Doe Ex Rel. Johnson v. South Carolina Department of Social Services

WILKINSON, Circuit Judge,

concurring in the judgment:

I agree with the affirmance of the grant of summary judgment to Thompson on all federal claims. * Maj. Op. at 179. I agree also with the vacatur of summary judgment and remand on the state claim, with explicit permission for the district court to further remand those claims back to state court. Maj. Op. at 179.

State court is the place where claims such as these should be heard. We should not further constitutionalize this intensely domestic area of the law. Every tort committed in this country by some public actor *180 was never meant to be a matter of constitutional import. See Paul v. Davis, 424 U.S. 693, 701 , 96 S.Ct. 1155 , 47 L.Ed.2d 405 (1976).

I disagree with the view that the Constitution imposes upon states a “duty to protect” children from harm by non-state actors in foster homes. Maj. Op. at 175-76. Such a holding is appealing in rhetoric and harmful in practice. The majority’s rule constitutes an unnecessary expansion of substantive due process that, however well-intentioned, will visit real misfortune upon abused and neglected children. The creation of this federal cause of action is not merely a misadventure in constitutional theory but a real-life loss for those most in need of help.

I.

Faced with the wrenching allegations of this case, it is understandably tempting to rush to Jane’s rescue and somehow right whatever wrongs were perpetrated against her. I certainly share the majority’s sympathy for any child subject to sibling sexual or physical abuse. However, I fear that the legal rule established, ironically and tragically, will have the exact opposite effect of that which is intended. Instead of helping children escape the desperate environment of abuse and neglect, it likely will hinder them in doing so. It will create a perverse set of incentives that will deter states and foster families from providing helpless children with the assistance they need.

More specifically, the new substantive due process right fashioned by the majority will lead to a combination of two separate but equally devastating effects: first, it will discourage states from taking legal custody of children in the first place, and second, it will discourage potential foster parents from becoming foster parents. Both outcomes disadvantage the countless children beaten, molested, or abandoned by their biological families — children who desperately need the safe haven provided by state custody and offered by willing foster families. In short, the “duty to protect” may well discourage protection and, ultimately, encourage harm.

A.

The first potential result of today’s decision is to dissuade states from assuming legal custody of children altogether. As DeShaney made clear, and as the majority admits, if a state avoids taking custody of a child, it cannot be held legally responsible for harm suffered by the child, even if the state knew about the harm and stood by to wratch. DeShaney v. Winnebago County Dep’t of Soc. Servs., 489 U.S. 189, 201 , 109 S.Ct. 998 , 103 L.Ed.2d 249 (1989); Maj. Op. at 171-72. A “duty to protect” arises, if at all, “only if the state takes an individual into custody; if there is no custodial relationship, then the state has no duty to protect.” Patten v. Nichols, 274 F.3d 829, 841 (4th Cir.2001).

By contrast, if a state takes legal custody of a child, it now potentially faces steep liability. Before this decision, a state taking custody of a child from her biological parents, against the parents’ wishes, already risked one form of litigation: a section 1983 suit by the biological parents for a violation of their substantive due process rights to “retain custody over and care for their children, and to rear their children as they deem appropriate.” Jordan ex rel. Jordan v. Jackson, 15 F.3d 333, 342 (4th Cir.1994) (citations omitted). Now, after this decision, a state taking custody of a child from her biological parents, against the parents’ wishes, is subject to a second form of litigation: a section 1983 suit by the child herself, perhaps brought many years after the initiation of custody, for a violation of the child’s substantive due pro *181 cess rights to “safety and general well-being.” Maj. Op. at 175 (quoting DeShaney, 489 U.S. at 200 , 109 S.Ct. 998 ). By piling on this additional cause of action, the majority has littered the path to protecting children with section 1983 landmines.

Thus, in deciding whether to assume custody of a child, states are faced with two possible choices: allow the child to continue to suffer, with no accompanying risk of liability, or rescue the child and quite possibly defend against two lawsuits. It doesn’t take a Law and Economics scholar to figure out how these prospects will affect rational state actors. The majority’s “duty to protect” essentially rewards states for the exact opposite: for not protecting and for doing nothing. As we have previously recognized, “[i]f section 1983 liability attaches too readily to removal and placement decisions, the course of public agencies would invariably become one of inaction, thus leaving children in abusive environments.” White ex rel. White v. Chambliss, 112 F.3d 731, 736 (4th Cir.1997). The danger of deterrence in this context is particularly acute. In the case of law enforcement, any deterrent effect caused by section 1983 suits is often mitigated by the inevitable public outcry over crime gone unaddressed. In the case of abused and neglected children, however, public outrage is often absent, since abuse occurring to the defenseless behind closed doors is typically invisible to public view.

We never learn, of course, precisely when the threat of liability produces inaction. The failure of agencies to take action never hits the judicial radar in quite the same manner as actions allegedly gone awry. But stacking the legal deck in the direction of inaction surely carries costs of its own. The skewing of incentives toward inaction can only harm those children who are badly mistreated by their biological families and who badly long to escape their hapless home surroundings. I need not detail the physical violence, sexual abuse, substance addiction, or outright abandonment and neglect that befall young children at the hands of those who should love them most. These sad stories are familiar to us all. See, e.g., DeShaney, 489 U.S. at 191-93 , 109 S.Ct. 998 .

For these children, the threat comes not from the state’s assumption of custody but from the state’s refusal to assume custody. A social worker may sometimes seem about as popular as a tax collector, but sometimes that same social worker is a child’s only lifeline, her sole hope of escaping a perfectly brutal and life-scarring state. By slamming social workers with multiple prospects for substantive due process suits, the majority’s opinion discourages them from exercising custody and encourages them to leave children “defenseless in the face of physical abuse and brutality.” White, 112 F.3d at 736 .

This ease illustrates the perils of a rule that discourages states from taking custody of children. Whatever SCDSS might or might not have done wrong once it removed Jane from her home, no one contends that SCDSS should have left her there. Jane’s biological mother neglected her, as the South Carolina Family Court determined after a probable cause hearing, and allegedly sexually abused her; Jane’s biological father was incarcerated; and moreover, Jane’s brother allegedly sexually abused her in their biological home even before he allegedly sexually abused her in foster homes. Thus, while it is all too easy to shovel opprobrium on SCDSS for its alleged shortcomings and make it the scapegoat for a terribly sad set of circumstances, it is worth remembering that the agency did not create the dangers of abuse or neglect. Those dangers were present long before SCDSS came on the scene, and ameliorating them is often no easy or sim *182 pie task. But leaving the child to her fate is not the answer, and I respectfully protest the creation of multiple causes of action that encourage agencies to do just that.

B.

The majority’s approach is likely to have a second and similarly unfortunate effect: causing some who would otherwise choose to become foster parents to abandon their plans. If and when states do decide to assume legal custody, they will now need to be more on guard than ever against section 1983 suits, and in an effort to fulfill their new “duty to protect,” states may become overly intrusive into the foster family unit. To be sure, some level of supervision is necessary. But to ensure that foster parents are not even unwittingly laying the basis for a later section 1983 action, states may go overboard, checking in on foster children in frequent, unexpected, inconvenient, and invasive visits to foster homes. States may subject foster parents to an onslaught of prying questions about their private lives. Further, lest they later face accusations of a constitutional nature, years after the fact, states may feel compelled to micro-manage the foster parenting process, dictating basic choices about diet, hygiene, education, activities, schedule, and interactions with peers — choices that are normally left to parents.

This is all contrary to the basic purpose of the Due Process Clause, which “was to protect the people from the State, not to ensure that the State protected them from each other.” DeShaney, 489 U.S. at 196 , 109 S.Ct. 998 . Instead of protecting foster children from the state, the majority uses the Due Process Clause to require states to protect foster children from other foster children, including, as here, their own biological siblings. Maj. Op. at 175-76. Leaving aside the fact that managing sibling relationships is among the foremost of parental duties, this rule might, at least on the margins, induce a state into deliberately placing siblings in different foster homes, just to be safe. Of course, there are undoubtedly cases in which we might want states to separate siblings, and if the facts alleged in this case turn out to be true, this case might be one of them. However, there are undoubtedly many other cases in which we would not want to skew legal incentives toward separating siblings. Children removed from their biological homes have already endured one significant dislocation, and we should not maximize the disruption of what remains of their family by placing law on the side of pulling them apart from their brothers and sisters.

For SCDSS, the only way to forestall a suit over whether an initial foster placement was good or bad is to continually monitor its consequences. But who would want to remain a foster parent under such a regime? The increased state intrusiveness encouraged by the majority’s decision will again not work to the advantage of children. Although such heightened oversight may ferret out abuse by foster parents in a few cases, it is just as likely, and probably more so, to deter many fíne individuals from ever agreeing to become foster parents in the first place. Although foster parents and biological parents do not share identical rights, they share similar purposes. As the Supreme Court has noted:

“[T]he importance of the familial relationship, to the individuals involved and to the society, stems from the emotional attachments that derive from the intimacy of daily association, and from the role it plays in ‘promot(ing) a way of life’ through the instruction of children.... No one would seriously dispute that a *183 deeply loving and interdependent relationship between an adult and a child in his or her care may exist even in the absence of blood relationship. At least [in certain cases], it is natural that the foster family should hold the same place in the emotional life of the foster child, and fulfill the same socializing functions, as a natural family.”

Smith v. Org. of Foster Families for Equality & Reform, 431 U.S. 816, 844 , 97 S.Ct. 2094 , 53 L.Ed.2d 14 (1977) (citations omitted). To be effective, foster parents need some freedom in deciding how to raise their children and need to bond with their children away from the state’s watchful eye. If agencies fearful of federal “duty to protect” suits are constantly making suspicion-laden inquiries and trying to direct decisions best made by parents, the satisfactions of foster parenting cannot be enhanced. And by making it less appealing to become a foster parent, fewer individuals will volunteer. Without willing foster families, children who are abused by their biological families will be increasingly deprived of a substitute home in which to find sanctuary.

The majority’s analysis further works to put off potential foster parents by painting foster homes in an unfortunate light. The majority concludes that a state’s removal of a child from her biological home is an exercise of custody sufficient to trigger the “duty to protect,” insofar as it deprives the child of liberty in a manner “analogous to that recognized in Estelle for prisoners and in Youngberg for the involuntarily committed.” Maj. Op. at 172. Indeed, the majority believes that the state’s custodial relationship continues even after the state has placed a child in the care of foster families. Maj. Op. at 172.

Any comparison of the foster home environment to that of prisons and mental health institutions for substantive due process purposes is misplaced. A foster home is not a prison; a child therein is not “in custody” in the sense that inmates are; and such an analogy is a poor foundation for an expansion of substantive due process. Foster parents, who are rarely state actors, see Milburn v. Anne Arundel County Dep’t of Soc. Servs., 871 F.2d 474, 479 (4th Cir.1989), would hardly find it flattering to learn that their efforts at raising children were being analogized by courts to those of prison wardens and mental health superintendents. While wardens and superintendents provide valuable functions, most foster parents approach their responsibilities in a different spirit. By extending custody to the case at bar, the majority has drifted far afield from the context of physical confinement envisioned by Estelle and Youngberg. Those cases use words such as “incarceration” and “institutionalization” to describe actionable physical restraints, see DeShaney, 489 U.S. at 200 , 109 S.Ct. 998 , terms which are singularly inapposite to the foster home setting. In expanding the concept of custody in this fashion, the majority has misapprehended the nature of foster parents’ services, all the while risking a variety of untoward effects.

The majority attempts to lull us into a false sense of security with its ruling that the state must have acted with “deliberate indifference” in its placement decisions in order for section 1983 liability to attach. Maj. Op. at 175-76. Under this standard, the majority reassures us, social workers will not be liable for “unknown harm or dangers” or every time a child suffers “incidental injuries or infrequent acts of abuse.” Maj. Op. at 175 (citations and internal quotations omitted). Therefore, in theory, states should fear liability only in those cases where they have acted in a particularly blameworthy manner.

*184 But these words are of little comfort. Standards of review are regrettably all too often in the eye-of-the-beholder. And words such as “incidental,” “infrequent,” and even “unknown” (which slips into “should have known”) are by their nature elastic and at best imperfect protection against even groundless litigation. Thus, in practice, states will be deterred both from taking custody of abused children and encouraged to intrude upon foster families to the detriment of that unit, not because such decisions were dictated by their merits, but because of a threat of future liability. The mere threat of litigation is enough to give officials pause, because even when state workers are ultimately absolved, litigation takes a toll. By diverting government officials from their duties, litigation “exacts heavy costs in terms of efficiency and expenditure of valuable time and resources,” Ashcroft v. Iqbal, — U.S. -, 129 S.Ct. 1937, 1953 , 173 L.Ed.2d 868 (2009), and potentially even “deter[s] ... able citizens from acceptance of public office,” or in this case social work. Harlow v. Fitzgerald, 457 U.S. 800, 814 , 102 S.Ct. 2727 , 73 L.Ed.2d 396 (1982).

To say the Constitution of the United States allows us to court these consequences, or even to flirt with the unpredictability of such serious harms, betrays an inflated conception of the federal judicial role. It is the. majority’s mistaken view of the boundaries of our function that I shall address in the next section and with which I also respectfully take issue.

II.

Law exists in part to guard against the overreaching of public authority, and from that general purpose the life-tenured federal courts are not exempt. When the many cautionary maxims of restraint are toppled like dominos, the chances of judicial miscalculation exponentially increase. The majority’s discovery of a constitutional “duty to protect” is badly misguided. But even if it were not — even if it were a great idea- — -we should not indulge in its creation. Federal courts simply do not have a roving warrant to adopt whatever policies they believe to be beneficial, all in the name of substantive due process. The project of creating such a duty should be left to the states, via statutes or the common law of tort. It does not belong to the federal government, via the Constitution. In creating this new “duty to protect,” the majority commandeers a role traditionally— and for good reason- — entrusted to state legislatures.

A.

To begin, by federalizing the “duty to protect,” the majority trespasses on the most traditional of state roles. The core legal question in this case- — -who may be held liable for physical and psychological harm inflicted by one foster child upon another foster child — is essentially one of tort and domestic relations law. Both legal realms have historically and justifiably been reserved to the states. Decisions in these areas require weighing a number of competing policy considerations. Deciding whether and to what extent to impose a “duty to protect,” for example, demands a delicate balancing of the interests of children, foster parents, biological parents, and states. Others have devoted far more thought to these hard problems than we ever could. Given their experience with, and acquired expertise over, matters of tort and family law, states are better suited than the federal government to make such intricate determinations. As the Supreme Court explained, the “host of policy choices” implicated in cases such as this one “must be made by locally elected representatives, rather than by federal judges *185 interpreting the basic charter of Government for the entire country.” Collins v. City of Harker Heights, 503 U.S. 115, 129 , 112 S.Ct. 1061 , 117 L.Ed.2d 261 (1992).

Indeed, most states have made such policy choices, enacting statutes addressing whether and when to provide recourse to children injured while in state custody or in the care of foster parents. South Carolina is one such state. See S.C.Code Ann. § 15-78-70. I run the risk of length in describing its efforts, but that is only to contrast it with the brief and episodic nature of the majority’s federal judicial foray. The South Carolina Tort Claims Act (SCTCA) is a comprehensive statutory scheme, which covers a breadth of situations in which an individual might be harmed by a state agency. S.C.Code Ann. § 15-78-10 et seq. It exhaustively delineates the circumstances in which a state agency, such as the South Carolina Department of Social Services (SCDSS), see Joubert v. S.C Dep’t of Soc. Servs., 341 S.C. 176 , 534 S.E.2d 1 (S.C.App.2000), may be held liable for torts. Not surprisingly, plaintiffs brought suit under the SCTCA as well as under 42 U.S.C. § 1983 .

South Carolina’s statute begins with a baseline rule that state agencies may be liable for torts to the same extent as private individuals. S.C.Code Ann. § 15-78-40. It then creates certain exceptions to the general rule, providing immunity in instances where an agency employee exercises discretion. S.C.Code Ann. § 15-78-60(5). The immunity, however, does not attach where the agency employee is guilty of “gross negligen[ee].” S.C.Code Ann. § 15-78-60(25). The SCTCA then fills in missing pieces with specific provisions addressing, for example, the statute of limitations, S.C.Code Ann. § 15-78-110, appropriate remedies and damages caps, S.C.Code Ann. § 15-78-120, and implications of state liability insurance. S.C.Code Ann. § 15-78-140, -150, -160.

When the South Carolina legislature has not spoken, the South Carolina courts have stepped up interstitially. See, e.g., City of Hartsville v. S.C. Mun. Ins. & Risk Financing Fund, 382 S.C. 535 , 677 S.E.2d 574 (2009) (discussing proper burden of proof under the SCTCA); Faile v. S.C. Dep’t of Juvenile Justice, 350 S.C. 315 , 566 S.E.2d 536 (2002) (discussing meaning of “gross negligen[ce]” under SCTCA); Jensen v. Anderson County Dep’t of Soc. Servs., 304 S.C. 195 , 403 S.E.2d 615 (1991) (discussing whether SCDSS’s decision regarding whether to investigate a report of child abuse was “discretionary”); Varn v. S.C. Dep’t of Highways & Public Transp., 311 S.C. 349 , 428 S.E.2d 895 (S.C.App. 1993) (per curiam) (discussing whether a court can award costs of litigation under the SCTCA). Moreover, SCDSS has further established an extensive set of policies that touch every aspect of a child’s care, developing numerous manuals to steer its many decisions affecting children, biological families, foster families, and adoptive families. See SCDSS, Manuals, http://dss. sc.gov/content/library/manuals/index/aspx. Its manual on “Foster Care,” for example, is 180 pages of detailed guidelines regarding such matters as intake procedures, foster children with special needs and health concerns, appropriate levels of ongoing supervision of children in foster care, and decisions regarding whether and when to remove children from foster care. See id.

All in all, the scheme is quite a comprehensive one, making all the more puzzling the majority’s intervention in this area. As the majority acknowledges, the South Carolina legislature intended its statute to be “the exclusive remedy for any tort committed by an employee of a governmental entity.” S.C.Code Ann. § 15 — 78—70(a); *186 Maj. Op. at 178. Yet the majority nonetheless provides an additional, federal remedy for torts committed by employees of state government entities. Maj. Op. at 172. The majority does so without ever telling us what deficiency in South Carolina law or practice has led it to intervene with a cause of action all its own. In fact, the majority says that SCDSS employees may be liable when they act with “deliberate indifference” to foster care placements. Maj. Op. at 172. This standard seems a rough approximation of the “gross negligen[ce]” standard, and South Carolina has already acted to deny immunity in cases where agency employees are grossly negligent. S.C.Code Ann. § 15-78-60(25).

Thus, the sole purpose of the majority’s cause of action, it would seem, is not to correct any identified deficiencies in state law but simply to allow plaintiffs access to federal courts, where federal judges may fashion over time their separate sets of policies which may converge or diverge from South Carolina’s own legal scheme. Moreover, the basis for the intrusion is none other than substantive due process, which unlike its procedural cousin, limits the range of options available not only to South Carolina but to every state in this circuit in balancing the claims of families, children, and those state agencies charged with protecting their welfare.

To make matters worse, the majority’s intrusion is undertaken pursuant to Pearson v. Callahan, — U.S. -, 129 S.Ct. 808 , 172 L.Ed.2d 565 (2009), which modified the Supreme Court’s earlier pronouncements in Saucier v. Katz, 533 U.S. 194 , 121 S.Ct. 2151 , 150 L.Ed.2d 272 (2001), to make unnecessary a federal declaration of rights in cases, such as this one, where qualified immunity is available to state actors and where the declaration can have no conceivable effect on the outcome of the case or controversy. Among its reasons for modifying Saucier, the Supreme Court noted that gratuitous proclamations of constitutional rights “may create a risk of bad decisionmaking” and “depart[ ] from the general rule of constitutional avoidance and run[ ] counter to the older, wiser judicial counsel not to pass on questions of constitutionality ... unless such adjudication is unavoidable.” Pearson, 129 S.Ct. at 820, 821 (citations and internal quotations omitted). Pearson’s cautionary signals were sent for good reason, namely to deflect advisory and ill-advised judicial ventures into policy, which, with all respect, is what we have here.

In this case, Jane may in fact be entitled to recovery under the South Carolina statute — an issue which appropriately may be litigated in state court. See Maj. Op. at 179. And perhaps Jane is not entitled to relief, either because the facts do not develop in Jane’s favor or because South Carolina may not allow recovery as a matter of law. But even if Jane cannot recover under state law — and perhaps especially if Jane cannot recover under state law— this court should not intervene to override that statutory result by allowing her to recover under federal constitutional tort. Any state court outcome may very well reflect a deliberate legislative decision regarding the relative merits of enhanced state liability versus the relative merits of permitting social workers some judgmental latitude in promoting an abused child’s welfare and best interest. We should not assume that our federal rule is ipso facto better than the state’s or that the South Carolina courts and legislature would gratefully welcome this court’s recalibration of the balance they have struck. In fact, in enacting the SCTCA, the South Carolina General Assembly expressly “recognize[d] the potential problems and hardships each governmental entity may face being subjected to unlimited and unqualified liability for its actions,” and therefore *187 decided upon a system of neither “total immunity” nor “unqualified liability.” S.C.Code Ann. § 15-78-20(a).

And if South Carolina is dissatisfied with the result under its statutory scheme, it may craft a new rule going forward. That decision, however, is the prerogative of the state, not this court. As the Supreme Court explained in DeShaney :

“The people of Wisconsin may well prefer a system of liability which would place upon the State and its officials the responsibility for failure to act in situations such as the present one. They may create such a system.... But they should not have it thrust upon them by this Court’s expansion of the Due Process Clause of the Fourteenth Amendment.”

DeShaney v. Winnebago County Dep’t of Soc. Servs., 489 U.S. 189, 203 , 109 S.Ct. 998 , 103 L.Ed.2d 249 (1989).

By federalizing the “duty to protect,” the majority has effectively scuttled one of the primary benefits of federalism: the tolerance of varied state approaches to our most stubborn social problems. See New State Ice Co. v. Liebmann, 285 U.S. 262, 286-87 , 52 S.Ct. 371 , 76 L.Ed. 747 (1932) (Brandeis, J., dissenting). Not only does the majority defeat the benefit of variation but it also incurs the cost of pile-on liability. By imposing a federal constitutional system of liability on top of a state system of liability, the majority has put into place a double whammy of deterrence coupled with a double wringer of litigation; states assuming children’s custody will now have to fear no less than two liability schemes— one federal, one state — and perhaps, as happened here initially, two separate lawsuits — one in federal court, one in state court. And as noted, this dual system of deterrence may work to the detriment of those very persons it was designed to protect — -the children whose chances in life depend on being rescued from the ravages of the most dysfunctional environments, even as multiple causes of action now counsel the rescuer to back off.

B.

While state legislatures are the traditional fora for deciding whether to impose a “duty to protect,” if the duty absolutely must be federalized, it should be federalized by Congress, not by us. Congress is the appropriate institution to create any federal “duty to protect,” since it can do so by statute rather than by Constitution, which is simply too blunt and rigid a tool for the imposition of general tort liability. If the matter is to be federalized, Congress could readily do so under its spending power, by creating rules for states and conditioning the receipt of federal funds on compliance with those rules. See South Dakota v. Dole, 483 U.S. 203, 206-07 , 107 S.Ct. 2793 , 97 L.Ed.2d 171 (1987). Congress can hold hearings and consult a wide range of public and expert opinion, thus enabling it to more accurately anticipate a proposed rule’s potential consequences and to develop a policy with sufficient nuance to work in diverse factual circumstances. None of these tools did this court have at its disposal.

Legislatures are also comparatively better positioned to monitor the impact of a legal rule, once enacted, and if need be, to adapt the rule in light of changing circumstances or new information. If a law is indeed having a negative effect, legislatures “may recognize degrees of evil and adapt [their] legislation accordingly.” See Packer Corp. v. Utah, 285 U.S. 105, 110 , 52 S.Ct. 273 , 76 L.Ed. 643 (1932). For judges, things are different. Because we are insulated from the democratic pulse, we learn of consequences, if at all, by fortuity. Although logic suggests that the majority’s opinion will have the unfortu *188 nate result of leaving more abused children in the hands of the abusers, the truth is, without any feedback mechanism in place to help us assess our rulemaking, we may never know the extent of the harm caused. And even if we could, the harm could not easily be undone- — -not by us, bound as we are by precedent, and not by a legislature, since the majority has rooted its “duty to protect” in the Constitution.

Neither the majority nor the parties point to any instance where Congress has laid down a rule to govern the conduct in this case, and it is wrong for a federal court to rush in where Congress has feared to tread. Additionally, because the textual underpinnings for the majority’s duty are unclear, the Constitution supplies this court with very little to direct our task. “[T]he Court has always been reluctant to expand the concept of substantive due process because guideposts for responsible decisionmaking in this uncharted area are scarce and open-ended. The doctrine of judicial self-restraint requires us to exercise the utmost care whenever we are asked to break new ground in this field.” Collins v. City of Harker Heights, 503 U.S. 115, 125 , 112 S.Ct. 1061 , 117 L.Ed.2d 261 (1992) (citation omitted).

By basing its newfound duty in the elusive Due Process Clause, the majority has triggered a myriad of issues that this court will eventually have to address, with virtually nothing either to constrain or to guide our task. The majority creates a new substantive due process duty but leaves the rest hanging. What exactly are a state’s affirmative obligations? What must it do to seek out information regarding a child’s circumstances? How often and how extensive must supervision be? How do limitations periods and tolling principles apply when a foster child raises claims much later in life? From whom and from what must the state protect a child? From which of life’s inevitable bumps and a family’s inevitable strains and difficulties? These questions and others have no easy or obvious answers, but by “extending] substantive due process to this area, these questions [will] be before us in short order, and it is hard to imagine what tools federal courts would use to answer them. At the end of the day, there is no reason to suppose that [our] answers to these questions would be any better than those of state courts and legislatures, and good rea son to suspect the opposite.” Dist. Atty’s Office v. Osborne, — U.S. -, 129 S.Ct. 2308, 2323 , 174 L.Ed.2d 38 (2009).

So those who must follow the scarcely formed federal rule are simply left to guess. I suspect that fear of this nebulous liability will produce many instances of excess caution in assuming legal custody of battered children and of excess intervention into new family units. And in the future, when we are forced to further define the contours of the majority’s “duty to protect,” we will be pushed further into the uncomfortable position of policy maker and will have to, bluntly, make stuff up. It is always tempting, I recognize, to respond to contentions that tug at the heartstrings, but I do not think state courts are cold or callous in a way that we are not, and I do think the states in their judicial and legislative capacities are where the very difficult tensions between familial and communal responsibilities should be resolved. Restraint in the judicial task can be the compassionate course. Allowing Jane to have her day in state court will both respect her individual claims to justice and avoid the heartache that creation of this new constitutional cause of action will involve.

Specifically, I concur in the majority's view that, at the very least, qualified immunity is appropriate here on plaintiffs' cause of action for failure to protect, Maj. Op. at 175-76, and in the majority's view that there is no federal cause of action for failure to disclose. Maj. Op. at 177.