Earnest Shields was an Illinois prisoner in 2008 when he was lifting weights and ruptured the pectoralis tendon in his left shoulder. Although he received some medical attention for the injury, he did not receive the prompt surgery needed for effective treatment. Instead, through a series of oversights and delays by various people responsible for his medical care, too much time passed for surgery to do any good. Shields now suffers from a serious and permanent impairment that could have been avoided. So we must assume, in any event, as we review the grant of summary judgment against Shields on his claims arising from the official response to his injury.
After his release from prison, Shields filed suit against numerous defendants under 42 U.S.C. § 1983. He alleges that all defendants were deliberately indifferent to his serious medical needs and thus violated his rights under the Eighth Amendment to the Constitution. On appeal, Shields is pursuing claims against two groups of defendants.
The first group consists of Wexford Health Sources, Inc., a private company that provides medical care to Illinois prisoners under contract with the Illinois Department of Corrections, and four doctors who worked for Wexford and were directly involved in treating or failing to treat Shields. The second group consists of two doctors employed by the Southern Illinois University School of Medicine who examined Shields and recommended physical therapy rather than surgery. Shields contends these SIU defendants violated the Eighth Amendment and committed medical malpractice under state law.
The district court granted summary judgment for defendants on all of Shields’ constitutional claims and then declined to exercise supplemental jurisdiction over the medical malpractice claims against the SIU doctors. After judgment was entered, Shields filed a motion for relief under Federal Rule of Civil Procedure 60 asking to amend his complaint to include state-law medical malpractice claims against Wexford and the doctors it employed. The district court denied the motion. Shields appeals both the grant of summary judgment on his constitutional claims and the denial of his post-judgment motion to amend.1
This case illustrates the often arbitrary gaps in the legal remedies under § 1983 for violations of federal constitutional rights. Viewing the evidence through the lens of summary judgment, we can and must assume that Shields is the victim of serious institutional neglect of, and perhaps deliberate indifference to, his serious medical needs. The problem he faces is that the remedial system that has been built upon § 1983 by case law focuses primarily on individual responsibility. Under controlling law, as a practical matter, Shields must come forward with evidence that one or more specific human beings acted with deliberate indifference toward his medical needs.
Shields has not been able to do so. The Illinois Department of Corrections and its *786medical services contractor, Wexford, diffused responsibility for Shields’ medical care so widely that Shields has been unable to identify a particular person who was responsible for seeing that he was treated in a timely and appropriate way. Several of the individual defendants employed by Wexford were aware of portions of Shields’ course of treatment, but no one person was responsible for ensuring that Shields received the medical attention he needed. No one doctor knew enough that a jury could find that he both appreciated and consciously disregarded Shields’ need for prompt surgery.
The problem Shields faces also raises a serious question about how we should evaluate the responsibility of a private corporation like Wexford for violations of constitutional rights. The question is whether a private corporation should be able to take advantage of the holding of Monell v. Department of Social Services, 436 U.S. 658, 98 S.Ct. 2018, 56 L.Ed.2d 611 (1978), which requires a plaintiff suing a local government under § 1983 to show that the violation of his constitutional rights was caused by a government policy, practice, or custom. Our prior cases hold, but without persuasive explanations, that the Monell standard extends from local governments to private corporations. As we explain below, however, that conclusion is not self-evident. We may need to reconsider it if and when we are asked to do so. As state and local governments expand the privatization of government functions, the importance of the question is growing.
Given the state of the controlling law, though, we must ultimately affirm the summary judgment for all defendants on the constitutional claims. Shields is also barred from appealing the denial of his post-judgment motion to amend his complaint because his appeal from that denial was untimely.
I. Facts for Purposes of Summary Judgment
We review de novo the grant of summary judgment, construing all facts in the light most favorable to the non-moving party. Greeno v. Daley, 414 F.3d 645, 652 (7th Cir.2005). Summary judgment is appropriate when there is no dispute of material fact and the moving party is entitled to judgment as a matter of law. Id. Because we are reviewing a grant of summary judgment, we must give Shields as the non-moving party the benefit of conflicts in the evidence and any reasonable inferences from the evidence. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 250-51, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986); Williams v. City of Chicago, 733 F.3d 749, 755 (7th Cir.2013). Our account of the facts therefore is not necessarily accurate in an objective sense but reflects the evidence through the lens of summary judgment.
In 2008, plaintiff Earnest Shields was a prisoner at Hill Correctional Center in Ga-lesburg, Illinois. He was transferred to the Stateville Correctional Center in Ro-meoville, Illinois, in January 2009. Inmates at both Hill and Stateville receive their medical care from Wexford, a company that contracted with the Illinois Department of Corrections to provide medical care to inmates. Defendants Arthur Funk, Robert Migliorino, Richard Shute, and Ronald Schaefer are all physicians who were employed by Wexford and had some involvement in treating Shields. Dr. Funk was the regional medical director in charge of overseeing medical care at Hill. Dr. Migliorino was the medical director for Hill until October 7, 2008. After Dr. Migliorino left Hill, the medical director position rotated among several doctors, including Dr. Schaefer. Dr. Shute was em*787ployed by Wexford as a traveling physician serving several prisons.
Southern Illinois University School of Medicine (“SIU”) is part of a state university with main campuses in Carbondale and Springfield. SIU employed defendant Dr. David Olysav. Dr. John Froelich also worked at SIU as a resident.
On June 16, 2008, Shields injured his shoulder while lifting weights at Hill. Dr. Migliorino examined Shields that same day, diagnosed a possible dislocated shoulder, and had him sent to a hospital emergency room where an MRI was taken. The MRI seemed to show a partial tear of the supraspinatus tendon in Shields’ left shoulder. Dr. Migliorino recommended that Shields be seen by an orthopaedic surgeon. As required by Wexford procedure, Dr. Migliorino conducted a “collegial review” with Dr. Funk to obtain approval for his recommendation. Collegial reviews frequently took place over the telephone, and Dr. Funk did not review patients’ charts as part of the collegial review.
Dr. Funk approved Dr. Migliorino’s referral recommendation, and Shields was seen by Dr. Schierer, an orthopaedic surgeon who is not a defendant. Dr. Schierer diagnosed a different problem, a ruptured left pectoralis tendon. He recommended that Shields see a shoulder specialist for surgery. Dr. Migliorino then secured Dr. Funk’s approval through collegial review to have Shields see Dr. Clark, a shoulder specialist who is also not a defendant. Dr. Clark confirmed Dr. Schierer’s diagnosis but said he did not feel comfortable performing the surgery himself. Dr. Clark recommended that Shields see Dr. Gibbons, another shoulder specialist who is not a defendant, for the surgery. Dr. Migliorino seconded Dr. Clark’s recommendation and obtained Dr. Funk’s approval via collegial review. Shields was scheduled to visit Dr. Gibbons, but before the visit took place, Dr. Gibbons notified Hill that he, too, would not feel comfortable performing the necessary surgery. Dr. Clark recommended finding another shoulder specialist to perform it. By this time, almost two months had passed since Shields’ injury.
At this point, a critical error occurred. The regional director responsible for Hill (Dr. Funk) and Hill’s medical director (Dr. Migliorino) did not select specific doctors for referrals. When a referral was authorized, staff in Hill’s Medical Records department selected the specific doctor for the referral from a list of local specialists that they maintained. When Dr. Migliori-no and Dr. Funk referred Shields to a shoulder specialist for the third time, Hill’s list did not contain any more shoulder specialists. Hill staff therefore contacted Wexford staff to find an out-of-area shoulder specialist to examine Shields. Wex-ford staff consulted their list of out-of-area shoulder specialists, drew the name of SIU’s Dr. Olysav from the list, and provided it to Hill. Dr. Olysav is a board-certified orthopaedic surgeon, but he is not a shoulder specialist. For the present we must assume he was included on Wexford’s list by mistake. No one detected the error, and Shields was sent to SIU for evaluation by Dr. Olysav and Dr. Froelich.
After Dr. Froelich conducted a physical examination and took Shields’ medical history, Dr. Olysav examined Shields. Dr. Olysav’s treatment recommendation differed dramatically. All the specialists who had examined Shields up to that point had recommended surgery to repair the torn tendon. Dr. Olysav, however, recommended only physical therapy. He did not recommend that a follow-up visit take place, nor did he indicate whether or under what conditions any further treatment might be needed. Dr. Funk and Dr. Mi-gliorino approved Dr. Olysav’s recommen*788dation. They also did not double-check Dr. Olysav’s credentials before approving his recommendation, missing an opportunity to discover that Olysav was not in fact a shoulder specialist.
On August 27, 2008, the day after being prescribed physical therapy, Shields filed a formal grievance because he was not receiving the surgery that several doctors had told him he needed. The Hill employee who reviewed Shields’ grievance spoke with a health care unit administrator, learned that the last “specialist” to examine Shields had prescribed physical therapy rather than surgery, and denied the grievance on that basis. She, too, did not double-check Dr. Olysav’s credentials. A designee of Hill’s warden approved the denial based solely on the grievance and the response. He did not examine any other documents or conduct any independent investigation into the counselor’s conclusions.
Shields did not begin physical therapy at Hill until October 2008. He was not able to complete the therapy. The physical therapist wrote on Shields’ chart that he was in too much pain to continue. The therapist suggested evaluation by an orthopaedist if such an evaluation had not already taken place. Despite the therapist’s notes, no follow-up examination was conducted and no further treatment was scheduled. (Dr. Migliorino left Hill while Shields’ therapy was in progress.) In fact, there is no indication that anyone checked whether Shields’ therapy had resolved his injury, or even read the therapist’s note indicating that the physical therapy had not been successful.
A couple of months later, Dr. Schaefer, a Wexford traveling physician filling in for the departed Dr. Migliorino, was asked to lift the medical hold that had been placed on Shields. (A medical hold prevents a prisoner from being moved to a different prison during medical treatment, to ensure continuity of care.) Dr. Schaefer reviewed Shields’ chart, asked Wexford staff if further treatment was planned, was told no, and accordingly lifted the medical hold on December 17, 2008. He made this decision without examining Shields.
Shields was transferred to Stateville in January 2009. After visiting the Stateville health care unit several times in April and May complaining of shoulder pain, Shields was referred to a shoulder specialist at the University of Illinois — Chicago. That doctor confirmed in July 2009 that his left pectoralis tendon had been ruptured. Unfortunately for Shields, though, too much time had passed for surgery to be effective. The result is that Shields’ left shoulder is permanently atrophied. His chest has sunk in around the left pectoralis tear, and he will never regain anything resembling full function in his left arm. We must assume that these permanent injuries would have been prevented by timely surgery. Surgery is the standard treatment for a pectoralis tear and typically results in a favorable outcome — but only if it is done promptly.
II. Analysis
A. Eighth Amendment
Shields alleges that Wexford and Drs. Funk, Migliorino, Shute, and Schaefer (“the Wexford defendants”), and SIU and Drs. Olysav and Froelich (“the SIU defendants”) were deliberately indifferent to his serious medical needs in violation of the Eighth Amendment and actionable under § 1983. We address each group of defendants in turn.
1. The Wexford Defendants
Shields argues that Wexford and the doctors it employed were deliberately indifferent to his medical needs, in violation *789of the Eighth Amendment’s prohibition on cruel and unusual punishment. He also argues that Drs. Funk, Migliorino, Shute, and Schaefer, in their individual capacities and as Wexford employees, were deliberately indifferent to those needs.
a. Wexford Health Sources, Inc.
We consider first the claim against the Wexford corporation itself. The question posed here is how § 1983 should be applied to a private corporation that has contracted to provide essential government services — in this case, health care for prisoners. The answer under controlling precedents of this court is clear. Such a private corporation cannot be held liable under § 1983 unless the constitutional violation was caused by an unconstitutional policy or custom of the corporation itself. Respondeat superior liability does not apply to private corporations under § 1983. E.g., Iskander v. Village of Forest Park, 690 F.2d 126, 128 (7th Cir.1982). Because Shields has no evidence of an unconstitutional policy or custom of Wex-ford itself, these precedents doom his claim against the corporation.
For reasons we explain below, however, Iskander and our cases following it on this point deserve fresh consideration, though it would take a decision by this court sitting en banc or pursuant to Circuit Rule 40(e), or a decision by the Supreme Court to overrule those decisions. We start with the background of § 1983 and the Supreme Court cases relevant to the issue, then turn to circuit court decisions, and finally discuss reasons to question those circuit decisions and adopt a different approach for private corporations.
The law now codified as 42 U.S.C. § 1983 was enacted as part of the Civil Rights Act of 1871, also known as the Ku Klux Klan Act, to provide a private right of action against persons acting under col- or of state law who violated constitutional rights. See 17 Stat. 13, § 1. The statute provides in relevant part: “Every 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 in an action at law, suit in equity, or other proper proceeding for redress.... ” The 42nd Congress enacted the law as part of a series of steps to protect freed slaves and their supporters from abuses of local and state government power in the Reconstruction era.
The statute was not used often until the Supreme Court held in Monroe v. Pape, 365 U.S. 167, 183, 81 S.Ct. 473, 5 L.Ed.2d 492 (1961), that § 1983 could provide a federal remedy for constitutional violations even if the defendant’s action also violated state law and even if a state remedy was available. After Monroe, § 1983 became the most important vehicle for enforcing federal constitutional rights against state and local governments and their agents. Monroe also held, however, that a local government was not a “person” that could be sued under § 1983. 365 U.S. at 187-92, 81 S.Ct. 473.
Most defendants under § 1983 are public employees, but private companies and their employees can also act under color of state law and thus can be sued under § 1983. E.g., Wyatt v. Cole, 504 U.S. 158, 161-62, 112 S.Ct. 1827, 118 L.Ed.2d 504 (1992); Lugar v. Edmondson Oil Co., Inc., 457 U.S. 922, 937, 102 S.Ct. 2744, 73 L.Ed.2d 482 (1982). In a case involving a private company, the Supreme Court took for granted that the corporate defendant would be liable under § 1983 for *790a constitutional tort committed by its employee. In Adickes v. S.H. Kress & Co., 398 U.S. 144, 90 S.Ct. 1598, 26 L.Ed.2d 142 (1970), a woman sued both a police officer and a private corporation under § 1983 for race discrimination. The plaintiff was a white teacher who had entered a restaurant in Mississippi with several African American students. She had been refused service and was then arrested when she left the restaurant.
The Supreme Court reversed summary judgment for the restaurant and explained that the plaintiff could recover from the restaurant if she could prove “that a Kress employee, in the course of employment, and a Hattiesburg policeman somehow reached an understanding to deny Miss Adickes service in the Kress store, or to cause her subsequent arrest because she was a white person in the company of Negroes.” 398 U.S. at 152, 90 S.Ct. 1598. In other words, the Court indicated that a private corporation could be held liable under § 1983 on a theory of respondeat superior liability. Interestingly, Adickes was decided at a time when a municipal government could not be held liable at all under § 1983.
For present purposes, the next pivotal decision was Monell v. Department of Social Services, 436 U.S. 658, 98 S.Ct. 2018, 56 L.Ed.2d 611 (1978). Monell first overruled Monroe in part and held that a local government could be sued as a person under § 1983. Id. at 690, 98 S.Ct. 2018. The Court then considered the issue of respondeat superior liability under § 1983, and held that “respondeat superior is not a basis for rendering municipalities liable under § 1983 for the constitutional torts of their employees.” Id. at 663 n. 7, 98 S.Ct. 2018. The Court held instead that a local government could be held liable under § 1983 only if the government’s own policy or custom had caused the violation. Id. at 694, 98 S.Ct. 2018.
In a number of decisions since Monell, our court has applied the Monell standard to private corporations. We said it first in Iskander: “Moreover, just as a municipal corporation is not vicariously liable upon a theory of respondeat superior for the constitutional torts of its employees, [Monell, 436 U.S. at 694, 98 S.Ct. 2018], a private corporation is not vicariously liable under § 1983 for its employees’ deprivations of others’ civil rights.” 690 F.2d at 128; see also Gayton v. McCoy, 593 F.3d 610, 622 (7th Cir.2010); Rodriguez v. Plymouth Ambulance Service, 577 F.3d 816, 822 (7th Cir.2009); Woodward v. Correctional Medical Services of Illinois, Inc., 368 F.3d 917, 927 (7th Cir.2004). All other circuits that have addressed the issue have reached the same conclusion, extending the Monell standard to private corporations.2
Such a unified phalanx of decisions from our own and other circuits is entitled to considerable respect. Upon closer examination, however, there are substantial grounds to question the extension of the Monell holding for municipalities to private corporations.
A close look at the reasoning of Monell provides no persuasive reason to extend its holding to private corporations. Monell *791gave two reasons for barring respondeat superior liability for municipalities under § 1983. First, the Court focused on the language of § 1983, which imposes liability on a person who “shall subject, or cause to be subjected,” any person to a deprivation of Constitutional rights:
The italicized language [of causation] 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.
436 U.S. at 692, 98 S.Ct. 2018.
Second, the Court concluded that the legislative history of the Civil Rights Act of 1871 showed that Congress did not intend to impose respondeat superior liability on municipalities. Id. at 693, 98 S.Ct. 2018. The Court focused on a rejected proposal known as the Sherman Amendment. Directed at Ku Klux Klan activity in the Reconstruction-era South, the amendment would have held a municipality liable for the torts of private citizens not under the municipality’s control, and thus would have imposed in essence a generalized duty to keep the peace. See 436 U.S. at 692-94 & n. 57, 98 S.Ct. 2018. The amendment was rejected largely due to concerns about its constitutionality. See id. at 678-79, 98 S.Ct. 2018. The Monell Court seems to have concluded that if the 1871 Congress rejected the Sherman Amendment on constitutional grounds, then it similarly would have thought that respondeat superior liability for municipalities was unconstitutional, so respondeat superior liability for municipalities must be implicitly barred under § 1983. See id. at 693, 98 S.Ct. 2018. While the Court’s discussion is opaque, it was clearly focused on municipalities and did not consider private corporations, such as in Adickes v. Kress.
The rejection of respondeat superior liability for municipalities in Monell has been the subject of extensive analysis and criticism. See Board of County Com’rs v. Brown, 520 U.S. 397, 430-37, 117 S.Ct. 1382, 137 L.Ed.2d 626 (1997) (Breyer, J„ dissenting) (calling for reconsideration of Monell rejection of respondeat superior liability); City of Oklahoma City v. Tuttle, 471 U.S. 808, 834-44, 105 S.Ct. 2427, 85 L.Ed.2d 791 (1985) (Stevens, J., dissenting) (same); see also, e.g., Jack M. Beermann, Municipal Responsibility for Constitutional Torts, 48 DePaul L.Rev. 627 (1999); Peter H. Schuck, Municipal Liability Under Section 1983: Some Lessons from Tort Law and Organization Theory, 77 Geo. L.J. 1753 (1989); Larry Kramer and Alan O. Sykes, Municipal Liability Under Section 1983: A Legal and Economic Analysis, 1987 S.Ct. Rev. 249 (1987); Susanah M. Mead, 42 U.S.C. § 1983 Municipal Liability: The Monell Sketch Becomes a Distorted Picture, 65 N.C. L.Rev. 517 (1987); Karen M. Blum, From Monroe to Monell: Defining the Scope of Municipal Liability in Federal Courts, 51 Temple L.Q. 409 (1978). (A reader of these critiques will find citations to many more.) These commentators have pointed out many critical problems with Monell’s conclusion that re-spondeat superior claims against municipalities are not permitted under § 1983.
Perhaps the most important criticism to emerge from this literature is that Monell failed to grapple, with the fact that respon-deat superior liability for employers was a *792settled feature of American law that was familiar to Congress in 1871, when § 1983 was enacted. Congress therefore enacted § 1983 against the backdrop of respondeat superior liability and presumably assumed that courts would apply it in claims against corporations under § 1983. Cf. Smith v. Wade, 461 U.S. 30, 38-45, 103 S.Ct. 1625, 75 L.Ed.2d 632 (1983) (considering common law in 1871 to decide standard for punitive damages under § 1983); Carey v. Piphus, 435 U.S. 247, 257-59, 98 S.Ct. 1042, 55 L.Ed.2d 252 (1978) (considering common law in 1871 to decide that actual injury is needed to recover compensatory damages 'under § 1983); see generally Jack M. Beermann, A Critical Approach to Section 1983 With Special Attention to Sources of Law, 42 Stanford L.Rev. 51, 66-73 (1989).
The Court’s reliance on the Sherman Amendment is also problematic. The rejection of the proposal to hold municipalities liable for actions of private citizens it could not control says little about whether a municipality should be held liable for constitutional torts committed by its own employees acting within the scope of their employment. (More about this below, when we discuss reasons not to extend the Monell holding to private corporations.) Finally, the Court gave only cursory and tentative treatment to the strongest foundation for respondeat superior liability: an employer should be held responsible for the torts of employees whose actions it can control and from whose actions it profits. See 436 U.S. at 694 & n. 58, 98 S.Ct. 2018.
Given these flaws on the surface of its reasoning, Monell is probably best understood as simply having crafted a compromise rule that protected the budgets of local governments from automatic liability for their employees’ wrongs, driven by a concern about public budgets and the potential extent of taxpayer liability.
Of course, the critiques of Monell’s rejection of respondeat superior liability for municipalities have not yet persuaded the Supreme Court to reconsider that rule. Given our position in the judicial hierarchy, then, we are bound to follow Monell as far as municipal liability is concerned. We need not extend that holding, however, to the quite different context of private corporate defendants.
As noted, respondeat superior liability, which makes employers liable for their employees’ actions within the scope of their employment, is an old and well-settled feature of American law. See, e.g., Restatement (3d) of Agency § 2.04 (2006); Kerl v. Dennis Rasmussen, Inc., 273 Wis.2d 106, 682 N.W.2d 328, 334 (2004) (Sykes, J.) (respondeat superior “has been well-settled in the law of agency for perhaps as long as 250 years.”); Sword v. NKC Hospitals, Inc., 714 N.E.2d 142, 147-48 (Ind.1999); Adames v.. Sheahan, 233 U1.2d 276, 330 Ill.Dec. 720, 909 N.E.2d 742, 754-55 (2009). It is often justified through a deterrence theory. E.g., Kerl, . 682 N.W.2d at 336. Employers are less likely than employees to be judgment-proof and thus are more likely to be deterred by potential liability. Id. Plus, while potential liability for a single tort may not be enough to cause an employee to take more care, the specter of massive aggregate liability might spur the employer to take precautions. Employers are in the better position to take cost-effective measures to avoid causing injury and can absorb the costs of those precautions more easily than their individual employees. Id. All of this suggests that making employers liable for their employees’ torts may result in less tortious behavior overall. We should not insulate employers from respondeat superior liability under § 1983 without powerful reasons to do so.
*793The text of § 1983 does not foreclose respondeat superior liability for corporations. “Cause” has many legal meanings, but it generally refers to proximate causation, which is something broader than immediate, direct causation. See National Union Fire Ins. Co. v. Mead Johnson & Co., LLC, 735 F.3d 539, 547 (7th Cir.2013); United States v. Laraneta, 700 F.3d 983, 990 (7th Cir.2012). The requirement of causation certainly does not generally preclude respondeat superior liability for a given tort. See Dobbs’ Law of Torts § 425 (2d ed); Kerl, 682 N.W.2d at 334; Sword, 714 N.E.2d at 147-48; Adames, 330 Ill.Dec. 720, 909 N.E.2d at 754-55. The causation requirement affects whether an individual employee can be found liable for a wrong in the first place, not whether his or her wrong can be imputed to the employer under respondeat superior. Courts routinely applying respondeat superior liability to corporations do not ask whether the corporation “caused” the wrong by its employee. They ask instead only whether the employee was acting within the scope of employment.
The Monell Court’s interpretation of the legislative history of the Civil Rights Act of 1871 similarly does not indicate that Congress rejected the idea of respondeat superior liability for corporations. The rejected Sherman Amendment, which the Monell Court relied on to reject responde-at superior liability for municipalities, would have made a “county, city, or parish” vicariously liable for acts of violence committed by private citizens. Monell, 436 U.S. at 667, 98 S.Ct. 2018. The amendment was designed to make municipalities vicariously liable for violence and property damage inflicted by the Ku Klux Klan, regardless of whether the municipality knew of the Klan’s planned activity in advance or had the power to stop it. Id. at 667-68, 98 S.Ct. 2018. That proposition simply is not analogous to imposing liability on private corporations for the tortious behavior of their own employees acting within the scope of employment. Nothing in the Monell treatment of the legislative history bars respondeat superior liability for corporations.
Other Supreme Court decisions also do not require the extension of Monell to this new context. Monell itself said nothing about whether its new “policy or custom” standard would apply to private companies sued under § 1983. Nor did Monell even mention Adickes’ almost reflexive application of respondeat superior liability to a private company under § 1983. Adickes remains good law, see Lugar, 457 U.S. at 930-31, 102 S.Ct. 2744 (quoting Adickes’ discussion of private liability under § 1983 at length and with approval), so current Supreme Court precedent seems to support rather than reject respondeat superi- or liability for private corporations under § 1983. Further, since Monell, the Supreme Court has never held that a private corporation may take advantage of the Monell standard that applies to local governments. That suggests that we should treat a private corporation like any other “person” who causes a constitutional violation and that respondeat superior liability should apply.
Moreover, in the related context of qualified immunity under § 1983, the Court has distinguished between employees of municipalities and employees of private corporations. In both Richardson v. McKnight, 521 U.S. 399, 412, 117 S.Ct. 2100, 138 L.Ed.2d 540 (1997), and Wyatt v. Cole, 504 U.S. 158, 167-68, 112 S.Ct. 1827, 118 L.Ed.2d 504 (1992), the Supreme Court based its conclusion on grounds of both history and policy, focusing on differences between private actors and governments. Despite a long history of private corporations performing state functions, there is no tradition of providing immunity to their employees. Richardson, 521 U.S. *794at 405, 117 S.Ct. 2100. Further, unlike municipalities, private corporations are subject to market pressures, which provide a set of incentives entirely different from those imposed by the democratic process. Id. at 409-11, 117 S.Ct. 2100. The conditions under which private corporations compete and provide government services are thus materially different from those affecting municipalities. Id. Due to these differences, private prison employees are barred from asserting qualified immunity from suit under § 1983. Id. at 412, 117 S.Ct. 2100.
The Court’s reasoning in Richardson and Wyatt suggests that we should not foreclose respondeat superior liability against private corporations under § 1983. Private prison employees and prison medical providers have frequent opportunities, through their positions, to violate inmates’ constitutional rights.3 It is also generally cheaper to provide substandard care than it is to provide adequate care. Private prisons and prison medical providers are subject to market pressures. Their employees have financial incentives to save money at the expense of inmates’ well-being and constitutional rights. The unavailability of qualified immunity for these employees is a deterrent against such conduct, but respondeat superior liability for the employer itself is likely to be more effective at deterring such actions. Insulating private corporations from responde-at superior liability significantly reduces their incentives to control their employees’ tortious behavior and to ensure respect for prisoners’ rights. The results of the current legal approach are increased profits for the corporation and substandard services both for prisoners and the public.
So the Supreme Court has not directly said whether Monell applies to private corporations, and there are powerful reasons to say no. Yet we and all other circuits that have considered the question have said yes. Why? It’s not easy to say. Our opinion in Iskander and virtually all of the circuit opinions after Monell simply cite one or more prior cases that all seem to trace back to the terse Fourth Circuit opinion in Powell v. Shopco Laurel Co., 678 F.2d 504 (4th Cir.1982). The relevant portion of that opinion said in full:
In Monell v. New York City Department of Social Services, 436 U.S. 658, 98 S.Ct. 2018, 56 L.Ed.2d 611 (1978), the Supreme Court held that a municipal corporation cannot be saddled with section 1983 liability via respondeat superior alone. We see this holding as equally applicable to the liability of private corporations. Two aspects of Monell exact this conclusion. The Court found section 1983 evincing a Congressional intention to exclude the imposition of vicarious answerability. For a third party to be' liable the statute demands of the plaintiff proof that the former “caused” the deprivation of his Federal rights. 436 U.S. at 691-92, 98 S.Ct. 2018. Continuing, the Court observed that the policy considerations underpinning the doctrine of respondeat superior insufficient to warrant integration of that doctrine into the statute. Id. at 694, 98 S.Ct. 2018. No element of the Court’s ratio decidendi lends support for distinguishing the case of a private corporation.
678 F.2d at 506.
There are good reasons to question the Powell conclusion. It overlooked the fact *795that Monell was focused on the Sherman Amendment, which would have imposed liability for mere failure to prevent harm caused by private citizens, not employees controlled by an employer. It also overlooked the fact that respondeat superior liability was already a well established part of the common law in 1871, so Congress could reasonably have expected the courts to apply the doctrine under § 1988. Perhaps most important, the Powell opinion simply overlooked the Monell Court’s special solicitude for municipalities and their budgets. These omissions counsel against adopting the Powell court’s conclusion.
For all of these reasons, a new approach may be needed for whether corporations should be insulated from respondeat superior liability under § 1983. Since prisons and prison medical services are increasingly being contracted out to private parties, reducing private employers’ incentives to prevent their employees from violating inmates’ constitutional rights raises serious concerns. Nothing in the Supreme Court’s jurisprudence or the relevant circuit court decisions provides a sufficiently compelling reason to disregard the important policy considerations underpinning the doctrine of respondeat superior. And in a world of increasingly privatized state services, the doctrine could help to protect people from tortious deprivations of their constitutional rights.4
If the Monell policy/custom standard did not apply here, we would reverse the district court’s grant of summary judgment for Wexford. Shields has offered evidence showing that the corporation was responsible for his health care. As an entity, the company knew of his injury, its severity, the need for surgery, and the complete failure of physical therapy. (Recall the therapist’s note that Shields had to stop the therapy because of pain.) Wexford nevertheless failed to ensure that Shields received the surgery he needed to avoid permanent impairment of his shoulder. These facts would support respondeat superior liability for Wexford and would lead us to reverse summary judgment for Wex-ford on that ground.
The facts in this case are also an excellent example of the problems generated by barring respondeat superior liability for corporations under § 1983. On the facts before us, it appears that Wexford structured its affairs so that no one person was responsible for Shields’ care, making it impossible for him to pin responsibility on an individual. If respondeat superior liability were available, Wexford could not escape liability by diffusing responsibility across its employees, and prisoners would be better protected from violations of their constitutional rights.
In view of these considerations, we have considered the possibility of circulating an opinion overruling Iskander and its progeny on this point for consideration by the entire court under Circuit Rule 40(e). Since Shields has not asked us to overrule those cases and Wexford has not had occasion to brief the issue, we have decided not to take that approach. A petition for re*796hearing en banc would provide an opportunity for both sides to be heard on this issue, and our decision is of course subject to review on certiorari. For now, this circuit’s case law still extends Monell from municipalities to private corporations. Iskander, 690 F.2d at 128; Gayton, 593 F.3d at 622; Rodriguez, 577 F.3d at 822. To recover against Wexford under our current precedent, Shields must offer evidence that his injury was caused by a Wexford policy, custom, or practice of deliberate indifference to medical needs, or a series of bad acts that together raise the inference of such a policy. Woodward, 368 F.3d at 927.
Shields attempts to proceed by showing a series of bad acts. He argues that mistakenly referring him to the wrong doctor (and failing to detect or correct that mistake), combined with failing to promptly discipline and eventually replace Dr. Migliorino, shows that Wexford was deliberately indifferent to his medical needs.
Such isolated incidents do not add up to a pattern of behavior that would support an inference of a custom or policy, as required to find that Wexford as an institution/corporation was deliberately indifferent to Shields’ needs. See Palmer v. Marion County, 327 F.3d 588, 596 (7th Cir.2003) (“proof of isolated acts of misconduct will not suffice; a series of violations must be presented to lay the premise of deliberate indifference”); Cornfield v. Consolidated High School Dist. No. 230, 991 F.2d 1316, 1326 (7th Cir.1993) (requiring “a pattern or series of incidents of unconstitutional conduct” in the absence of an explicit policy). Under our existing case law, summary judgment was properly granted for Wexford on this claim.
b. Drs. Funk, Migliorino, Shute, and Schaefer
Shields contends that Drs. Funk, Mi-gliorino, Shute, and Schaefer were all deliberately indifferent to his medical needs. However, he has not come forward with evidence that would allow a reasonable jury to find that any one of the Wexford doctors both knew there was a risk of harm to Shields and consciously disregarded that risk. See Mathis v. Fairman, 120 F.3d 88, 91 (7th Cir.1997).
The initial response to Shields’ injury was not deliberately indifferent. Dr. Migliorino saw Shields the day he was injured and sent him to an outside hospital emergency room. When an MRI revealed an injury, Dr. Migliorino referred Shields to an orthopaedic surgeon, and Dr. Funk approved the referral. There is no indication that this response was inadequate, let alone deliberately indifferent. Rather, Drs. Funk and Migliorino seem to have taken prompt action to diagnose Shields’ injury and to seek treatment for it.
Shields argues that Drs. Funk and Migliorino deliberately delayed the surgery he needed, referring him to endless doctors in search of a different, cheaper treatment recommendation. However, Drs. Funk and Migliorino referred Shields to the type of doctor who seemed well qualified to perform his surgery: an ortho-paedic surgeon specializing in the shoulder. Given that Drs. Funk and Migliorino did not select specific doctors for referrals, referring Shields to the type of doctor who would be able to treat his injury competently, with surgery or otherwise, was adequate. Each time a particular doctor said he was unwilling to try the surgery himself, Drs. Funk and Migliorino scheduled another referral promptly. We see no basis for holding Drs. Funk and Migliorino responsible for the specialists’ reluctance to do the surgery themselves. Under these circumstances, the multiple referrals *797do not permit an inference of deliberate indifference.
We must assume the referral to Dr. Olysav was a mistake since he was not the proper sort of specialist. The problem for Shields is that § 1983 requires proof of individual responsibility. There is no indication that Dr. Funk or Dr. Migliorino had any involvement in that mistaken referral, let alone that either one acted with deliberate indifference to Shields’ health in allowing the referral to go forward.
Approving Dr. Olysav’s treatment recommendation of physical therapy also was not deliberately indifferent. Shields points to the fact that all previous shoulder specialists who had seen his records had recommended surgery. However, choosing one treatment recommendation over another does not amount to deliberate indifference where both recommendations are made by qualified medical professionals, as Drs. Funk and Migliorino believed to be the case here. See Estate of Cole v. Fromm, 94 F.3d 254, 261 (7th Cir.1996) (“Mere differences of opinion among medical personnel regarding a patient’s appropriate treatment do not give rise to deliberate indifference.”).
Neither Dr. Funk nor Dr. Migliorino was deliberately indifferent to Shields’ medical needs after approving him for physical therapy as recommended by Dr. Olysav. Dr. Migliorino left Hill before Shields’ physical therapy was stopped because of pain, and there is simply no evidence that Dr. Funk knew of any continuing injury after he approved Dr. Migliorino’s recommendation that Shields undergo physical therapy. In particular, there is no evidence that Dr. Funk saw the physical therapist’s report that Shields could not complete the prescribed physical therapy because of the pain it caused. On these undisputed facts, a reasonable jury could not find that either Dr. Funk or Dr. Migliorino personally realized there was a risk that Shields’ injury was not being treated properly, and so could not find that either consciously disregarded that risk.
Finally, there is no evidence that would support a finding that Dr. Shute or Dr. Schaefer was deliberately indifferent to Shields’ medical needs. Dr. Schaefer’s job was limited to determining whether further treatment was planned. After determining it was not, he lifted the medical hold on Shields. He did not personally examine Shields, so he did not know that his injury had not been adequately addressed. Shields makes no arguments about Dr. Shute on appeal, so any claim against him is waived. Puffer v. Allstate Ins. Co., 675 F.3d 709, 718 (7th Cir. 2012).
2. The SIU Defendants
Summary judgment was also properly granted on Shields’ § 1983 claims against SIU, Dr. Olysav, and Dr. Froelich. Shields makes no arguments regarding SIU itself on appeal, so any claim against SIU is waived. Puffer, 675 F.3d at 718. Shields also has not shown that either Dr. Olysav or Dr. Froelich acted under color of state law within the meaning of § 1983. Whether a medical provider is a state actor is a functional inquiry, focusing on the relationship between the state, the medical provider, and the prisoner. Rodriguez, 577 F.3d at 826. A business like Wexford that contracts to provide medical care to prisoners undertakes “freely, and for consideration, responsibility for a specific portion of the state’s overall [constitutional] obligation to provide medical care for incarcerated persons,” id. at 827, and thus acts under color of state law for purposes of § 1983. On the other hand, we have held that medical providers who have “only an incidental or transitory relationship” *798with the penal system generally are not considered state actors. Id. at 827.
, The undisputed facts show here that Dr. Olysav and Dr. Froelich each had only an incidental and transitory relationship with the penal system. Neither was acting under color of state law for purposes of § 1983. Wexford referred Shields to Dr. Olysav for a one-time examination, which he performed with the help of Dr. Froelich. Dr. Olysav recommended physical therapy and had nothing more to do with the patient. Neither Dr. Olysav nor Dr. Froelich scheduled follow-up appointments with Shields or retained responsibility for his course of treatment, so they did not have a sufficiently direct relationship with him to find that they were acting under color of state law. See Rodriguez, 577 F.3d at 828. Their relationship with Wexford was similarly too attenuated to support the conclusion that they were acting under color of state law.
Shields correctly points out that Drs. Olysav and Froelich both indicated that they had treated inmates before and that Wexford had made the arrangements for those treatments. However, there is no evidence that Drs. Olysav and Froelich had a contract with Wexford or the prison, that their practices focused on treating inmates, or even that they regularly treated inmates as part of their practices. Standing alone, merely having treated inmates before does not establish the kind of close relationship between the doctors and Wexford required to find that they were state actors. In other words, the undisputed facts show that Dr. Olysav and Dr. Froelich had only incidental and transitory relationships with both Wexford and Shields. The undisputed facts do not allow a reasonable inference that these doctors acted under color of state law when they took the referral from Wexford. Summary judgment was properly granted for them on Shields’ § 1983 claims. Accordingly, the district court also did not abuse its discretion by declining to exercise supplemental jurisdiction over the state law claims for medical malpractice against these defendants. See 28 U.S.C. § 1367(c)(3).
B. Rule 60 Motion for Relief
Finally, Shields appeals the district court’s denial of his Rule 60 motion for relief. After the district court issued its opinion and final judgment, Shields filed a motion to amend his complaint to add state medical malpractice claims against the Wexford defendants.5 The court found that the motion to amend his complaint was brought under Rule 15(b), which directs courts to grant leave to amend freely where the opposing party will not be prejudiced by the amendment. Fed.R.Civ.P. 15(b)(1). The court then denied the motion, holding that Shields’ operative complaint (the second amended complaint) did not give the Wexford defendants fair notice of a medical malpractice claim against them. The court also held that allowing such a late amendment would prejudice the Wexford defendants by requiring them to devote resources to investigating and defending the claim after having already taken discovery and moved successfully for summary judgment on the claims actually asserted against them. On appeal, Shields argues that the Wexford defendants were on sufficient notice that medical malpractice claims were being asserted against them, so that having to defend those claims would not have prejudiced them.
*799Shields’ appeal of the district court’s denial of his motion to amend came too late. In civil cases where the federal government is not a party, a party ordinarily must file a notice of appeal within 30 days of the district court’s entry of judgment. Fed. R.App. P. 4(a)(1)(A). Filing a Rule 60 motion will toll the commencement of that time limit until the motion is resolved, as long as the Rule 60 motion is filed within 28 days of the district court’s entry of judgment. Fed. R.App. P. 4(a)(4)(A)(vi). Filing a post-judgment Rule 15(b) motion does not similarly toll the time to file an appeal. See Fed. R.App. P. 4(a)(4)(A).
Shields filed his Rule 60 motion 30 days after the district court entered its final judgment. If we consider his motion to amend as having been brought under Rule 60, he is not entitled to the tolling provided for in Federal Rule of Appellate Procedure 4(a)(4)(A)(vi) because he did not file his motion within 28 days of the district court’s final judgment, as required by the rule. If we consider Shields’ motion und§r Rule 15(b), then he did not file an appeal of the resolution of that motion within 30 days of the court’s entry of judgment in his case, as required by Rule 4(a)(1)(A). Either way, Shields’ appeal is untimely and his claims regarding the motion are barred.
Conclusion
There is ample evidence here that plaintiff Shields was the victim of delayed medical care that has left him with a serious and permanent injury that could have been avoided. The evidence suggests that he is the victim not of any one human being’s deliberate indifference but of a system of medical care that diffused responsibility for his care to the point that no single individual was responsible for seeing that he received the care he needed in a timely way. As a result, no one person can be held liable for any constitutional violation. Finally, Shields’ efforts to rely on state medical malpractice law against the Wex-ford defendants appear to have run afoul of procedures courts must follow for the timely and fair resolution of cases. As the case is presented to us, the judgment of the district court must be and is AFFIRMED.
. Shields also sued the Illinois Department of Corrections and two wardens of prisons where he was housed. The state agency itself is not subject to a suit for damages under § 1983, see Will v. Michigan Dep’t of State Police, 491 U.S. 58, 109 S.Ct. 2304, 105 L.Ed.2d 45 (1989), and the wardens did not have direct responsibility for Shields' medical care. The district court dismissed the claims against these non-medical defendants, and Shields does not challenge those dismissals on appeal.
. See Iskander, 690 F.2d at 128; Rojas v. Alexander’s Dept. Store, Inc., 924 F.2d 406, 408-09 (2d Cir.1990); Powell v. Shopco Laurel Co., 678 F.2d 504, 506 (4th Cir.1982); Street v. Corrections Corp. of America, 102 F.3d 810, 818 (6th Cir.1996); Lux v. Hansen, 886 F.2d 1064, 1067 (8th Cir.1989); Tsao v. Desert Palace, Inc., 698 F.3d 1128, 1139 (9th Cir.2012); DeVargas v. Mason & Hanger-Silas Mason Co., 844 F.2d 714, 723 (10th Cir.1988); Harvey v. Harvey, 949 F.2d 1127, 1129-30 (11th Cir.1992). See also Lyons v. National Car Rental Systems, Inc., 30 F.3d 240, 246 (1st Cir.1994); Defreitas v. Montgomery County Corr. Facility, 525 Fed.Appx. 170, 176 (3d Cir.2013).
. Although Richardson involved a private prison, some circuits (including our own) have applied Richardson to private medical providers, holding that they are similarly barred from asserting immunity under § 1983. See, e.g., Currie v. Chhabra, 728 F.3d 626, 631-32 (7th Cir.2013) (affirming denial of qualified immunity for private health care providers for jail); McCullum v. Tepe, 693 F.3d 696 (6th Cir.2012); Jensen v. Lane County, 222 F.3d 570 (9th Cir.2000); Hinson v. Edmond, 192 F.3d 1342 (11th Cir.1999).
. For more detailed critiques of the extension of Monell to private corporations, and for more detailed reviews of the policy considerations and the nuances in the case law, see Richard Frankel, Regulating Privatized Government Through § 1983, 76 U. Chi. L.Rev. 1449 (2009), and Barbara Kritchevsky, Civil Rights Liability of Private Entities, 26 Cardozo L.Rev. 35 (2004); see also Jack M. Beermann, Why Do Plaintiffs Sue Private Parties Under Section 1983?, 26 Cardozo L.Rev. 9, 27 (2004). As just one example of additional problems, the Monell policy/custom rule is difficult to apply to a private corporation. How does a court identify the relevant "final policymaker” in a corporation? Is it the CEO, the board of directors, the shareholders? What if the corporation is a subsidiary of another? See Kritchevsky, 26 Cardozo L.Rev. at 56-60.
. Shields also asked the district court to amend its judgment, a request properly brought under Rule 60. The court declined, and Shields does not appeal that denial, so we do not discuss these claims further.