put more at stake here than the mere "contractual relationship" dismissed by the majority. See ante at 290-91. Rather, just as in West, any alleged constitutional harm suffered by Ricky Holly was caused "in the sense relevant for [government]-action inquiry" by the government's power to punish and incarcerate him, depriving him of alternative means of care independent of the government. West, 487 U.S. at 55, 108 S.Ct. 2250. And, as in West, the fact that the defendants are not direct employees of the government does not in any way alter their fundamental obligations, for it is their "function within the [federal] system, not the precise terms of [their] employment" that makes their conduct governmental action. Id. Accordingly, again as in West, the defendants are "possessed of [governmental] authority" and so must be considered governmental actors. Id. at 56 n. 15, 108 S.Ct. 2250.
The Supreme Court has never retreated from its holding in West. Indeed, only four years ago, in its most recent Bivens case, the Court specifically acknowledged that individual private correctional providers (not just private prison doctors) are government actors. See Corr. Servs. Corp. v. Malesko, 534 U.S. 61, 122 S.Ct. 515, 151 L.Ed.2d 456 (2001). In Malesko, the Court did not reach the precise question of whether a Bivens action would lie against individual "private correctional providers," but it did expressly recognize that such individuals are government actors whose conduct is attributable to the government; thus, they would, in the appropriate case, be subject to suit on that basis. Id. at 72 n. 5, 122 S.Ct. 515 (noting that prisoners "enjoy a right of action against private correctional providers under 42 U.S.C. § 1983").
The majority ignores Malesko's express acknowledgment that individual private correctional providers are government actors and unconvincingly dismisses the strikingly similar West precedent as "very different." Ante at 294. Instead of following Malesko and West, binding Supreme Court precedent as to the liability of private correctional employees as government actors, the majority relies on a case involving a question not presented here — whether private persons sued as government actors under § 1983 enjoy qualified immunity for their actions. See Richardson v. McKnight, 521 U.S. 399, 117 S.Ct. 2100, 138 L.Ed.2d 540 (1997).
This assertedly critical difference is that while in West the government had a contract with the private doctor who cared for the prisoners, here the government contracts with a private entity which in turn employs those who care for prisoners. Ante at 294. The majority thus suggests that by adding an additional layer, the government can contract away its constitutional duties. West itself expressly rejects this disturbing contention, instructing that "[c]ontracting out" care " does not relieve" the government of its "constitutional duty" to provide adequate care or "deprive inmates of the means to vindicate their Eighth Amendment rights." West, 487 U.S. at 55-56, 108 S.Ct. 2250 (emphasis added).
The majority posits that Richardson (which, of course, predates Malesko) "precludes argument that the operation of a prison is a traditionally exclusive state function" and that therefore, private correctional providers are not government actors for purposes of imposing liability. Ante at 293. But Richardson does nothing of the sort; the case deals only with a private person's immunity from liability. Richardson, 521 U.S. at 401, 413, 117 S.Ct. 2100. That question, as the Supreme Court has acknowledged, does not control the separate question of whether a private party can be liable as a governmental actor. See Wyatt v. Cole, 504 U.S. 158, 168-69, 112 S.Ct. 1827, 118 L.Ed.2d 504 (1992) (holding certain "private defendants" not entitled to the qualified immunity afforded government officials but remanding for determination as to whether they are liable as government actors); see also, United States v. Thomas, 240 F.3d 445, 448-49 (5th Cir. 2001) (holding employee of private detention center a "public official" for purposes of the Federal Bribery Statute and distinguishing Richardson on the ground that "policy considerations supporting private corrections officers' not being entitled to qualified immunity are quite different from those concerning whether they are `public officials' for purposes of the federal bribery statute").
The fact is that, at least in this country, incarceration of those charged with committing crimes is, and always has been, the province and prerogative of the government. That historically immunity has not been afforded those performing some correctional duties demonstrates only that the government has delegated some of its correctional functions to private actors. Richardson, 521 U.S. at 405-07, 117 S.Ct. 2100. These correctional functions have not been "exclusively public," id., only in the sense that private individuals have long been empowered by the government to fulfill the tasks involved in the fundamentally governmental function of incarceration of criminals. But this government delegation of some duties to private persons or entities does not change the public character of the underlying function performed by "private correctional providers," as the Court recognized in Malesko, 534 U.S. at 72 n. 5, 122 S.Ct. 515.
Indeed, in Richardson itself, the Court recognized that its historical discussion did not apply to questions of governmental action. After concluding that the defendants lacked qualified immunity, the Richardson Court remanded for a determination of whether the defendants were, in fact, liable as governmental actors for their operation, confinement, and care of inmates. Richardson, 521 U.S. at 413, 117 S.Ct. 2100. If the Court's historical analysis of "public function" for immunity purposes were meant to control the "public function" determination for liability purposes — as the majority holds today — the Court would not have needed to remand the case at all.
Certainly, incarcerating and caring for those confined by the government for criminal conduct is just as much a public function as maintaining a park, Evans v. Newton, 382 U.S. 296, 86 S.Ct. 486, 15 L.Ed.2d 373 (1966), running an election, Terry v. Adams, 345 U.S. 461, 73 S.Ct. 809, 97 L.Ed. 1152 (1953), or operating a company town, Marsh v. Alabama, 326 U.S. 501, 66 S.Ct. 276, 90 L.Ed. 265 (1946). All of these functions "traditionally serve the community" no matter who performs them, and surely the "predominant character and purpose" of a prison is as governmental as these other functions. Evans, 382 U.S. at 302, 86 S.Ct. 486. Thus, those performing the fundamental governmental function at issue here, like those in Evans, Terry, and Marsh, are governmental actors whether publicly or privately employed and are subject to liability as such.
It is worth noting that, despite the majority's "reluctan[ce] to create an anomaly whereby private defendants face greater constitutional liability than public officials," see ante at 294, the majority creates its own anomaly. That is, the majority holds that even though an individual guard employed by a private correctional facility under contract with the state government is a governmental actor liable for his constitutional wrongs, an individual guard employed by a private correctional facility under contract with the federal government somehow is not. Thus the majority places inmates in private correctional institutions under contract with a state government "in a more favorable position than their counterparts" in private correctional facilities under contract with the federal government, id.; the former could bring a § 1983 action in the proper case to recover damages for constitutional wrongs, while the latter could never bring the analogous federal action.
Numerous courts have so held, both before and after Richardson. See, e.g., Rosborough v. Mgmt. Training Corp., 350 F.3d 459, 461 (5th Cir. 2003) (holding that employees of a private correctional facility perform a "fundamentally government function" — "confinement of wrongdoers" — and so are government actors for § 1983 purposes); Street v. Corr. Corp. of Am., 102 F.3d 810, 814 (6th Cir. 1996) (same); Ancata v. Prison Health Servs., Inc., 769 F.2d 700, 703 (11th Cir. 1985) (explaining that employees of private health corporation are state actors when performing the public function of providing required medical services to inmates); Sarro v. Cornell Corr., Inc., 248 F.Supp.2d 52, 60-62 (D.R.I. 2003) (holding that employees of a private operator of a prison facility perform a government function traditionally reserved to the government and are government actors for Bivens purposes); Palm v. Marr, 174 F.Supp.2d 484, 487-88 (N.D.Tex. 2001) (holding that employees of a private prison facility exercise powers that are traditionally the exclusive prerogative of the state and so are governmental actors for § 1983 purposes); Giron v. Corr. Corp. of Am., 14 F.Supp.2d 1245, 1249 (D.N.M. 1998) (holding that "[t]he function of incarcerating people, whether done publicly or privately, is the exclusive prerogative of the state"); Kesler v. King, 29 F.Supp.2d 356, 370-71 (S.D.Tex. 1998) (concluding that "the incarceration of inmates . . . falls within the exclusive responsibility of the state" and so holding that the warden of a private correctional facility is a governmental actor subject to § 1983 liability); Nelson v. Prison Health Servs., Inc., 991 F.Supp. 1452, 1463 (M.D.Fla. 1997) (explaining that nurses employed by private company providing medical services for a jail are considered state actors for § 1983 liability).
In holding to the contrary, the majority disregards all of this authority and creates a circuit split. Indeed, like the en banc majority in West, the majority's view stands alone among the federal circuits addressing this point. Cf. Rosborough, 350 F.3d at 461; Street, 102 F.3d at 814; Ancata, 769 F.2d at 703. Even more disturbingly, the majority, again like the en banc majority in West, misreads and misunderstands Supreme Court precedent. Pursuant to that precedent, the defendants here were clearly exercising authority fairly attributable to the government and so are government actors for liability purposes.
Notably, the only other circuit to consider the precise question presented here — whether an inmate can bring a Bivens action against private correctional employees — does not share the majority's view on this point. See Peoples v. CCA Detention Ctrs., 422 F.3d 1090 (10th Cir. 2005). The Tenth Circuit, unlike the majority here, did not hold that the Bivens action failed because the private defendants were not governmental actors. Rather, the Tenth Circuit (adopting the same approach I do here) refused to imply a Bivens action solely because it concluded that in the case before it, an alternative remedy precluded the Bivens action. Id. at 1108.
The only substantive response the majority offers to this concurrence is an argument that because § 1983 actions are statutory and Bivens actions judicially created, I err in relying on § 1983 precedent in this Bivens case. See ante at 294-95 n. 4. This argument utterly fails.
The majority claims that this improper "intermingling" of § 1983 and Bivens cases leads me to make allegedly "contradictory" assertions about Malesko and to "create causes of action." See ante at 294-95 n. 4. Actually it is the majority that misunderstands Malesko. There is nothing "contradictory" in noting that Malesko did not reach the precise question as to whether a Bivens action would lie against individual private correctional providers but did recognize that they are government actors, because, as the case at hand demonstrates, it may be that no Bivens action lies even when defendants are government actors. Like the Malesko Court, I "create" no "cause of action" at all, but rather conclude that no Bivens action lies here. Government action is a necessary but not sufficient pre-requisite to a Bivens action.
While different in origin, § 1983 and Bivens actions are identical in the respect critical here — both involve imposition of liability on governmental actors for constitutional violations. In view of the scarcity of cases discussing governmental actor status under federal law, § 1983 cases determining the liability of private persons as state actors provide valuable guidance when determining the liability of private persons as federal actors. See e.g., Brown v. Philip Morris Inc., 250 F.3d 789, 801 (3rd Cir. 2001) (explaining that a Bivens action is "the federal equivalent of the § 1983 cause of action," and holding in a Bivens case that "[i]n order to determine whether the conduct of a private party should be attributable to the federal government, courts apply the `state action' analysis set forth by the Supreme Court" in § 1983 cases). See also, Edmonson, 500 U.S. at 621-22, 111 S.Ct. 2077 (explaining, in a case involving a federal actor, that "certain principles of general application" should be applied to determine "whether a particular course of conduct is governmental in character"). In fact, despite its lengthy footnote criticizing me, earlier in its opinion the majority itself cites § 1983 "state action" cases as providing the principles to be followed in determining whether the private defendants here are government actors. See ante at 290-91, 292-93.
Indeed, the majority builds its entire holding that the defendants are not government actors on a § 1983 case. In doing so, however, the majority eschews the relevant § 1983 precedent it initially cites, and instead commits the very sin it accuses me of — "intermingling" cases from different "worlds." Id. at 294-95 n. 4. For the majority can conclude that the defendants are not government actors only by holding that Richardson, a § 1983 qualified immunity case, "is controlling" on the very different question presented here — whether a private person is liable as a governmental actor. Id. That the majority holds a defendant's entitlement to qualified immunity "controls" his liability — when the Supreme Court expressly refused to do so both in Wyatt and Richardson itself — speaks volumes.
Thus, in its attempt to counter my concurrence, the majority is forced into the untenable position of disavowing clearly relevant precedent on which it itself otherwise relies, and then finding "controlling" an inapposite case dealing with a wholly different issue. The majority must steadfastly rely on an immunity case and resolutely refuse to acknowledge the applicability of the cases dealing with the liability of government actors because to do otherwise makes the conclusion I have reached — that the defendants are indeed government actors — inescapable.
Although Supreme Court precedent thus dictates that the defendants here are governmental actors, the Court has indicated that a court should not imply a Bivens cause of action against government actors if the plaintiff possesses an alternative remedy. Because Holly possesses an alternative remedy against the defendants under state tort law, I must concur in the judgment.
This rationale is more than sufficient to address the majority's policy concerns with extending Bivens actions. See ante at 289-90. Thus, no policy concern requires us to become the only circuit court to hold that individual private correctional providers are not governmental actors — a holding that may have untoward and far-reaching consequences beyond the Bivens context.
In Peoples v. CCA Detention Ctrs., 422 F.3d 1090, 1109-1111 (10th Cir. 2005) (Ebel, J., concurring and dissenting), Judge Ebel forcefully set forth a contrary view — that to be sufficient to preclude a Bivens action, the alternative remedy "must be a constitutional cause of action," and so state tort law is inadequate. Some of the language and reasoning in Bivens, 403 U.S. at 391-92, 91 S.Ct. 1999, and Carlson, 446 U.S. at 24, 100 S.Ct. 1468, support this position.
In the years since Bivens and Carlson, however, the Court has retreated from this view, instead limiting a Bivens cause of action to cases in which no alternative remedy — constitutional or otherwise — exists. See Malesko, 534 U.S. at 69, 122 S.Ct. 515 (noting that the Court has "rejected the claim that a Bivens remedy should be implied simply for want of any other means for challenging a constitutional deprivation in federal court"); Schweiker v. Chilicky, 487 U.S. 412, 421, 108 S.Ct. 2460, 101 L.Ed.2d 370 (1988) (noting that "[t]he absence of statutory relief for a constitutional violation . . . does not by any means necessarily imply that courts should award money damages against the officers responsible for the violation"); Bush v. Lucas, 462 U.S. 367, 386-88, 103 S.Ct. 2404, 76 L.Ed.2d 648 (1983) (refusing to find a Bivens cause of action because meaningful, albeit incomplete, relief was available through the administrative system created by Congress).
Because Holly has an alternative remedy under North Carolina tort law, this more recent Supreme Court precedent requires that we reverse the district court's order refusing to dismiss this action.
I therefore concur — but only in the judgment.