Dodds v. Richardson

FILED United States Court of Appeals Tenth Circuit August 6, 2010 PUBLISH Elisabeth A. Shumaker Clerk of Court UNITED STATES COURT OF APPEALS TENTH CIRCUIT THOMAS CARL DODDS, JR., Plaintiff-Appellee, v. No. 09-6157 RANDY RICHARDSON, Sheriff, individually, Defendant-Appellant, LOGAN COUNTY SHERIFF’S DEPARTMENT; DAVID LANDMAN, Deputy, official capacity, JOHN DOE, Deputies 1–6 (arresting deputies and jailers), individually and in their official capacities; LOGAN COUNTY SHERIFF, sued as “The Sheriff of Logan County in his Official Capacity,” Defendants. APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE WESTERN DISTRICT OF OKLAHOMA (D.C. No. 5:08-CV-00333-R) James Richard McClure, Muskogee, Oklahoma, for Plaintiff-Appellee. Eric Devalson Cotton (Christopher James Collins with him on the briefs), Collins, Zorn & Wagner, P.C., Oklahoma City, Oklahoma, for Defendant-Appellant. Before TYMKOVICH, SEYMOUR, and BALDOCK, Circuit Judges. BALDOCK, Circuit Judge. Plaintiff Thomas Carl Dodds, Jr. brought this 42 U.S.C. § 1983 suit, alleging Defendant former Logan County, Oklahoma Sheriff Randy Richardson violated his Fourteenth Amendment due process rights by depriving him of his protected liberty interest in posting bail. The district court denied Defendant’s claim to qualified immunity in the context of summary judgment, and Defendant appealed. Exercising jurisdiction pursuant to 28 U.S.C. § 1291, we affirm. I. Logan County, Oklahoma sheriff’s deputies arrested Plaintiff pursuant to a felony arrest warrant and placed him in the Logan County Jail on Friday, April 6, 2007. An Oklahoma state district court judge set bond for Plaintiff in the arrest warrant in the amount of $5,000. Plaintiff alleges Logan County Jail employees told two individuals who inquired about posting bond on his behalf that he could not post the preset bail until after he was arraigned by a judge. Plaintiff was not arraigned until Monday, April 9. At that time, another state district judge reset bail at $10,000. Plaintiff did not post bail. Later that week, the district court nevertheless released Plaintiff on a personal recognizance bond. The charges against Plaintiff underlying the arrest warrant were eventually dismissed in September 2007. Subsequently, Plaintiff filed suit under 42 U.S.C. § 1983 against, among others, Defendant in his individual capacity. Plaintiff asserted in his complaint he “had a liberty interest in 2 the pre-set bond under the Fourteenth Amendment,” but Defendant denied him “such, without explanation, in violation of the [United States] Constitution.” Defendant filed a motion for summary judgment claiming he was entitled to qualified immunity because the policy of the Logan County court clerk or district judges prevented individuals charged with a felony from posting bond after hours. According to Defendant, this policy rather than any action taken by him personally caused the alleged deprivation of Plaintiff’s liberty. Dodds v. Logan County Sheriff’s Dep’t., No. 08-CV-00333-R, Order at *2 (W.D. Okla. Aug. 3, 2009) (Doc. #75). In an affidavit submitted to the district court and incorporated into Defendant’s motion, the Logan County court clerk stated that for “at least the past eighteen years” “Logan County has [had] a local rule preventing individuals charged with a felony from posting bond until they have gone before a judge and been arraigned.” Aplt’s App. at 138. The clerk also confirmed that “[i]t is the policy of the Court Clerk’s office not to permit the Sheriff’s office to accept bonds after hours on felony warrants.” Id. Evidently, however, no one submitted to the district court or this Court a copy of these local policies or stated definitively who promulgated them. See Dodds, Order at *2 (Aug. 3, 2009). In response to Defendant’s summary judgment motion, Plaintiff did not allege Defendant was one of the jail employees who told him or the individuals who inquired about posting bail on his behalf that he may not post the preset bail until he had been arraigned by a judge. Instead, Plaintiff responded that an Oklahoma sheriff 3 is responsible for his county jail and has a duty to allow an arrestee such as Plaintiff to post bond. Id. (citing 57 Okla. Stat. § 47 (“The sheriff . . . shall have charge of the county jail of his county and of all persons by law confined therein, and such sheriff . . . is hereby required to conform, in all respects, to the rules and directions promulgated pursuant to [74 Okla. Stat. § 192] and of the district judge and communicated to him by the proper authority.”); 22 Okla. Stat. § 1101(A) (“Except as otherwise provided by law, bail . . . shall be admitted upon all arrests in criminal cases where the offense is not punishable by death and in such cases it may be taken by and of the persons . . . authorized by law to arrest, [and] to imprison offenders . . . .”); Okla. Att’y. Gen. No. 69-138 (1969) (“In criminal cases except cases punishable by death . . . a sheriff is required to accept bail, under the terms of [22 Okla. Stat. § 1101 (1961)], for those persons jailed at times other than the normal working hours of the Court, provided proper bail has been set as provided by law.”)). Plaintiff further asserted that the court clerk had no authority to create or maintain its policies at the jail or to dictate bail policy to the sheriff. Id. Therefore, according to Plaintiff, by acquiescing in the operation of the clerk’s non-binding policies at the jail, Defendant breached the duties imposed by Oklahoma law to accept bail and to maintain the jail himself, and deprived Plaintiff of his liberty interest in posting the preset bail, in violation of the Fourteenth Amendment. Id. The district court denied Defendant’s motion for summary judgment, concluding: By accepting or acquiescing in a policy set by the Logan County Court 4 Clerk or district court judges purportedly prohibiting individuals who have been arrested from posting bond until they have appeared before a judge and have been arraigned and or prohibiting the sheriff’s office from accepting bond, Defendant Richardson knew or should have known that Logan County deputies and jailers would violate the constitutional rights of arrestees like Plaintiff whose bail had been pre- set in his arrest warrant by refusing to allow them to post bail in the amount set or accept bail, because under Oklahoma law, a sheriff is required to accept bail which has already been set for persons jailed at times other than the normal working hours, and individuals have a liberty interest in being freed of detention once the amount of their bail is set. See Gaylor v. Does, 105 F.3d 572, 576 (10th Cir. 1997). It may reasonably be inferred that Defendant Richardson, who was the supervisor of the deputies and jailers for Logan County, exhibited deliberate indifference to the due process rights of arrestees whose bail had been pre-set to be free of detention by acquiescing in the Logan County policy and that his acquiescence caused or contributed in causing the deprivation of Plaintiff’s due process rights by another or others. Accordingly, Defendant Richardson has failed to show that he cannot be liable for participating or acquiescing in the deprivation of Plaintiff’s Fourteenth Amendment rights. See e.g., Serna v. Colorado Department of Corrections, 455 F.3d 1146, 1151–52 (10th Cir. 2006). Id. at *3–*4. The district court had previously concluded in another order that when Plaintiff was prevented from paying the bond set in his arrest warrant pursuant to these aforementioned policies, he “was unnecessarily detained without a legitimate goal (none is asserted) in violation of his due process rights.” Dodds v. Logan County Sheriff’s Dep’t., No. 08-CV-00333-R, Order at *4, (W.D. Okla. July 9, 2009) (Doc. #61) (denying the motion for summary judgment by Defendant Sheriff of Logan County in his official capacity). Defendant appeals, challenging the district court’s denial of qualified immunity. First, in his opening brief he maintains Plaintiff has failed to show that 5 he was personally involved in preventing Plaintiff from posting bail. Second, he asserts Plaintiff has not demonstrated that he acted with the state of mind (“knowingly or with ‘deliberate indifference’ that a constitutional violation would occur”) required to impose § 1983 supervisory liability upon him. Third, Defendant argues our decision in Gaylor v. Does, 105 F.3d 572 (10th Cir. 1997), does not clearly establish that his compliance with a policy that prevents an arrestee from posting preset bail subjects him to liability in his individual capacity. II. The Supreme Court has recognized a number of immunities from § 1983 suit and liability, including qualified immunity. Lawrence v. Reed, 406 F.3d 1224, 1229 (10th Cir. 2005). “The doctrine of qualified immunity protects government officials ‘from liability for civil damages insofar as their conduct does not violate clearly established statutory or constitutional rights of which a reasonable person would have known.’” Pearson v. Callahan, 129 S. Ct. 808, 815 (2009) (quoting Harlow v. Fitzgerald, 457 U.S. 800, 818 (1982)). Once a defendant asserts qualified immunity, the plaintiff bears the burden of satisfying a “‘strict two-part test.’” McBeth v. Himes, 598 F.3d 708, 716 (10th Cir. 2010) (quoting Bowling v. Rector, 584 F.3d 956, 964 (10th Cir. 2009)). “The plaintiff must establish ‘(1) that the defendant violated a constitutional or statutory right, and (2) that this right was clearly established at the time of the defendant’s conduct. . . .’” Id. (quoting Bowling, 584 F.3d at 964). 6 We possess “‘interlocutory jurisdiction over denials of qualified immunity at the summary judgment stage to the extent that they turn on an issue of law.’” Zia Trust Co. ex rel. Causey v. Montoya, 597 F.3d 1150, 1152 (10th Cir. 2010) (quoting Fogarty v. Gallegos, 523 F.3d 1147, 1153 (10th Cir. 2008)). Therefore, we usually only “examine the facts presented on summary judgment in the light most favorable to the plaintiff, to determine whether they amount to a violation of a clearly- established right.” Walker v. City of Orem, 451 F.3d 1139, 1155 (10th Cir. 2006); Lewis v. Tripp, 604 F.3d 1221, 1225 (10th Cir. 2010) (“At the summary judgment stage . . ., it is generally the district court’s exclusive job to determine which facts a jury could reasonably find from the evidence presented to it by the litigants. After doing so, the district court and we may then consider the ‘abstract’ legal questions whether those facts suffice to show a violation of law and whether that law was clearly established at the time of the alleged violation.” (internal citations omitted)). Generally, we may not “‘review a district court’s factual conclusions, such as the existence of a genuine issue of material fact for a jury to decide, or that a plaintiff’s evidence is sufficient to support a particular factual inference.’” Zia Trust Co., 523 F.3d at 1153 (quoting Fogarty, 523 F.3d at 1154). Within this limited jurisdiction, “‘[w]e review de novo the district court’s denial of a summary judgment motion asserting qualified immunity.’” McBeth, 598 F.3d at 715 (quoting Bowling, 584 F.3d at 963). III. 7 Although we are now at liberty to embark upon the two-part qualified immunity analysis in any order we choose, we begin in this case at the beginning: Has Defendant violated Plaintiff’s federally protected rights? See Pearson, 129 S. Ct. at 818 (deciding that courts may undertake the qualified immunity analysis in any order they choose). In doing so, we “consider[] whether the facts taken in the light most favorable to the plaintiff show that the defendant’s conduct violated a constitutional right” cognizable under § 1983. Poolaw v. Marcantel, 565 F.3d 721, 728 (10th Cir. 2009). A. Plaintiff’s claim falls into a category of claims which unfortunately have become so common that they have acquired their own term of art: “‘overdetention,’” i.e., when “the plaintiff has been imprisoned by the defendant for longer than legally authorized, whether because the plaintiff’s incarcerative sentence has expired or otherwise.” Holder v. Town of Newton, 638 F. Supp. 2d 150, 153 (D.N.H. 2009) (citing Davis v. Hall, 375 F.3d 703, 714 (8th Cir. 2004) and Barnes v. District of Columbia, 242 F.R.D. 113, 117 (D.D.C. 2007)). “[T]he right of an accused to freedom pending trial is inherent in the concept of a liberty interest protected by the due process clause of the Fourteenth Amendment.” Meechaicum v. Fountain, 696 F.2d 790, 791–92 (10th Cir. 1983). Consequently, the denial of bail must comport with the requirements of due process. See id. at 792 (“[S]tate statutes restricting bail must be rational, reasonable, and nondiscriminatory. . . . [B]ail may not be denied 8 ‘without the application of a reasonably clear legal standard and the statement of a rational basis for the denial.’” (quoting Atkins v. Michigan, 644 F.2d 543, 549 (6th Cir. 1981))). In Gaylor v. Does, 105 F.3d 572 (10th Cir. 1997), we further clarified the due process protections involving bail. We declared that an arrestee obtains a liberty interest in being freed of detention once his bail is set because the setting of bail accepts the security of the bond for the arrestee’s appearance at trial and “hence the state’s justification for detaining him fade[s].” Id. at 576. To avoid depriving an arrestee of due process, the government may only interfere with this protected liberty interest, for instance by refusing to accept lawfully set bail from the arrestee and detaining him until some later time, if its actions reasonably relate “to a legitimate goal.” Id. Otherwise, the detention of such an arrestee would constitute punishment prior to trial, in violation of due process. Id. Other courts have similarly concluded that arrestees whose bail have been set have a protected liberty interest in posting bail and being freed from detention and, thus, depriving such arrestees of that liberty interest may violate due process. See e.g., Campbell v. Johnson, 586 F.3d 835, 586 (11th Cir. 2009) (explaining that a county sheriff may be held liable for violating the due process rights of an arrestee if he acts with deliberate indifference in personally refusing to accept the arrestee’s court-set bail or if his actions were causally connected to his subordinates’ refusal of the arrestee’s bail); Doyle v. Elsea, 658 F.2d 512, 516–17 n.6 (7th Cir. 1981) (“For due process purposes, the constitutional liberty interest in release on bail arises after a magistrate 9 has determined that an accused may be released upon deposit of whatever sum of money will ensure the accused’s appearance for trial.”). The district court concluded in this case that when Plaintiff was prevented from paying the bond set in his arrest warrant pursuant to these abovementioned policies, he “was unnecessarily detained without a legitimate goal (none is asserted) in violation of his due process rights.” Dodds, Order at *4, (July 9, 2009). For the following reasons, we agree. Defendant has yet to proffer any reason, let alone a “legitimate goal,” for refusing to allow Plaintiff to post bail and detaining Plaintiff for three days, other than the assertion that the longstanding policies or customs at the jail, allegedly set by either the court clerk or the district judges, prohibited individuals charged with a felony from posting bond until they had been arraigned by a judge and from posting bond after hours. Gaylor teaches that Plaintiff’s liberty interest in being freed from pretrial detention once his bail had been set may not be denied just because an official says it has been his practice to do so for a long time and the practice of his predecessors for an even longer time. Gaylor, 572 F.2d at 576–77 (explaining that due process dictates that government officials may only interfere with an arrestee’s liberty interest in being freed from detention in a manner reasonably related “to a legitimate [governmental] goal, such as insuring his appearance for trial or protecting others from him”). We note Defendant’s counsel essentially conceded at oral argument the unconstitutionality of the policies that prevented Plaintiff from posting bail. When asked “[d]oesn’t Gaylor make it clear 10 that where bail has been set that it is a violation of due process not to grant it?” Defendant’s counsel responded “Yes, I believe Gaylor does.” We therefore conclude Plaintiff has set forth facts that, if proven to be true, state a violation of his constitutional rights. Of course, that conclusion alone does not merit denying Defendant qualified immunity. B. We must now determine whether this Defendant deprived Plaintiff of that right and whether he may be held liable for that deprivation. See McBeth, 598 F.3d at 716 (providing that to overcome an official’s assertion of qualified immunity and thereby subject her to suit and liability “[t]he plaintiff must establish ‘. . . that the defendant violated a constitutional or statutory right’” (quoting Bowling, 584 F.3d at 964)). To recover under § 1983, a plaintiff must establish “the violation of a right secured by the Constitution and laws of the United States. . . .” West v. Atkins, 487 U.S. 42, 48 (1988). 1 Because a plaintiff can neither recover under § 1983 from a government official nor overcome the official’s assertion of qualified immunity without demonstrating that official violated his constitutional or statutory rights, the legal analysis required to surmount these separate obstacles is often related, if not 1 A § 1983 plaintiff must also “show that the alleged deprivation was committed by a person acting under color of state law.” West, 487 U.S. at 48. Neither party in this case disputes that the facts presented establish Defendant acted under color of state law. They do, however, dispute whether Defendant himself, rather than only his subordinates, violated Plaintiff’s constitutional rights. 11 identical. As Justice Thomas explained: In conducting qualified immunity analysis . . ., courts do not merely ask whether, taking the plaintiff’s allegations as true, the plaintiff’s clearly established rights were violated. Rather, courts must consider as well whether each defendant’s alleged conduct violated the plaintiff’s clearly established rights. For instance, an allegation that Defendant A violated a plaintiff’s clearly established rights does nothing to overcome Defendant B’s assertion of qualified immunity, absent some allegation that Defendant B was responsible for Defendant A’s conduct. Hope v. Pelzer, 536 U.S. 730, 751 n.9 (2002) (Thomas, J., dissenting). In particular, as will become apparent from our discussion, determining whether a plaintiff has demonstrated a defendant-supervisor violated his constitutional rights and whether § 1983 allows a plaintiff to hold a defendant-supervisor liable for that violation may depend on whether that defendant-supervisor, rather than only his subordinates, violated the plaintiff’s constitutional rights. For this reason, we properly address this question of supervisory liability now as part of the qualified immunity analysis. See al-Kidd v. Ashcroft, 580 F.3d 949, 964–65 (9th Cir. 2009) (explaining that questions of supervisory liability, though part of the substance of a § 1983 claim, are also properly a part of the qualified immunity inquiry). Defendant maintains that in order to show he violated Plaintiff’s clearly established constitutional rights, and therefore overcome his assertion of qualified immunity as well as hold him liable under § 1983, Plaintiff must demonstrate that he personally participated in such a violation with a sufficiently culpable state of mind. Defendant points out that Plaintiff does not allege Defendant was one of the 12 jail employees who told him and the individuals who inquired about posting bail on his behalf that he could not post the bail set in his arrest warrant until he had been arraigned by a judge. Nor does Plaintiff contend Defendant personally instructed those employees to refuse to accept bail from Plaintiff the weekend of Friday, April 6, 2007. According to Defendant in his opening brief, the “policy of the court clerk’s office, and no action” by him deprived Plaintiff of his federally protected rights. Defendant argues, as a result, Plaintiff has not shown he committed any act which violated Plaintiff’s rights or that he acted with deliberate indifference to Plaintiff’s rights. 1. Defendant’s argument implicates important questions about the continuing vitality of supervisory liability under § 1983 after the Supreme Court’s recent decision in Ashcroft v. Iqbal, 129 S. Ct. 1937 (2009). 2 Prior to Iqbal, we held that § 1983 does not allow a plaintiff to hold an individual government official liable “under a theory of respondeat superior.” Gagan v. Norton, 35 F.3d 1473, 1476 n.4 (10th Cir. 1994) (citing Monell v. Dep’t of Social Serv., 436 U.S. 658 (1978)); see also Poolaw, 565 F.3d at 732 (“[A] supervisory relationship alone is insufficient for liability under § 1983.”). We explained “‘[t]his does not mean that a supervisor may not be liable for the injuries caused by the conduct of one of his subordinates. It 2 Neither the district court nor the parties in their appellate briefs discussed the import of the Supreme Court’s May 2009 decision in Iqbal. 13 does mean that his liability is not vicarious, that is, without fault on his part.’” Serna, 455 F.3d at 1151 (quoting Scull v. New Mexico, 236 F.3d 588, 600 (10th Cir. 2000)). “A crucial difference exists between liability as master (respondeat superior) and direct liability.” McClelland v. Facteau, 610 F.2d 693, 695 (10th Cir. 1979). While respondeat superior imposes liability for public policy reasons upon masters though they are not at fault in any way, direct liability only imposes liability where the plaintiff has shown the supervisor himself “breached a duty to plaintiff which was the proximate cause of the injury.” Id. We consequently concluded that “‘[i]ndividual liability under § 1983 must be based on personal involvement in the alleged constitutional violation’” but “[p]ersonal involvement is not limited solely to situations where a defendant violates a plaintiff’s rights by physically placing hands on him.” Fogarty, 523 F.3d at 1162 (quoting Foote v. Spiegel, 118 F.3d 1416, 1423 (10th Cir. 1997)). Personal involvement does not require direct participation because § 1983 states “‘[a]ny official who “causes” a citizen to be deprived of her constitutional rights can also be held liable.’” Buck v. City of Albuquerque, 549 F.3d 1269, 1279 (10th Cir. 2008) (quoting Snell v. Tunnell, 920 F.2d 673, 700 (10th Cir. 1990)). Even before Iqbal, it was not enough in our circuit “for a plaintiff merely to show defendant was in charge of other state actors who actually committed the violation. Instead, . . . the plaintiff must establish a deliberate, intentional act by the supervisor to violate constitutional rights.” Serna, 455 F.3d at 1151 (internal 14 quotations omitted). In sum, to impose § 1983 liability the plaintiff first had to establish “the supervisor’s subordinates violated the [C]onstitution.” Id. (internal quotations omitted). Then, the plaintiff must demonstrate “an ‘affirmative link’ between the supervisor and the violation . . . .” Id. Over time, this “affirmative link” requirement came to have three related prongs: (1) personal involvement, (2) sufficient causal connection, and (3) culpable state of mind. A plaintiff could establish the defendant-supervisor’s personal involvement by demonstrating his “‘personal participation, his exercise of control or direction, or his failure to supervise,’” Poolaw, 565 F.3d at 732 (quoting Green v. Branson, 108 F.3d 1296, 1302 (10th Cir. 1997)), or his “knowledge of the violation and acquiesce[nce] in its continuance.” Jenkins v. Wood, 81 F.3d 988, 995 (10th Cir. 1996)). A defendant- supervisor’s promulgation, creation, implementation, or utilization of a policy that caused a deprivation of plaintiff’s rights also could have constituted sufficient personal involvement. See Meade v. Grubbs, 841 F.2d 1512, 1528 (10th Cir. 1988) (stating § 1983 liability may be imposed on a supervisor who either “established or utilized an unconstitutional policy or custom” or “breached a duty imposed by state or local law which caused the constitutional violation”); Martin A. Schwartz, Section 1983 Litigation: Claims and Defenses, § 7.19[C] (4th ed. 2010) (“[S]upervisory officials who promulgate[d] policies that [were] enforced by subordinates [were] liable if the enforcement of the policy cause[d] a violation of federally protected 15 rights.”). 3 A plaintiff then had to establish the “‘requisite causal connection’” by showing “‘the defendant set in motion a series of events that the defendant knew or reasonably should have known would cause others to deprive the plaintiff of her constitutional rights.’” Poolaw, 565 F.3d at 732–33 (quoting Snell, 920 F.2d at 700). And, finally, the plaintiff also had to show the supervisor had a culpable state of mind, meaning “the supervisor acted knowingly or with ‘deliberate indifference’ that a constitutional violation would occur.” Serna, 455 F.3d at 1151, 1154. 4 We did not 3 Accord Hernandez v. Keane, 341 F.3d 137, 145 (2d Cir. 2003) (“The liability of a supervisor under § 1983 can be shown [by] . . . creation of a policy or custom that sanctioned conduct amounting to a constitutional violation, or allowing such a policy or custom to continue.”); Zatler v. Wainwright, 802 F.2d 397, 401 (11th Cir. 1986) (“Personal participation . . . is only one of several ways to establish the requisite causal connection. An official may also be liable where a policy or custom that he established or utilized results in deliberate indifference to an inmate’s constitutional rights.”); Wagner v. Bonner, 621 F.2d 675, 679 (5th Cir. 1980) (concluding that plaintiffs satisfied their burden against a sheriff under § 1983 when they presented evidence from which a reasonable jury could have concluded that sheriff’s deputies acted pursuant to policies implemented by the sheriff); Duchesne v. Sugarman, 566 F.2d 817, 831 (2d Cir. 1977) (“It is not necessary for § 1983 liability that the appellees directed any particular action with respect to these specific individuals, only that they affirmatively promoted a policy which sanctioned the type of action which caused the violations.”). 4 In City of Canton v. Harris, 489 U.S. 378, 388 (1989), the Supreme Court held “the inadequacy of police training may serve as the basis for § 1983 liability only where the failure to train amounts to deliberate indifference to the rights of persons with whom the police come into contact.” After Canton, “deliberate indifference or reckless disregard” became the primary governing standard for supervisory liability. 1 Sheldon Nahmod, Civil Rights & Civil Liberties Litigation: The Law of Section 1983 § 3:100 (4th ed. 1997 & Supp. 2009). See also Woodward v. City of Worland, 977 F.2d 1392, 1399 & n.11 (10th Cir. 1992) (explaining that after Canton’s “explicit holding . . . of a scienter requirement for § 1983 liability” a plaintiff must establish supervisory liability by alleging more than gross negligence by demonstrating “an intentional, conscious, and deliberate act by the defendant participating in, or knowingly acquiescing in, the unconstitutional behavior”); Sample v. (continued...) 16 view these requirements as necessarily distinct. Proof of a supervisor’s personal direction or knowledge of and acquiescence in a constitutional violation often sufficed to meet the personal involvement, causal connection, and deliberate indifference prongs of the affirmative link requirement for § 1983 supervisory liability. See Jenkins, 81 F.3d at 995 (concluding a plaintiff may satisfy the requirements for imposing supervisory liability under § 1983 by “showing the defendant-supervisor personally directed the violation or had actual knowledge of the violation and acquiesced in its continuance”); Woodward, 977 F.2d at 1399–1400 & n.11 (implying that personal direction or actual knowledge and acquiescence demonstrates deliberate indifference).5 4 (...continued) Diecks, 885 F.2d 1099, 1117–18 (3d Cir. 1989) (“Although the issue here is one of individual liability rather than of the liability of a political subdivision, we are confident that . . . the standard of individual liability for supervisory public officials will be found to be no less stringent than the standard of liability for the public entities that they serve. In either case, a ‘person’ is not the ‘moving force [behind] the constitutional violation’ of a subordinate . . . unless that ‘person’ . . . has exhibited deliberate indifference to the plight of the person deprived.” (quoting Canton, 489 U.S. at 389)); contra Hernandez, 341 F.3d at 145 (“The liability of a supervisor under § 1983 can be shown [by] . . . grossly negligent supervision of subordinates who committed a violation.” (emphasis added)). 5 Our sister circuits developed similar requirements. See Bennett v. City of Eastpointe, 410 F.3d 810, 818 (6th Cir. 2005) (“At a minimum, a § 1983 plaintiff must show that a supervisory official at least implicitly authorized, approved or knowingly acquiesced in the unconstitutional conduct of the offending subordinate.” (internal quotations omitted)); Whitfield v. Meléndez-Rivera, 431 F.3d 1, 14 (1st Cir. 2005) (“[A] supervisor may only be held liable [under § 1983] where (1) the behavior of [his] subordinates results in a constitutional violation and (2) the [supervisor’s] action or inaction was affirmatively link[ed] to the behavior in the sense that it could be characterized as supervisory encouragement, condonation or acquiescence or gross negligence . . . amounting to deliberate (continued...) 17 Based at least in part on this jurisprudence, the district court denied Defendant qualified immunity. The district court concluded Defendant possessed responsibility for running the jail, acquiesced in the continuing operation of the court clerk’s or district judges’ policies at the jail, and that “his acquiescence caused or contributed in causing the deprivation of Plaintiff’s due process rights by another or others” because the policies prohibited felony arrestees, including those whose bond had been set, from posting bond after hours or until they had seen a judge. Dodds, Order at *3–*4 (Aug. 3, 2009). From Defendant’s acquiescence in the continued operation 5 (...continued) indifference.” (internal quotations omitted)); Hernandez, 341 F.3d at 145 (“The liability of a supervisor under § 1983 can be shown [by] . . . : (1) actual direct participation in the constitutional violation, (2) failure to remedy a wrong after being informed through a report or appeal, (3) creation of a policy or custom that sanctioned conduct amounting to a constitutional violation, or allowing such a policy or custom to continue, (4) grossly negligent supervision of subordinates who committed a violation, or (5) failure to act on information indicating that unconstitutional acts were occurring.”); Randall v. Prince George’s County, 302 F.3d 188, 206 (4th Cir. 2002) (stating that “supervisory liability may attach under § 1983 if a plaintiff can establish . . . . (1) that the supervisor had actual or constructive knowledge that his subordinate was engaged in conduct that posed a pervasive and unreasonable risk of constitutional injury to citizens like the plaintiff; (2) that the supervisor’s response to that knowledge was so inadequate as to show deliberate indifference to or tacit authorization of the alleged offensive practices; and (3) that there was an affirmative causal link between the supervisor’s inaction and the particular constitutional injury suffered by the plaintiff” (internal quotations omitted)); Baker v. Monroe Twp., 50 F.3d 1186, 1194 (3d Cir. 1995) (explaining that an individual may be liable under § 1983 if he participated in violating the plaintiff’s rights, or he directed others to violate them, or as a supervisor he had knowledge of and acquiesced in his subordinates’ violations); Jones v. City of Chicago, 856 F.2d 985, 992–93 (7th Cir. 1988) (concluding that to establish supervisory liability under § 1983 “[t]he supervisors must know about the conduct and facilitate it, approve it, condone it, or turn a blind eye for fear of what they might see. They must in other words act either knowingly or with deliberate, reckless indifference”). 18 of the clerk’s or district judges’ policies at the jail, the district court found a reasonable jury could infer Defendant had acted with “deliberate indifference to the due process rights of arrestees whose bail had been pre-set to be free of detention.” Id. at *4. As a result, the district court determined Defendant could “be liable for participating or acquiescing in the deprivation of Plaintiff’s Fourteenth Amendment rights.” Id. (citing Serna, 455 F.3d at 1151–52 2. But then, as the saying will surely go, came Iqbal. Federal officials arrested and detained Javaid Iqbal, a Pakistani citizen, in the United States shortly after September 11, 2001. Iqbal, 129 S. Ct. at 1942. In his Bivens suit, he alleged inter alia that John Ashcroft, the former Attorney General, and Robert Mueller, the Director of the Federal Bureau of Investigation (FBI), violated his equal protection rights. Id. at 1942–43. Justice Kennedy, writing for the Supreme Court, explained that because Bivens is “‘the federal analog’” to § 1983 suits, “Government officials may not be held liable for the unconstitutional conduct of their subordinates under a theory of respondeat superior.” Id. at 1948 (citing Monell, 436 U.S. at 691). This statement alone adds nothing new to our law or that of our sister circuits. But, the Court then declared “[b]ecause vicarious liability is inapplicable to Bivens and § 1983 suits, a plaintiff must plead that each Government-official defendant, through the official’s own individual actions, has violated the Constitution.” Id. It explained that “[t]he factors necessary to establish a Bivens violation” and presumably a 19 § 1983 violation, “will vary with the constitutional provision at issue.” Id. Consequently, when a plaintiff alleges “invidious discrimination in contravention of the First and Fifth Amendments” she “must plead and prove that the defendant acted with discriminatory purpose.” Id. Iqbal, as a result, had to plead that Ashcroft and Mueller “adopted and implemented the . . . policies at issue not for a neutral . . . investigative reason but for the purpose of discriminating on account of race, religion, or national origin.” Id. at 1948–49. Ashcroft and Mueller’s alleged deliberate indifference to or knowledge and acquiescence in their subordinates’ unconstitutional conduct or discriminatory animus, alone, did not amount to the state of mind required to establish Ashcroft and Mueller violated equal protection guarantees—purposeful discrimination—and the Court dismissed Iqbal’s claims against them. Id. at 1949, 1952. Thus, when a plaintiff sues an official under Bivens or § 1983 for conduct “arising from his or her superintendent responsibilities,” the plaintiff must plausibly plead and eventually prove not only that the official’s subordinates violated the Constitution, but that the official by virtue of his own conduct and state of mind did so as well. Id. at 1949. Much has been made about this aspect of Iqbal, but consensus as to its meaning remains elusive. Justice Souter in his dissent, joined by Justices Stevens, Ginsburg, and Breyer, concluded the majority did not merely narrow, but rather eliminated supervisory liability altogether. Id. at 1957 (Souter, J., dissenting). He surmised that even if Iqbal’s complaint sufficiently alleged Ashcroft and Mueller’s 20 “knowledge and deliberate indifference, [the majority] presumably would still conclude that the complaint fails to plead sufficient facts and must be dismissed.” Id. at 1958 (Souter, J., dissenting). Some agree with Justice Souter’s view of the majority’s opinion: “[T]he Supreme Court in Ashcroft v. Iqbal . . . appears to have ruled that even deliberate indifference with actual knowledge [of subordinates’ unconstitutional conduct] may not be sufficient for supervisory liability.” Nahmod, supra note 4, at § 3:100. Others posit the circuit courts’ supervisory liability standards “only survive Iqbal to the extent they authorize § 1983 liability against a supervisory official on the basis of the supervisor’s own unconstitutional conduct or, at least, conduct that set the unconstitutional wheels in motion. The issue, then is one of causation, i.e., whether the supervisor’s conduct was a proximate cause of the violation of the plaintiff’s constitutional rights.” Schwartz, supra III.B.1, at § 7.19[D]. We have already acknowledged that Iqbal may have changed the § 1983 supervisory liability landscape. See Lewis, 604 F.3d at 1227 n.3 (“At one end of the spectrum, the Iqbal dissenters seemed to believe that the majority opinion ‘eliminates . . . supervisory liability entirely,’ . . . . At the other end of the spectrum, the Ninth Circuit has read Iqbal as possibly holding that ‘purpose . . . is required’ merely in cases of alleged racial discrimination by governmental officials, given that Iqbal itself involved allegations of racial discrimination and such discrimination only violates the Constitution when it is intentional. . . . Many intermediate positions are also surely plausible.” (quoting al-Kidd, 580 F.3d at 976 n.25 (internal citations 21 omitted)). 6 But because our cases since Iqbal have thus far only presented allegations that do not satisfy our pre-Iqbal liability standard, we have not yet had occasion to determine what allegations of personal involvement and mental state do meet Iqbal’s stricter liability standard. 7 See id., 604 F.3d at 1227 n.3; Gallagher, 587 F.3d at 1069–71. Whatever else can be said about Iqbal, and certainly much can be said, we conclude the following basis of § 1983 liability survived it and ultimately resolves this case: § 1983 allows a plaintiff to impose liability upon a defendant-supervisor who creates, promulgates, implements, or in some other way possesses responsibility for the continued operation of a policy the enforcement (by the defendant-supervisor or her subordinates) of which “subjects, or causes to be subjected” that plaintiff “to the deprivation of any rights . . . secured by the Constitution . . . .” (quoting 42 6 We recognize that some panels of this Court after Iqbal, but without discussing Iqbal’s statements about supervisory liability, have continued to state part of our pre-Iqbal standard: “[T]here must be ‘an affirmative link . . . between the constitutional deprivation and either the supervisor’s personal participation, his exercise of control or direction, or his failure to supervise.’” Gallagher v. Shelton, 587 F.3d 1063, 1069 (10th Cir. 2009); see also Barrett v. Orman, No. 10-7000, 2010 WL 1499586, *3 (10th Cir. Apr. 15, 2010) (unpublished). 7 See also Barrett, 2010 WL 1499586 at *3 (concluding that because the plaintiff’s allegations against the defendant-supervisor do not show an affirmative link between the constitutional deprivation and his personal involvement, the plaintiff failed to state a claim on which relief could be granted against the defendant); Arocho v. Nafziger, No. 09-1095, 2010 WL 681679 at *11 (10th Cir. Mar. 1, 2010) (unpublished) (noting the plaintiff’s allegations do not satisfy our pre-Iqbal standard of § 1983 supervisory liability). In this section, we cite these unpublished cases not for their precedential value (because, of course, they have none), but merely as examples of our lingering pre-Iqbal jurisprudence. 22 U.S.C. § 1983). See Schwartz, supra III.B.1, at § 7.19[C] (positing that imposing liability upon officials for their promulgation of a policy the enforcement of which violates individuals’ federally protected rights holds such officials “responsible for their own wrongs rather than on the basis of respondeat superior liability” and, therefore, comports with Iqbal); see also Davis v. City of Aurora, __F. Supp. 2d__, __, 2010 WL 1348450, *17 (D. Colo. 2010) (“The exercise of control which may create the ‘affirmative link’ does not need to be the sort of on-the-ground, moment-to-moment control that defendants appear to suggest. Rather, the establishment or utilization of an unconstitutional policy or custom can serve as the supervisor’s ‘affirmative link’ to the constitutional violation. . . . [W]here an official with policymaking authority creates, actively endorses, or implements a policy which is constitutionally infirm, that official may face personal liability for the violations which result from the policy’s application.”). A plaintiff may therefore succeed in a § 1983 suit against a defendant-supervisor by demonstrating: (1) the defendant promulgated, created, implemented or possessed responsibility for the continued operation of a policy that (2) caused the complained of constitutional harm, and (3) acted with the state of mind required to establish the alleged constitutional deprivation. 8 See Summum v. City of Ogden, 297 F.3d 995, 1000 (10th Cir. 2002) 8 We do not mean to imply that these are distinct analytical prongs, never to be intertwined. The Supreme Court explained in the context of § 1983 municipal liability: Where a plaintiff claims that a particular municipal action itself violates (continued...) 23 (“Under 42 U.S.C. § 1983, Summum must establish (1) a violation of rights protected by the federal Constitution or created by federal statute or regulation, (2) proximately caused (3) by the conduct of a person (4) who acted under color of any statute, ordinance, regulation, custom[,] or usage, of any State or Territory or the District of Columbia.” (internal quotations and citations omitted)). Denying 8 (...continued) federal law, or directs an employee to do so, resolving these issues of fault and causation is straightforward. Section 1983 itself ‘contains no state-of-mind requirement independent of that necessary to state a violation’ of the underlying federal right. In any § 1983 suit, however, the plaintiff must establish the state of mind required to prove the underlying violation. Accordingly, proof that a municipality’s legislative body or authorized decisionmaker has intentionally deprived a plaintiff of a federally protected right necessarily establishes that the municipality acted culpably. Similarly, the conclusion that the action taken or directed by the municipality or its authorized decisionmaker itself violates federal law will also determine that the municipal action was the moving force behind the injury of which the plaintiff complains. Board of County Comm’rs v. Brown, 520 U.S. 397, 404–05 (1997) (quoting Daniels v. Williams, 474 U.S. 327, 330 (1986)); see also Barney v. Pulsipher, 143 F.3d 1299, 1307 (10th Cir. 1998) (“The Supreme Court observed in Brown that when an official municipal policy itself violates federal law, issues of culpability and causation are straightforward; simply proving the existence of the unlawful policy puts an end to the question.”). We think the same logic applies when the plaintiff sues a defendant-supervisor who promulgated, created, implemented or possessed responsibility for the continued operation of a policy that itself violates federal law. See Dodge v. Shoemaker, __F. Supp. 2d__, __, 2010 WL 924249, *13 (D. Colo. 2010) (“A supervisor’s liability under § 1983 must be predicated on the supervisor’s deliberate indifference and a plaintiff must show that an affirmative link exists between the [constitutional] deprivation and either the supervisor’s personal participation, his exercise of control or direction, or his failure to supervise . . . . Supervisory liability exists even without personal participation in the offensive act if supervisory officials implement a policy so deficient that the policy itself is a repudiation of constitutional rights and is the moving force of the constitutional violation.” (internal quotations and citations omitted)). 24 qualified immunity on the basis of such a showing complies with Iqbal’s requirement that § 1983 liability only be imposed upon those defendants whose own individual actions cause a constitutional deprivation because it requires plaintiffs prove each defendant took some act with the constitutionally applicable state of mind that caused the alleged constitutional violation. Moreover, Iqbal does not purport to overrule existing Supreme Court precedent. While Iqbal may very well have abrogated § 1983 supervisory liability as we previously understood it in this circuit in ways we do not need to address to resolve this case, we do not believe it altered the Supreme Court’s previously enunciated § 1983 causation and personal involvement analysis. Section 1983’s supervisory liability journey arguably began with the Supreme Court’s decision in Rizzo v. Goode, 423 U.S. 362 (1976). In Rizzo, the Court concluded a mayor, police commissioner, and other city officials could not be held liable under § 1983 for constitutional violations committed by unnamed individual police officers because: As the facts developed, there was no affirmative link between the occurrence of the various incidents of police misconduct and the adoption of any plan or policy by petitioners—express or otherwise—showing their authorization or approval of such misconduct. Instead, the sole causal connection found by the District Court between petitioners and the individual respondents was that in the absence of a change in police disciplinary procedures, the incidents were likely to continue to occur, not with respect to them, but as to the members of the classes they represented. Id. at 371. This reasoning implies defendant-supervisors may be liable under § 1983 where an “affirmative” link exists between the unconstitutional acts by their 25 subordinates and their “adoption of any plan or policy . . .—express or otherwise—showing their authorization or approval of such misconduct.” Rizzo, 423 U.S. at 371. Then Justice Rehnquist writing for the Court contrasted the facts presented by Rizzo with those in Hague v. CIO, 307 U.S. 496 (1939). In Hague, the Court affirmed the imposition of § 1983 liability upon defendants, including a mayor and chief of police, for adopting and enforcing deliberate policies “of excluding and removing the plaintiff’s labor organizers . . . . implemented ‘by force and violence’ on the part of individual policemen.” Rizzo, 423 U.S. at 374. And, in Allee v. Medrano, 416 U.S. 802 (1974), the plaintiffs alleged the complained of conduct “‘was but one part of a single plan by the defendants . . . . The numerous incidents of misconduct on the part of the named Texas Rangers . . . found beyond peradventure not only a ‘persistent pattern’ but one which flowed from an intentional, concerted, and indeed conspiratorial effort to deprive the organizers of their First Amendment rights and place them in fear of coming back.” Rizzo, 423 U.S. at 374–75. According to the majority in Rizzo, “[t]he focus in Hague and Medrano, was . . . on . . . a ‘pervasive pattern of intimidation’ flowing from a deliberate plan by the named defendants to crush the nascent labor organizations.” Id. at 375 (emphasis added). We conclude Rizzo’s explanation of these cases and its reasoning confirm we properly impose § 1983 liability upon individual defendants who act with the requisite degree of culpability to promulgate, create, implement, or 26 otherwise possess responsibility for the continued operation of policies that cause the deprivation of persons’ federally protected rights. 9 Not much later, the Supreme Court again addressed § 1983 liability in Monell v. Dep’t of Social Serv., 436 U.S. 658 (1978). After conducting an extensive review of the legislative history of § 1983, the Court overruled “Monroe v. Pape [365 U.S. 167 (1961)] insofar as it holds that local governments are wholly immune from suit under § 1983” id. at 663, and instead concluded plaintiffs may sue municipalities and other local government entities as “persons” under § 1983. “However, [it upheld] Monroe v. Pape, insofar as it holds that the doctrine of respondeat superior is not a basis for rendering municipalities liable under § 1983 for constitutional torts of their employees.” Id. at 663–64, n.7. Based in large part upon the language of § 1983 itself, the Court declared “it is when execution of a [local] government’s policy or custom, whether made by its lawmakers or by those whose edicts or acts may fairly be said to represent official policy, inflicts the injury that the government as an entity is responsible under § 1983.” Id. at 694. The Court explained that § 1983’s 9 Long before Iqbal, we interpreted Rizzo as imposing an “‘affirmative link’ requirement” that meant “that before a superior may be held [liable under § 1983] for acts of an inferior, the superior, expressly or otherwise, must have participated or acquiesced in the constitutional deprivations of which complaint is made.” Kite v. Kelley, 546 F.2d 334, 337 (1976); see also Buck, 549 F.3d at 1287 (stating that Rizzo found “that supervisory personnel could be held liable where the plaintiff demonstrates an ‘affirmative link’ between the constitutional violation and the defendant’s actions, typically through ‘the adoption of any plan or policy . . . showing authorization or approval.’” (quoting Rizzo, 423 U.S. at 371)). 27 language: “[A]ny person who, under color of any law . . . of any State, shall subject, or cause to be subjected, any person . . . to the deprivation of any rights . . . secured by the Constitution . . . shall . . . be liable to the party injured” . . . . plainly imposes liability on a government that, under color of some official policy, ‘causes’ an employee to violate another’s constitutional rights. At the same time, that language cannot be easily read to impose liability vicariously on governing bodies solely on the basis of the existence of an employer-employee relationship with a tortfeasor. Indeed, the fact that Congress did specifically provide that A’s tort became B’s liability if B ‘caused’ A to subject another to a tort suggests that Congress did not intend § 1983 liability to attach where such causation was absent. Id. at 691–92. 10 Since Monell, the Supreme Court has “consistently refused to hold municipalities liable under a theory of respondeat superior” but nonetheless has imposed liability upon municipalities when the enforcement of their policies or customs by their employees causes a deprivation of a person’s federally protected rights. Brown, 520 U.S. at 403. Therefore, the Court “require[s] a plaintiff seeking to impose liability on a municipality under § 1983 to identify a municipal ‘policy’ 10 Although Monell does not explicitly apply its prohibition on respondeat superior liability to suits against individual government officials, because its holding relied in large part on the language of § 1983—“any person who subjects or causes to be subjected”—many concluded the decision “made clear that respondeat superior cannot be applied either to superiors or to local government entities.” Nahmod, supra note 4, at § 3:93; see also Schwartz, supra III.B.1, at § 7.19[C] (“A combined reading of Rizzo and Monell thus supported the conclusion that § 1983 supervisory liability cannot be based upon respondeat superior.”); Iqbal, 129 S. Ct. at 1948 (citing Monell, 436 U.S. at 691, for the proposition that there exists “no vicarious liability for a municipal ‘person’ under 42 U.S.C. § 1983”). 28 or ‘custom’ that caused the plaintiff’s injury.” Id. “That is, a plaintiff must show that the municipal action was taken with the requisite degree of culpability and must demonstrate a direct causal link between the municipal action and the deprivation of federal rights.” Id. at 404 (emphasis added). Monell and its progeny clearly stand for the proposition that the very language of § 1983 provides for the imposition of liability where there exists an “affirmative” or “direct causal” link between a municipal person’s adoption or implementation of a policy and a deprivation of federally protected rights, and that imposing liability upon such a basis does not implicate respondeat superior. Nothing in Iqbal contradicts this longstanding interpretation of § 1983’s language. 3. Regardless of the fate of other theories of supervisory liability in Iqbal’s wake, we conclude the facts presented on summary judgment taken in the light most favorable to Plaintiff establish Defendant violated his federally protected rights, satisfying the first part of the qualified immunity test. Defendant contends that at most Plaintiff has demonstrated he failed to act and that he cannot be held liable under § 1983 for a failure to act. We disagree with his characterization of his role in the unconstitutional deprivation of Plaintiff’s liberty interest. Defendant may not have personally informed Plaintiff or the individuals who inquired on his behalf that he could not post the preset bail until he had seen a judge. And, Defendant may not have actually known of his subordinates’ enforcement of these policies with regard 29 to Plaintiff in particular. Nonetheless, Defendant admits to the existence and operation of the policies of not allowing felony arrestees to post lawfully set bail after hours or until they had seen a judge. Defendant also admits that these policies are the reason Plaintiff was detained despite the fact a judge had already approved bail in the amount of $5,000 in Plaintiff’s arrest warrant. Regardless of who first drafted the policies, Oklahoma law charged Defendant as sheriff with the responsibilities of running the county jail and accepting bail from all arrestees not charged with death-penalty eligible crimes. See 57 Okla. Stat. § 47 (“The sheriff . . . shall have charge of the county jail of his county and of all persons by law confined therein, and such sheriff . . . is hereby required to conform, in all respects, to the rules and directions promulgated pursuant to [74 Okla. Stat. § 192] and of the district judge and communicated to him by the proper authority.”); 22 Okla. Stat. § 1101(A) (“Except as otherwise provided by law, bail . . . shall be admitted upon all arrests in criminal cases where the offense is not punishable by death and in such cases it may be taken by and of the persons . . . authorized by law to arrest, [and] to imprison offenders . . . .”); Okla. Att’y. Gen. No. 69-138 (1969) (“In criminal cases except cases punishable by death . . . a sheriff is required to accept bail, under the terms of [22 Okla. Stat. § 1101 (1961)], for those persons jailed at times other than the normal working hours of the Court, provided proper bail has been set as provided by law.”); see also Meade, 841 F.2d at 1528 (“Under Oklahoma law, a sheriff is responsible for the proper management of a jail in his county and the conduct of his 30 deputies.”). Defendant evidently does not dispute this point. When asked at oral argument “And, in Oklahoma, isn’t the sheriff the final policymaker for the jail?” Defendant’s counsel answered “Yes.” When asked “So as the final policymaker [Defendant] was required at least in his official capacity to make sure that the policies of the jail followed the Constitution?” Defendant’s counsel responded, “I agree as in his official capacity he would be the final policymaker as to the jail.” Thus, as Defendant’s counsel conceded at oral argument, Oklahoma law made Defendant responsible for the policies that operated and were enforced by his subordinates at the jail. And under his watch, as he admits, the policies which caused Plaintiff’s constitutional injury continued to operate. Defendant has not provided any reason to conclude these policies were binding upon him as sheriff in particular or upon the jail in general. 11 Therefore, the facts, taken in the light most 11 When asked in a deposition who set the policy of not releasing felony arrestees until they had been arraigned by a judge despite their bond having been set, Defendant responded: “As far as I know, it was set by the district judges. I think it may have had something to do with the Court Clerk’s office also, because there is no one—no one there like at nighttime or weekends, stuff like that to accept bonds. So it has been there—it was that way when I got there, and I have been told it has been there for 20 years or 25 years.” Aplt’s App. at 63. When asked at oral argument if Defendant knew for a fact that the district court judges rather than the court clerk set the policies, Defendant’s counsel responded: “That was his understanding. I wouldn’t say that he knows for a fact or had anything beyond just his understanding from when he came into office, but it was his understanding that it initially started with the district court and that, again, this all before he came into office, of course.” Defendant did not provide the district court or this Court with a copy of these policies or even state definitively who first authored them. The court clerk did not shed any more light in her affidavit. She only stated that “Logan County has a local rule preventing individuals charged with a felony from posting bond until they have gone before a judge and (continued...) 31 favorable to Plaintiff, show Defendant may have played more than a passive role in the alleged constitutional violation—he may have deliberately enforced or actively maintained the policies in question at the jail. Plaintiff has thereby presented facts that establish personal involvement by Defendant in the alleged constitutional violation sufficient to satisfy § 1983. By Defendant’s own admission, the policies’ enforcement caused the constitutional violation before us. As a result, the facts show Defendant’s maintaining these policies at the jail caused Plaintiff to be deprived of his due process rights. 4. Now that we have concluded Plaintiff has shown facts that, if proven at trial, suffice to establish Defendant’s personal involvement caused the misconduct complained of, we address whether the facts show Defendant acted with the state of 11 (...continued) have been arraigned. To my knowledge, this rule has been in effect for at least the past 18 years. (See Ex.2, Bondable Offenses)” and that “it is the policy of the Court Clerk’s office not to permit the Sheriff’s office to accept bonds after hours on felony warrants.” Aplt’s App. at 138. She did not provide these local rules to the court or state who promulgated them. Exhibit 2 is titled “BONDABLE OFFENSES.” The document lists offenses that are presumably bondable, and then states “ALL MISDEMEANOR BONDS ARE $1000 PER CHARGES.” As best as we can tell, this is a list of some misdemeanor offenses and indicates the bond for those offenses is set at $1,000. This document does not say anything about the sheriff being forbidden to accept bail set by a judge in a warrant from felony arrestees after hours or until the arrestee has been arraigned. Furthermore, Defendant’s counsel seems to have essentially conceded that the court clerk does not have authority over him. In response to the panel’s statement that the court clerk “[c]ertainly doesn’t have authority to set jail policy” he stated “No, your honor. Absolutely not. I agree with that. The county clerk would not have the authority to run the county jail.” 32 mind required to establish Defendant committed a constitutional violation. The Court in Iqbal explained that the factors necessary to establish a § 1983 violation depend upon the constitutional provision at issue, including the state of mind required to establish a violation of that provision. Iqbal, 129 S. Ct. at 1949. Iqbal’s allegations that Ashcroft and Mueller knew and acquiesced in their subordinates’ unconstitutional conduct did not allege they themselves purposefully discriminated on the basis of race, religion, or national origin—the state of mind required to establish an equal protection violation. Id. As a result, Iqbal’s claims were insufficient to impose Bivens liability upon them without improperly implicating respondeat superior. Id. We therefore conclude that after Iqbal, Plaintiff can no longer succeed on a § 1983 claim against Defendant by showing that as a supervisor he behaved “knowingly or with ‘deliberate indifference’ that a constitutional violation would occur” at the hands of his subordinates, unless that is the same state of mind required for the constitutional deprivation he alleges. Serna, 455 F.3d at 1151 (quoting Green, 108 F.3d at 1302); see also Sandra T.E. v. Grindle, 599 F.3d 583, 588 (7th Cir. 2010) (concluding that after Iqbal when a plaintiff claims the defendant-supervisor violated her constitutional rights she must allege the defendant- supervisor acted with whatever state of mind is required to state the underlying constitutional violation she alleges); compare Daniels, 474 U.S. at 329–30 (holding “that § 1983 . . . contains no state-of-mind requirement independent of that necessary to state a violation of the underlying constitutional right. . . . But in any given 33 § 1983 suit, the plaintiff must still prove a violation of the underlying constitutional right; and depending on the right, merely negligent conduct may not be enough to state a claim” (internal citations omitted)); Brown, 520 U.S. at 405 (“Section 1983 itself ‘contains no state-of-mind requirement independent of that necessary to state a violation’ of the underlying federal right. In any § 1983 suit, however, the plaintiff must establish the state of mind required to prove the underlying violation.” (quoting Daniels, 474 U.S. at 330)). But given Plaintiff alleges a substantive due process violation, it appears Plaintiff must establish that Defendant acted with deliberate indifference to Plaintiff’s due process right to post preset bail. 12 See Daniels, 474 U.S. at 330–31 (concluding something more than simple negligence is required to establish an official deprived an individual of life, liberty, or property under the Fourteenth Amendment); Green v. Post, 575 F.3d 1294, 1301 (10th Cir. 2009) (explaining that when a plaintiff alleges a substantive due process violation and “when actual deliberation is practical, we will employ a deliberate indifference standard” (internal quotations and citations omitted)); Webber v. Mefford, 43 F.3d 12 Plaintiff does not specify that he alleges a “substantive” due process claim. But we construe his due process claim as a substantive one because, as we have explained many times, “[w]hen government action deprives a person of life, liberty, or property without fair procedures, it violates procedural due process.” U.S. v. Deters, 143 F.3d 577, 582 (10th Cir. 1998). Plaintiff, however, does not complain about a lack of procedure and, “thus, []he does not implicate the procedural component of the Due Process Clause.” Id. But “[t]he Clause ‘cover[s] a substantive sphere as well, barring certain government actions regardless of the fairness of the procedures used to implement them.’” Seegmiller v. Laverkin City, 528 F.3d 762, 766–67 (10th Cir. 2008). 34 1340, 1342 (10th Cir. 1994) (“[A] government official violates an individual’s Fourteenth Amendment rights by injuring his or her life, liberty, or property with deliberate or reckless intent.”). However, we need not say anything further as to what state of mind Plaintiff’s substantive due process claim demands because neither party challenges the district court’s conclusion that Plaintiff must show Defendant acted with deliberate indifference (albeit likely based upon our pre-Iqbal explanation of § 1983 supervisory liability given its citation to Serna, 455 F.3d at 1151–52). Therefore, we should not consider the issue. See Hernandez v. Starbuck, 69 F.3d 1089, 1093 (10th Cir. 1995) (“A court of appeals is not required to manufacture an appellant’s argument on appeal when it has failed in its burden to draw our attention to the error below. In the event of such a failure, the court will ordinarily consider the appellant’s point waived.” (internal quotations and citations omitted)). So, let us be clear: We do not pass judgment at this time on the state of mind required to establish a substantive due process violation based upon preventing an arrestee from posting preset bail. We assume, without deciding, deliberate indifference constitutes the required state of mind. In an appeal of a district court’s denial of qualified immunity at the summary judgment stage, we may not review the district court’s conclusion that the facts alleged support a particular factual inference. Zia Trust Co., 597 F.3d at 1152. But we may decide whether the facts “presented on summary judgment in the light most favorable to the plaintiff . . . amount to a violation of a clearly established right.” 35 Walker, 451 F.3d at 1155. In this case, conducting that analysis requires us to determine whether the facts support the view that Defendant acted with deliberate indifference to Plaintiff’s due process rights, i.e., support the view that Defendant knew his actions created a substantial risk of constitutional injury. See Serna, 455 F.3d at 1154–55 (explaining that the plaintiff “must point to evidence that would establish [the defendant] knew he was creating a situation that created a substantial risk of constitutional harm” to establish the defendant acted with deliberate indifference); Campbell, 586 F.3d at 840 (determining that a plaintiff may show the defendant acted with the deliberate indifference required to establish a Fourteenth Amendment due process violation by providing evidence the defendant “had subjective knowledge of a risk of harm and disregarded that risk by actions beyond mere negligence”). Plaintiff has shown facts from which a reasonable jury could infer Defendant knowingly created a substantial risk of constitutional injury to people like Plaintiff. Oklahoma law made Defendant, rather than the clerk or district judges, responsible for controlling the jail and accepting bail from arrestees like Plaintiff. Defendant admits that while he served as the sheriff he maintained policies that prevented felony arrestees whose bail had been set from posting bail after hours and before arraignment. Plaintiff had a liberty interest in being released once his bail had been set. Defendant does not suggest any “legitimate goal” behind preventing felony arrestees whose bail had been set from posting bail. We therefore agree with the district court that Plaintiff has shown facts that taken in the light most 36 favorable to him establish that Defendant acted with deliberate indifference and thereby violated his Fourteenth Amendment due process rights. IV. We now address whether Plaintiff has established Defendant violated a clearly established right. “The law is clearly established when a Supreme Court or Tenth Circuit decision is on point, or if the clearly established weight of authority from other courts shows that the right must be as plaintiff maintains.” Harman v. Pollock, 586 F.3d 1254, 1261 (10th Cir. 2009) (internal quotations omitted). We have explained that “clearly established” means “‘[t]he contours of the right must be sufficiently clear that a reasonable official would understand that what he is doing violates that right. This is not to say that an official action is protected by qualified immunity unless the very action in question has previously been held unlawful, but it is to say that in the light of pre-existing law the unlawfulness must be apparent.’” Id. (quoting Anderson v. Creighton, 483 U.S. 635, 640 (1987)). That Plaintiff had a liberty interest based upon federal law in being freed from detention once bail had been set and that his continued detention despite that liberty interest must be reasonably related to a legitimate goal to pass constitutional muster have been clearly established in our circuit (and others) since at least 1997 when we published Gaylor. See supra III.A. Plaintiff’s right to be free from unjustified detention after his bail was set was clearly established such that a reasonable official in Defendant’s position in April 2007 would have understood that his deliberately indifferent 37 maintenance of the policies that prevented arrestees from posting preset bail for no legitimate reason violated the Constitution. Nonetheless, Defendant contends that Gaylor does not clearly establish that he could be liable in his individual capacity for the jail’s policies which prevented arrestees whose bail had been set from posting bail. First, the “clearly established” prong of the qualified immunity inquiry asks whether the “‘[t]he contours of the right’” the plaintiff claims the defendant violated are “‘sufficiently clear that a reasonable official would understand that what he is doing violates that right.’” Harman, 586 F.3d at 1261 (quoting Anderson, 483 U.S. at 640) (emphasis added). Plaintiff has demonstrated at this stage in the litigation that the contours of the right he claims Defendant violated were sufficiently clear that a reasonable official in Defendant’s position would know his maintenance of policies that prevent arrestees with preset bail from posting bail for no legitimate reason violates the Fourteenth Amendment right to due process. Second, while Gaylor admittedly involved municipal liability, other cases of ours and the great weight of authority from other circuits clearly established by 2007 that officials may be held individually liable for policies they promulgate, implement, or maintain that deprive persons of their federally protected rights. See supra III.B.1. Thus, we think Plaintiff has shown facts that, if proven at trial, establish Defendant violated his clearly established rights. V. 38 Lastly, we note Defendant has repeatedly argued throughout this litigation that he relied on the court clerk’s or district court judges’ policies and that these policies caused Plaintiff’s continued detention, rather than any action on his part. But he has never claimed prior to oral argument that he is entitled to qualified immunity because his conduct was nonetheless objectively reasonable in light of his reasonable reliance on policies set by the court clerk or district judges. 13 While such an argument may be an obvious one given Defendant’s repeated assertions that he relied on these policies, he has simply failed to make it or even cite any supporting precedent. For this reason, the district court never considered whether Defendant’s reliance “on a state statute, regulation, or official policy” constituted extraordinary circumstances sufficient to warrant granting qualified immunity. Roska, 328 F.3d at 1251. Of course, when pointedly asked at oral argument whether it was objectively reasonable for Defendant to rely on these policies or whether our case law supported such a 13 Generally speaking, “a reasonably competent public official should know the law governing his conduct.” Harlow, 457 U.S. at 919. Once a plaintiff demonstrates the right at issue was clearly established, the defendant then bears the “burden to prove that her conduct was nonetheless objectively reasonable.” Roska ex rel. Roska v. Peterson, 328 F.3d 1230, 1251 (10th Cir. 2003) (citing Cannon v. City & County of Denver, 998 F.2d 867, 874 (10th Cir. 1993)) (emphasis added). Rarely, a defendant can demonstrate her conduct was nonetheless objectively reasonable if she demonstrates extraordinary circumstances that prove she “neither knew or should have known of the relevant legal standard.” Id. at 1247, 1251. “‘The circumstances must be such that the defendant was so “prevented” from knowing that his actions were unconstitutional that he should not be imputed with knowledge of a clearly established right.’” Id. at 1251 (quoting Cannon, 998 F.2d at 874). A defendant’s reliance “on a state statute, regulation, or official policy” may constitute just such extraordinary circumstances. Id. 39 claim, Defendant’s counsel answered the questions. But, again, Defendant’s counsel did not raise the argument. In addition, in his responses to those questions he generally couched the inquiry as a matter of official versus individual liability, i.e., whether a former sheriff can be held individually liable for reliance on that policy. 14 Moreover, “issues may not be raised for the first time at oral argument.” United States v. Abdenbi, 361 F.3d 1282, 1290 (10th Cir. 2004). And, we “should neither raise sua sponte an argument not advanced by a party either before the district court or on appeal, nor then advocate a particular position and resolve the appeal based on that advocacy.” Id.; see also Perry v. Woodward, 199 F.3d 1126, 1141, n.13 (explaining that although we may affirm a district court’s decision for any reason supported by the record, because the defendants had not adequately developed the alternative argument we refused to address it). “Given [Defendant’s] failure to properly raise, brief, and argue” that he is nonetheless entitled to qualified immunity because he reasonably relied on the policies at issue, which he bears the burden of 14 Panel: “Let me ask you a question about Roska. Roska also says that officials are generally held to have constructive knowledge of established law. Here we have, do you concede we have a clearly established unconstitutional policy under Gaylor? Defendant’s Counsel: “I agree that Gaylor stands for the proposition that where you have an official capacity claim that the policy discussed in Gaylor . . .” *** Panel: “So our question is whether . . . it’s reasonable to say that a court clerk’s policy can undermine a clearly established constitutional right . . . of which the sheriff has constructive knowledge according to Cannon?” Defendant’s Counsel: “I believe the question would be can the former sheriff be held individually liable for reliance on that policy.” 40 proving, we will not consider the argument. Id.; see also Roska, 328 F.3d at 1251 (noting that once a plaintiff demonstrates the right at issue was clearly established, the defendant then bears the “burden to prove that her conduct was nonetheless objectively reasonable.”). The judgment of the district court is hereby AFFIRMED. 41 09-6157, Dodds v. Richardson TYMKOVICH, J., concurring. I fully agree with the majority that the complaint sufficiently alleges former Sheriff Richardson violated clearly established law when he implemented the county court’s unconstitutional bail policy. 1 I write separately to further note the lack of clarity in the law of supervisory liability, and my view of how this may have been affected by the Supreme Court’s recent decision in Ashcroft v. Iqbal, 129 S. Ct. 1937 (2009). Federal law provides that “[e]very person who, under color of any statute, ordinance, regulation, custom, or usage, of any State or Territory or the District of Columbia, subjects, or causes to be subjected, any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges, or immunities secured by the Constitution and laws, shall be liable to the party injured.” 42 U.S.C. § 1983 (emphasis added). The phrase “causes to be subjected” suggests that liability exists for officials who did not directly violate constitutional rights, but, as the majority illustrates, the standard for 1 I do not intend to address situations in which the right at issue was not clearly established. Nor do I address the debate over whether the theory of supervisory liability advanced must itself be clearly established at the time of the violation. See Murrell v. School District, 186 F.3d 1238, 1251 (10th Cir. 1999); see also Poe v. Leonard, 282 F.3d 123 (2d Cir. 2002) (plaintiff must show that “the supervisory liability doctrine under which [she] wishes to hold [the supervisor] liable” is clearly established). Rather, I emphasize that the question of supervisory liability is more coherently viewed through the lens of causation. demonstrating that a supervisory official has caused a violation is far from clear. A short review of pertinent case law illustrates the lack of clarity. I. As an initial matter, the Supreme Court has long held that municipalities are not liable for the constitutional torts of their employees merely on a respondeat superior basis. Monell v. Dep’t of Soc. Servs., 436 U.S. 658, 691 (1978). Instead, municipalities are only liable for constitutional violations that they have directly caused. To prove that a municipality caused a violation of her constitutional rights, a plaintiff must show that an injury was caused by an “official municipal policy of some nature.” Id. An official policy need not be explicitly codified—in some contexts, decisions of high-ranking officials, even with regard to isolated events, are rightly viewed as official policy and therefore subject the municipality to liability. See Pembaur v. Cincinnati, 475 U.S. 469 (1986). This concept of “official policy” is even flexible enough to include situations in which the potential for constitutional violations was so obvious that a municipality can be held liable for failing to adequately train its employees. See Canton v. Harris, 489 U.S. 378 (1989). But it is essential to recall that in all these situations, liability is imposed “only where the municipality itself causes the constitutional violation at issue.” Id. at 385 (citing Monell, 436 U.S. at 694–95) (emphasis in original). These examples—official policy, decisions of high-ranking officials, and failure to 2 adequately train employees—are not rightfully regarded as theories of liability but should instead be viewed as theories of causation. As to supervisory liability, the added level of removal between the violation and the supervisor makes questions of causation even more difficult. The Supreme Court has yet to speak with much clarity on the theories of causation that could demonstrate a supervisory official’s liability for the constitutional violations carried out by a subordinate. Whether a supervisor has violated the plaintiff’s rights is dependent on whether the subordinate violated the Constitution—the supervisor cannot be liable if there was no violation. See City of Los Angeles v. Heller, 475 U.S. 796, 799 (1986) (“if the [employee] inflicted no constitutional injury on respondent, it is inconceivable that [the city] could be liable to respondent.”). And in some cases, the determination of whether a violation occurred turns on the subordinate’s state of mind. 2 What remains 2 The elements necessary to establish a § 1983 violation “will vary with the constitutional provision at issue.” Iqbal, 129 S. Ct. at 1948. This variation is required because § 1983 “is not itself a source of substantive rights, but a method for vindicating federal rights elsewhere conferred by those parts of the United States Constitution and federal statutes that it describes.” Baker v. McCollan, 443 U.S. 137, 145 n.3 (1979). For example, certain violations of the Eighth Amendment will require proof of deliberate indifference, see Estelle v. Gamble, 429 U.S. 97, 104 (1976), whereas the Equal Protection clause is only violated when the official had a discriminatory purpose, see Washington v. Davis, 426 U.S. 229, 240 (1976). But procedural due process violations focus on the sufficiency of the procedural protections afforded the plaintiff, not the state of mind of the officials who establish or apply the policies. See, e.g., Wilkinson v. Austin, 545 U.S. 209, 224–25 (2005). The state of mind required to prove a constitutional violation is distinct from the state of mind that may be required to show that a municipality caused a constitutional (continued...) 3 unclear is the state of mind that the supervisor must possess to be liable for causing such a violation. As the majority points out, the Supreme Court recently muddied further these already cloudy waters. In Ashcroft v. Iqbal, 129 S. Ct. 1937 (2009), the Court reiterated the long-standing rule from Monell that “[g]overnment officials may not be held liable for the unconstitutional conduct of their subordinates under a theory of respondeat superior.” 129 S. Ct. at 1948. In doing so, the Court explained that in “a § 1983 suit or a Bivens action—where masters do not answer for the torts of their servants—the term ‘supervisory liability’ is a misnomer. Absent vicarious liability, each Government official, his or her title notwithstanding, is only liable for his or her own misconduct.” Id. at 1949. Iqbal involved a claim of invidious discrimination, in which a plaintiff must demonstrate that an official undertook “a course of action ‘because of,’ not merely ‘in spite of,’ [the action’s] adverse effects upon an identifiable group.” Id. In that context—where the intent to discriminate is an element of the constitutional violation—the Supreme Court rejected the proposition that supervisors can be held liable for mere “knowledge and acquiescence in their subordinates’ use of discriminatory criteria to make classification decisions.” Id. 2 (...continued) violation. See Canton, 489 U.S. at 388 n.8 (“[T]he proper standard for determining when a municipality will be liable under § 1983 for constitutional wrongs does not turn on any underlying culpability test that determines when such wrongs have occurred.”). 4 at 1949. To prevail against these supervisory officials, the plaintiff would have to demonstrate that they “purposefully adopted a policy of [discriminating against plaintiffs] because of their race, religion, or national origin.” Id. at 1952. That is, the plaintiffs must show that the supervisory officials themselves acted with an impermissible motive, not merely that they knew of their subordinates’ impermissible motives and did not put a stop to their actions. Since the plaintiffs in Iqbal failed to sufficiently allege discriminatory intent on the part of the supervisors, their claims against the supervisory officials failed. The dissenters in Iqbal stated, perhaps hyperbolically, that the majority’s view “eliminat[ed] . . . supervisory liability entirely.” Id. at 1957 (Souter, J., dissenting). 3 Iqbal unfortunately did not provide a unified theory for the variety of supervisory liability cases we face. We do know supervisory liability under § 1983 is still only appropriate where the plaintiff can prove that the supervisor caused the violation. And in a case like Iqbal, where the constitutional violation 3 Our recent cases have noted this potentially significant development, but have not determined the scope of its impact on our case law. In Lewis v. Tripp, 604 F.3d 1221 (10th Cir. 2010), we noted the “significant debate about the continuing vitality and scope of supervisory liability, not only in Bivens actions, but also in § 1983 suits.” Id. at 1227 n.3; see also Mink v. Knox, No. 08-1250, ___ F.3d ___, 2010 WL 2802729 (10th Cir. July 19, 2010); Sheldon Nahmod, Constitutional Torts, Over-Deterrence and Supervisory Liability after Iqbal, 14 Lewis & Clark L. Rev. 279, 294–98 (2010) (discussing some alternatives). In both Lewis and Mink we avoided delving further into the morass given allegations that the supervisor in each case had directly participated in the constitutional violations at issue. 5 requires discriminatory intent, a supervisor does not cause a violation unless he or she actually intended for his or her subordinates to invidiously discriminate. Mere knowledge and acquiescence of, or even “deliberate indifference” towards, the discriminatory actions of employees now appears insufficient to prove causation, and thereby prove liability. Although the Supreme Court did not explicitly state it in this manner, the discriminatory intent of subordinates actually breaks the chain of causation if the supervisor did not also intend to discriminate. This analysis fits well with the Court’s precedent noting that “claims brought pursuant to § 1983 sound in tort,” and counseling therefore that § 1983 should be interpreted “in light of the ‘background of tort liability.’” Monterey v. Del Monte Dunes, 526 U.S. 687, 709 (1999) (quoting Monroe v. Pape, 365 U.S. 167, 187 (1961)). But Iqbal does not address constitutional violations that are based on a state of mind other than specific intent—for instance, a procedural due process violation, or an Eighth Amendment violation based on an official’s deliberate indifference. A supervisor is liable for these actions only when the supervisor can be fairly said to have caused the violation, but determining when this is the case can be tricky, to say the least. At the most basic level of formulation, our law requires an “affirmative link” between the constitutional violation and “either the supervisor’s personal participation, his exercise of control or direction, or his failure to supervise.” 6 Meade v. Grubbs, 841 F.2d 1512, 1527 (10th Cir. 1988) (quoting Specht v. Jensen, 832 F.2d 1516, 1524 (10th Cir. 1987)) (internal quotation marks omitted). Ordinarily, “there must be cause in fact between the conduct complained of and the constitutional deprivation.” Snell v. Tunnell, 920 F.2d 673, 700 (10th Cir. 1990). Applying this “cause in fact” reasoning to supervisory defendants, we concluded that the “requisite causal connection is satisfied if the defendant set in motion a series of events that the defendant knew or reasonably should have known would cause others to deprive the plaintiff of her constitutional rights.” Id.; see also Buck v. City of Albuquerque, 549 F.3d 1269, 1279–80 (10th Cir. 2008). At odds with the “set in motion” formulation, we also caution it is not enough to “merely show[] that a supervisor ‘should have known’ that a subordinate was violating someone’s constitutional rights.” Woodward v. City of Worland, 977 F.2d 1392, 1399 (10th Cir. 1992). Instead, only a supervisor’s actual knowledge of his subordinates’ behavior will demonstrate the requisite “deliberate, intentional act by the supervisor to violate constitutional rights.” Id. Negligence—even gross negligence—is insufficient to prove that the supervisor caused a violation. The exact method of demonstrating a causal link depends on the actions of the supervisor in relation to the subordinate that led to the violation. Supervisors sometimes directly order their subordinates to take an action, either in a specific 7 case, or by establishing some sort of policy. They may also learn of conduct taken by their subordinates and acquiesce in it after the fact or simply ignore it. Some supervisors may never learn of the unconstitutional actions of their subordinates, not because their subordinates were successful in hiding their behavior, but because the supervisor was “willfully blind” or deliberately indifferent. And supervisors may have a responsibility, as do municipalities, to ensure that their subordinates are properly trained—failure to carry out this duty may in some cases result in a violation. Just as there are various ways in which a supervisor can be said to have caused a violation, as outlined above, there are different levels of fault associated with these actions. We consider some of these actions to be blame-worthy enough that the supervisor should be liable. Our cases have suggested that a supervisor could be liable if he or she “established or utilized an unconstitutional policy or custom.” Meade, 841 F.2d at 1528. We have also stated, perhaps contradictorily, that supervisory liability under § 1983 “must be predicated on the supervisor’s deliberate indifference, rather than mere negligence.” Green v. Branson, 108 F.3d 1296, 1302 (10th Cir. 1997). We have even suggested that, in some cases, supervisors could be liable for violations that resulted from recklessness—“a conscious acceptance of a known, serious risk”—because this “requires proof that the defendant focused upon the risk of unconstitutional 8 conduct and deliberately assumed or acquiesced in such risk.” Woodward, 977 F.2d at 1399 n.11 (emphasis in original). In sum, our precedent has established, with varying levels of clarity, that a supervisor is only liable for violations that he caused, and that causation requires at least some degree of fault on the supervisor’s part. Exactly how this causation can be shown varies depending on the type of violation and the facts of the case. II. Iqbal holds that at least for some constitutional violations requiring specific intent, a supervisor cannot be liable for a subordinates’ violations unless the supervisor possessed that intent as well. Thus, in cases where an element of the constitutional violation by the subordinate is “discriminatory intent,” such as the racial discrimination claimed in Iqbal, the supervisor must share the state of mind. But in situations where the constitutional violation does not require specific intent, the supervisor “is only liable for his or her own misconduct.” Iqbal, 129 S. Ct. at 1949. Following this logic, several theories of liability are possible. First, a supervisor may directly order a subordinate to violate the plaintiff’s rights—for example, by deliberately creating a policy that denies the plaintiff due process, or utilizing an already-existing policy that has the same effect. See Meade, 841 F.2d at 1528; Hernandez v. Keane, 341 F.3d 137, 145 (2d Cir. 2003) (supervisory liability can be based on “creation of a policy or custom that sanctioned conduct 9 amounting to a constitutional violation, or allowing such a policy or custom to continue”); Cottone v. Jenne, 326 F.3d 1352, 1360 (11th Cir. 2003) (supervisory liability “when a supervisor’s custom or policy results in deliberate indifference to constitutional rights”) (internal punctuation omitted). Where the actions of the subordinates directly flow from the participation or conduct of the supervisor, liability may be appropriate. The supervisor deliberately caused the violation. Next, some cases say a supervisor may cause violations when he or she has actual knowledge of past constitutional violations being carried out by a subordinate, and does nothing to stop future occurrences. See Hernandez, 341 F.3d at 145 (supervisory liability can result from “failure to act on information indicating that unconstitutional acts were occurring”); Chavez v. Ill. State Police, 251 F.3d 612, 651 (7th Cir. 2001) (liability is appropriate for supervisors who “know about the conduct and facilitate it, approve it, condone it, or turn a blind eye for fear of what they might see”); Hartley v. Parnell, 193 F.3d 1263, 1269 (11th Cir. 1999) (widespread abuse that is “obvious, flagrant, rampant and of continued duration” can lead to supervisory liability if the supervisor does not act to correct these behaviors); but see Sheehee v. Luttrell, 199 F.3d 295, 300 (6th Cir. 1999) (even if supervisors know of unconstitutional behavior, “liability under § 1983 must be based on active unconstitutional behavior and cannot be based upon a mere failure to act”) (internal quotation marks omitted). 10 Finally, a series of cases requires a standard of deliberate indifference. Those types of cases include the failure to train, the failure to supervise, and potentially other supervisory shortcomings. See, e.g., Carr v. Castle, 337 F.3d 1221, 1228 (10th Cir. 2008) (analyzing a failure to train theory of liability in the context of an excessive force claim); Hernandez, 341 F.3d at 145 (supervisory liability can be based on “grossly negligent supervision of subordinates who committed a violation”); Brown v. Muhlenberg Twp., 269 F.3d 205, 216 (3d Cir. 2001) (supervisor can be liable if he was aware of an unreasonable risk of injury, a specific supervisory practice could have averted this injury, the supervisor was indifferent to that risk, and the injury resulted from the failure to employ the supervisory practice); Smith v. Brenoettsy, 158 F.3d 908, 911–12 (5th Cir. 1998) (supervisory liability is appropriate if the supervisor exercised deliberate indifference in failing to supervise or train, and the injury resulted from this failure); Otey v. Marshall, 121 F.3d 1150, 1155 (8th Cir. 1997) (a supervisor can be held liable “if his failure to train or supervise the offending actor caused the deprivation”). In those cases, we may find that a supervisor has somehow caused the violation to occur by an egregious failure to act. See, e.g., Canton, 489 U.S. at 390 n.10 (hypothetically, equipping officers with firearms but neglecting to train them on their appropriate use could result in liability); Board of Cnty. Comm’rs v. Brown, 520 U.S. 397, 407–08 (1997) (“[T]he existence of a pattern of tortious conduct by inadequately trained employees may tend to show that the 11 lack of proper training . . . is the ‘moving force’ behind the plaintiff's injury.”) (citing Canton, 489 U.S. at 390–91). In sum, our decisions hold that supervisors are liable for constitutional violations they cause. The exact contours of causation—especially regarding an official’s state of mind sufficient for liability—are uncertain in light of Iqbal. But for purposes of this case, Dodds alleges the sheriff deliberately implemented an unconstitutional bail policy that violated his clearly established rights as a pre- trial detainee and thereby caused him injury. As the majority ably demonstrates, his allegations are enough to survive summary judgment. 12