The present controversy inhabits the twilight area of developing law concerning the constitutional rights of the involuntarily committed mentally retarded. Nicholas Romeo appeals, through his next friend, from a jury verdict for the defendants, officials of the Pennhurst State School and Hospital, in a suit brought pursuant to 42 U.S.C. § 1983 (1976). Plaintiff alleges trial errors in the admission and exclusion of evidence, in the court's instructions to the jury and in the manner in which the trial was conducted. Because of the improper exclusion of relevant expert medical testimony and critical flaws in the standards that were employed in charging the jury, we vacate the judgment of the district court and remand for a new trial.

While courts in the past decade have carefully focused on the procedural protections applicable to the initial commitment of the mentally handicapped, see Parham v. J. R., 442 U.S. 584, 99 S. Ct. 2493, 61 L. Ed. 2d 101 (1979); Addington v. Texas, 441 U.S. 418, 99 S. Ct. 1804, 60 L. Ed. 2d 323 (1979), relatively little has been resolved with respect to conditions of confinement or the extent of the state's duty to protect and to treat the institutionalized. Specifically at issue here is the judicial responsibility to enforce constitutional guarantees governing the incarceration of the institutionalized retarded.1 This, in turn, calls on us to deal with the question of what standards of proof are required in a § 1983 suit for damages, in which a mentally retarded plaintiff claims that the defendants improperly shackled him, failed to provide adequate protection for him, and did not make appropriate treatment available to him. In defining the principles relating to claims for protection and treatment of the retarded, carefully crafted instructions must be utilized that will reflect the duty of courts to safeguard the constitutional rights of those confined, but also will be sensitive to the prerogative of the medical community to exercise its professional judgment and to the undeniable fiscal and administrative concerns of the state.

Romeo is a profoundly retarded person. Although he is physically thirty years old, he suffers from a chemical imbalance of the brain that renders his mental capacity approximately that of an eighteen month old child. For the first twenty-six years of his life Romeo lived with his parents in South Philadelphia. On May 10, 1974 his father died. Within a month his mother, finding herself unable to care for him, applied to the Philadelphia Common Pleas Court for his admission to a mental retardation facility. On July 11, 1974, the court committed Romeo to the Pennhurst State School and Hospital, pursuant to the involuntary commitment provision of the Pennsylvania Mental Health & Mental Retardation Act of 1966. Pa.Stat.Ann. tit. 50, § 4406 (Purdon) (1969).

It is not contested that, while confined at Pennhurst, Romeo was injured on over seventy occasions. These injuries were both self-inflicted and the result of attacks by other residents, some in retaliation against Romeo's aggressive behavior. The injuries included a broken arm, a fractured finger, injuries to sexual organs, human bite marks, lacerations, black eyes, and scratches. Moreover, some of plaintiff's injuries became infected, either from inadequate medical attention or from contact with human excrement that the Pennhurst staff failed to clean up.

Since Romeo is incompetent, this action was brought on his behalf by his mother as next friend. The § 1983 complaint seeks damages for the described injuries from three officials at Pennhurst: C. Duane Youngberg, then superintendent, Richard Matthews, director of resident life, and Marguerite Conley, director of the plaintiff's assigned unit at the time most of the injuries occurred. There is evidence which indicates that each defendant knew of some or all of the seventy-plus injuries suffered by Romeo.

After the case was filed, the district court permitted the plaintiff to amend the complaint to include allegations that, since the initiation of the suit, defendants had kept Romeo shackled to a bed or a chair in the hospital at Pennhurst for long periods each day. The amended complaint, which posited a violation of plaintiff's constitutional right to treatment occasioned by the shackling, exposure to attacks and inappropriate treatment,2 again sought compensatory and punitive damages from the defendants.3

At the time of trial, the district court refused to permit plaintiff's two experts. Dr. Foxx and Dr. Grover, to testify about the lack of programming and activities on Romeo's ward, which they believed accounted for his numerous injuries, and about alternative methods of treatment that would have reduced the frequency of attacks.4 One of the experts would have testified further that the restraints served no medical purpose and were used solely for the convenience of the staff. The court sustained objections to all of this proffered medical and psychiatric testimony on the theory that admission of such evidence would transform a § 1983 action into a malpractice case.5 In addition, the court rejected the plaintiff's proposed jury instruction which maintained that the plaintiff had a right to treatment in the least restrictive environment. The court decided instead that defendants' shackling practices and duty to protect Romeo should be evaluated solely on the basis of an Eighth Amendment standard. Further, in what was described as an attempt to distinguish this § 1983 suit from a malpractice case, the court subjected plaintiff's claims of inadequate treatment to an Eighth Amendment standard of "deliberate indifference to the serious medical needs of the resident."

Because we believe that the Eighth Amendment which limits the scope of judicial review of conditions of incarceration for the criminally convicted to a "cruel and unusual" threshold is inappropriate in the context of civil as distinguished from criminal confinement, the result reached in the district court must be vacated. Moreover, the uncharted legal issues which have arisen and the potential ramifications of this case impel us to set out in some detail the level of judicial scrutiny that should be accorded the intricate set of legal, medical and societal interests that intersect in the situation presented here.

Of critical importance in this appeal is the recognition that this is a due process case, not a controversy to be governed by "cruel and unusual" principles. Although the complaint alleged Eighth as well as Fourteenth Amendment violations, Ingraham v. Wright6 and Bell v. Wolfish7 would appear to preclude reliance on the minimal safeguards of the Eighth Amendment in a non-criminal context. Ingraham held that the Cruel and Unusual Punishments Clause is inapplicable in the context of corporal punishment in public schools; Wolfish found Eighth Amendment scrutiny inappropriate for evaluating conditions of pretrial detention.8 Indeed, Wolfish explicitly recognizes a right of innocent persons to be free from punishment, a proposition directly grounded in the word "liberty" that appears in the Due Process Clause. See 441 U.S. at 535, 99 S.Ct. at 1872, id. at 580, 99 S.Ct. at 1896. (Stevens, J., dissenting). Thus, it is the Fourteenth Amendment's prohibition of state deprivation of life, liberty or property without due process of law that is the appropriate fulcrum of our concerns today.

It is necessary, of course, to determine initially whether the asserted individual interests are encompassed within the Fourteenth Amendment's protection of life, liberty and property. If the answer to that inquiry is in the affirmative, we must then determine what level of judicial scrutiny is applicable to the various interests in issue here.

We are fully cognizant that the propriety of a § 1983 claim, in the present situation, turns on whether a constitutional right is at issue. The concurrence is undeniably correct in requiring that we distinguish between constitutional violations and ordinary malpractice claims. But the implication that the existence vel non of a state remedy for example a malpractice action is relevant to the determination of whether a § 1983 claim exists, would appear to be unfounded. Cf. Paul v. Davis, 424 U.S. 693, 715, 96 S. Ct. 1155, 1167, 47 L. Ed. 2d 405 (1976) (Brennan, J., dissenting). Nor does the existence of treatment issues in a claim of constitutional infringement, on account of a linguistic similarity, transform the claim into a malpractice action.9 Admittedly, the plaintiff's confinement in a state institution does not dignify every complaint with constitutional stature. Cf. Estelle v. Gamble, 429 U.S. 97, 106, 97 S. Ct. 285, 292, 50 L. Ed. 2d 251 (1976). Nevertheless, there is a distinction for constitutional purposes between conduct by state actors and private citizens. Therefore, understandable concerns with stemming the federalization of common law tort actions, see Baker v. McCollan, 443 U.S. 137, 99 S. Ct. 2689, 61 L. Ed. 2d 433 (1979), must not overcome a court's duty to safeguard legitimate constitutional rights.10 As Justice Harlan realized, " 'liberty' is not a series of isolated points pricked out in terms of (the Bill of Rights) .... It is a rational continuum which, broadly speaking, includes a freedom from all substantial arbitrary impositions and purposeless restraints." Poe v. Ullman, 367 U.S. 497, 543, 81 S. Ct. 1752, 1776, 6 L. Ed. 2d 989 (1961) (dissenting opinion).

Once a protected interest is found to exist, the proper level of judicial scrutiny is ascertained by the nature of that interest. A court must not be overquick to equate the scope of the right to protection or treatment for the involuntarily confined retarded with the already articulated scope of such rights for the criminally incarcerated.11 Nor are analogies to state or common law precedents controlling it is federal law that answers the question of what process is due under the Constitution.12

The confinement of an individual to an institution for either the mentally ill or mentally retarded entails a "massive curtailment of liberty." Humphrey v. Cady, 405 U.S. 504, 509, 92 S. Ct. 1048, 1052, 31 L. Ed. 2d 394 (1972). In consequence it is circumscribed by due process protections. Addington v. Texas, 441 U.S. 418, at 425, 99 S. Ct., at 1809. And once inside the institution an individual's liberty interest is not summarily extinguished.13 Rather, those aspects of personal autonomy recognized from the time of Blackstone the power of locomotion without restraint and the right to personal security14 as well as the right to freedom from punishment, require continued respect. These fundamental liberties may be legitimately encroached upon only when justified by an overriding,15 non-punitive state interest related to the reasons for confinement.

Courts have reached a general consensus on three legitimate state justifications for the confinement of the mentally ill and mentally retarded. These rationales have evolved historically from (1) the protection of society from individuals who constitute a danger, a concept rooted in the state's police power; to (2) the protection of individuals unable to care for themselves or prone to self-destructive acts, based on police power and parens patriae rationales; and finally to (3) rehabilitation often limited to habilitation in the case of the mentally retarded also an exercise of the state's parens patriae authority.16 See Addington v. Texas, 441 U.S. 418, at 426, 99 S. Ct. 1804, 1809, 60 L. Ed. 2d 323, O'Connor v. Donaldson, 422 U.S. 563, 573-74, 95 S. Ct. 2486, 2493-94, 45 L. Ed. 2d 396 (1974).

Noticeably lacking, and logically inapplicable to the mentally retarded, are the traditional deterrence and retribution underpinnings of the criminal system.17 Obviously, the state has no right to punish an innocent individual, and the concept of deterrence has a minimal effect on a severely retarded person, such as the plaintiff here. In seeking to explain claimed infringements of fundamental liberty interests, the state is limited to protection and treatment rationales of a compelling or substantial nature. Involuntary commitment in the civil context, then, quite clearly implicates a constitutional right to treatment and protection.18 Alternatively, if a claimed post-commitment abridgement does not reach the status of a gross deprivation or squarely cut across constitutionally protected liberty interests, but only tangentially implicates such an important interest, a mutual accommodation between institutional objectives and constitutional provisions is needed. This is so since the very nature of an institution requires some limitation on the retained rights of those who are confined. Cf. Wolff v. McDonnell, 418 U.S. at 556, 94 S.Ct. at 2974. In such instances, security concerns, fiscal constraints and administrative necessities may be adduced to demonstrate that a claimed encroachment is necessary to a facility's internal operations.19

In the present case, Romeo's complaints may be appropriately conceptualized as (1) a right to be free from undue bodily restraint; (2) a right to personal security and protection; and (3) a right to adequate treatment. The first two are undiluted legal concerns, relating to protected liberty interests; as such, they are entitled to heightened judicial scrutiny. The third entails mixed questions of law and medical judgment, and thus requires a more flexible standard of judicial review and suitable deference to informed medical opinion.

The basis of Romeo's first claim, that he was unduly shackled,20 is clearly inimical to the right of an unconvicted citizen to be free from punishment.21 Even though shackling may not be punishment per se, it raises a presumption of a punitive sanction.22 It squarely collides with a traditional liberty interest in freedom from bodily restraint.23 A valid involuntary commitment ex necessitate extinguishes a retarded person's right to freedom from confinement. Nevertheless, a residuum of liberty remains which is entitled to due process protection. In Vitek v. Jones, which dealt with the transfer of a prisoner from a prison to a mental hospital, the Supreme Court found that an involuntary commitment to a mental hospital is "qualitatively different from the punishment characteristically suffered by a person convicted of crime." 445 U.S. 480, 499, 100 S. Ct. 1254, 1266, 63 L. Ed. 2d 552 (March 25, 1980). Despite pre-existing confinement, a substantial, additional loss of liberty occurred, for which the Court required observance of due process safeguards. Similarly, in the situation before us, shackling is not normally within the range of conditions of confinement contemplated in habilitative institutionalization.24 Neither in Vitek nor in the case at hand do the asserted privations inhere in the original rationales for confinement. Because of the fundamental right at issue here, as well as the substantial risk of error25 and the possibility of significant harm, the proper judicial posture is one of careful scrutiny.26

Plaintiff requested that his shackling claim be analyzed under the rubric of a "right to treatment under the least restrictive conditions consistent with the purpose of the commitment." In this regard, he submitted the following charge:

Plaintiff contends that he was shackled to his bed or chair for long periods of time each day after he filed this lawsuit. If you find that he could have been provided treatment under less restrictive conditions than those imposed on him, then you must find that his constitutional rights were violated.

The court, however, gave the following instruction:

In order to prove his case the plaintiff must show acts or omissions sufficiently harmful to evidence a deliberate indifference to the serious medical needs of the resident. It is only such indifference that can offend the standards of decency required by the 8th Amendment. App. 2-236a 2-237a.

We have already recorded our disapproval of the adoption of the Eighth Amendment standard in the mental retardation area. However, we cannot assent as the concurrence suggests, to an analysis which scrutinizes shackling and a right to treatment by a standard that is essentially the same. The Pennsylvania statute generally prohibits such shackling practices, see fn. 21, and similarly, mental retardation professionals have relegated the use of physical restraints to the closets of an earlier age. Logically, Romeo's shackling claim centers on a liberty interest: a right to be free from bodily restraint.

The trial judge, therefore, should instruct the jury that such shackling may be justified only by a compelling necessity, i. e., that the shackling was essential to protect the patient or to treat him. It should be explained that, except in emergency situations, inadequate resources or administrative rationales offer an insufficient basis for intrusions of this kind on a fundamental liberty interest. As the courts have repeatedly proclaimed, "Humane considerations and constitutional requirements are not, in this day, to be measured or limited by dollar considerations." Jackson v. Bishop, 404 F.2d 571, 580 (8th Cir. 1968) (Blackmun, J.), quoted in Rozecki v. Gaughan, 459 F.2d 6, 8 (1st Cir. 1972) quoted in Wyatt v. Aderholt, 503 F.2d 1305, 1315 (5th Cir. 1974). Plaintiff is also entitled to a charge, in the alternative, that to absolve the defendants from liability on the shackling claim it would be necessary to find that shackling was the least restrictive method of dealing with the patient,27 in light of his problems and the surrounding environment. A "least restrictive" charge will not only insure that compelling treatment explanations, as opposed to fiscal concerns or staff convenience, were the basis for the shackling, but also that the institution considered and rejected alternative methods of restraining the resident, if some restraint indeed was required.28

A comparison with the result that might be obtained by employing the unitary standard proposed by the concurrence is apposite here. Under the standard advanced by the concurrence, the jury would be charged that shackling is permissible so long as there was not "such a substantial departure from accepted professional judgment, practice, or standards as to demonstrate that the defendants did not base their conduct on a professional judgment." Arguably, such an instruction assumes that physical restraint for the convenience of the staff would constitute a substantial departure from accepted professional judgment. Yet it is not logically evident that concerns of staff convenience are the sort of departures from medical judgment that rise to the level of being a sham or otherwise illegitimate, as defined in the concurring opinion. More fundamentally, although the standard proposed by the concurrence would probably prevent use of shackling as punishment,29 it would not preclude its use as a substitute for more effective treatment programs. Thus such a standard would fail to give adequate weight to the resident's substantial interest in freedom from bodily restraint. In addition, it would conflict with the thrust of the state's interest as proclaimed in § 4422 of the Mental Health and Mental Retardation Act, in prohibiting the use of mechanical restraints except in limited situations.

Our holding that the district court erred with respect to the jury instruction for the shackling claim, and our establishment of a compelling necessity standard for review of such troubling interferences with bodily freedom, necessarily require an additional finding of error in the district court's exclusion of relevant expert testimony. Under the compelling necessity or least restrictive standards of proof, both sides are, of course, permitted to adduce evidence. Although defendants must justify the shackling as the least restrictive means of handling or protecting the individual, the plaintiff may produce evidence, for example, that restraints were utilized largely because of convenience to the staff. Consequently, Dr. Foxx's proffered testimony regarding the inappropriate reasons and counterproductive results connected with the restraint of Romeo, was improperly excluded by the trial court.

From Colonial times to the present day, the concept of liberty has embraced the "right to be free from and to obtain judicial relief for unjustified intrusions on personal security." Ingraham v. Wright, 430 U.S. 651, 673, 97 S. Ct. 1401, 1413, 51 L. Ed. 2d 711 (1977). Plaintiff's second claim, the right to protection from attack, undeniably falls within the compass of this right. The record discloses that while confined at Pennhurst Romeo was injured on over seventy occasions. Some of these injuries were self-inflicted, some resulted from attacks by fellow residents, and some may even have been by the staff. The evidence also suggests that the defendants knew, or had reason to know, of some or all of the injuries suffered by the plaintiff. It would be anomalous to find that the right to a secure environment, which federal courts have often intervened to protect in the context of penal institutions, did not extend to facilities for the mentally retarded.30

The scope and nature of the right of the plaintiff to protection from attack is also capable of judicial demarcation. Both the individual's right to personal security and the state's interest in providing care converge to support a right to protection from attack. The state cannot simply confine a person, without more, because he may be a danger to himself or to third parties.31 Only after the commission of a discrete criminal act and conviction in accordance with procedural protections is incarceration without more warranted. Consequently, in order to survive constitutional scrutiny, involuntary commitment of a mentally retarded person must combine an interest in insulating society from dangerous behavior with a promise, either express or by implication, to care for and to treat the individual.32 Alternatively, the commitment of a retarded person for treatment purposes or because he is unable to care for himself, necessarily entails the provision of care and protection. The parens patriae power to provide care and protection is thus inextricably bound up in the involuntary commitment of the retarded regardless of the original rationale for the confinement.

Further, in Romeo's case, commitment was pursuant to Penn.Stat.Ann. tit. 50, Mental Health and Mental Retardation Act of 1966. The state, in confining Romeo, represented that it was "willing and able" to care for him.33 The statute provides an entitlement to protection34 which, like other state-created entitlements, is guarded by constitutionally acceptable procedures. The right to protection is not activated by an isolated mishap, or called into question by each bruise that a patient may suffer. But a pattern of attacks, injuries or violent behavior such as we have here would create a claim to such a right. While no one is guaranteed an injury-proof life, Romeo, as an involuntarily committed resident of Pennhurst, had a right to humane care and protection, bottomed both on the Pennsylvania statute and the Constitution.

Institutions for the mentally retarded are rarely "open" facilities. Those involuntarily confined are not free to return home, and indeed, many are bereft of any support by family and friends.35 This absence of openness or significant community surveillance and oversight underscores the need for the courts to discharge their traditional function of safeguarding constitutional rights.36

Therefore, with respect to the protection claim, we conclude that the trial court erred in its charge. It instructed the jury that:

Under the 14th ... Amendment, state officials at a state mental hospital have a duty to protect involuntarily committed residents from repeated attacks by other patients and staff.

Plaintiff ... contends that these defendants violated such a duty to protect him because they were aware of such attacks and failed to take such reasonable steps as (were) required to protect him.

If you find that the defendants were deliberately indifferent to the medical and psychological needs of (the plaintiff), then you may find that plaintiff's ... 14th Amendment rights were violated.

To find for the plaintiff you need not find that the defendants personally participated in any attack upon the plaintiff. If you find that the defendants were aware of repeated attacks upon plaintiff and failed within their sphere of authority to take reasonable steps to protect the plaintiff, then you may find that the defendants are liable for a violation of the plaintiff's constitutional rights. (emphasis supplied)

As we see it, the jury should be informed that the plaintiff has a right to have his physical safety protected. The plaintiff sought an instruction that defendants had an obligation to "take reasonable steps to protect plaintiff." Chief Judge Seitz maintains that such a charge is virtually indistinguishable from a state malpractice standard. Infra at 177. However, analysis of this particular claim for constitutional purposes would proceed as follows: because this is a fundamental interest which conflates plaintiff's right to personal security with defendants' duty to protect if the defendants failed to provide for Romeo's personal security, such failure can be justified, in a § 1983 case, only by substantial necessity. Substantial necessity is more appropriate than the compelling necessity standard employed in connection with the shackling claim, for it enables a court and jury to distinguish between isolated incidents and inadvertent accidents, on the one hand, and persistent disregard of patients' needs, on the other.37 If the defendants disregarded plaintiff's injuries or failed to take steps to protect plaintiff then they should be liable unless they can offer explanations based on important state interests.38 However, the least restrictive charge, which is applicable to the shackling claim, is a less meaningful analytic tool in the protection setting because of the very existence of a right to personal security and a duty on the part of the state to protect. We cannot assume that all patients desire and are capable of dealing with the freedoms provided by least restrictive treatment. Moreover, inherent inconsistencies arise in the use of such a concept in the protection area. The least restrictive treatment, which might be the least confining, might provide insufficient protection and care for a particular resident's wants and needs.

In view of the stance we have adopted regarding the plaintiff's right to personal security and protection from attack, we hold that it was error for the district court to exclude the testimony of Dr. Foxx and Dr. Grover concerning the availability of treatment programs which would minimize the aggressive behavior and attacks that were unnecessarily prevalent at Pennhurst. Such evidence is relevant in enabling the jury to resolve whether the defendants were properly attending to, or overlooking, the right of the plaintiff to protection from attack.39

Plaintiff's third complaint turns essentially on a claim of inadequate treatment and a right to treatment in the least restrictive manner. As previously observed, unlike a right to freedom from bodily restraint or to personal security, treatment does not present a purely legal issue. See p. 159 supra. Questions of a court's relative competence concerning medical decisions and respect for medical judgment, as well as concerns of comity with the legislative and executive branches, place this claim in a different perspective. The move from preventing degrading impositions on human dignity, such as shackling and exposing a resident to continuous attacks, to requiring treatment consonant with individual needs, transfers the court to a subjective realm of decision making. Nevertheless, appropriate deference to medical expertise does not diminish the judicial duty to safeguard liberty interests implicated in treatment decisions.

The right to treatment or habilitation in the case of the mentally retarded40 arises when an individual is involuntarily committed, regardless of whether parens patriae or police power grounds provide the major premise for the confinement. See pp. 158, 162 supra. A court performs two functions with respect to such a right: it requires that treatment be provided to those who require and are willing to accept it, and it places limits on the state's power to impose such treatment on those who do not desire it. A right to treatment does not create a corresponding duty to submit to any treatment whatsoever; such a simple equation would sanction unacceptable invasions of personal autonomy.41

At one end of the spectrum, an absolute failure to treat, when treatment is the reason for commitment, raises clear and serious constitutional problems. As the Court noted in O'Connor v. Donaldson, 422 U.S. 563, 95 S. Ct. 2486, 45 L. Ed. 2d 396 (1974), "where 'treatment' is the sole asserted ground for depriving a person of liberty, it is plainly unacceptable to suggest that courts are powerless to determine whether the asserted ground is present." Id. at 574 n.10, 95 S. Ct. at 2493 n.10. In such instances, it is the court's duty to insist on appropriate treatment or that the plaintiff be released.42 At the other end of the spectrum is the judicial role in dealing with involuntary treatment which occasions severe intrusions on individual dignity. Nonreversible physical operations, such as a vasectomy, or permanent physical alterations by surgical intervention, such as a prefrontal lobotomy, or the administration of powerful antipsychotic drugs may well constitute fundamental liberty violations.43 Like shackling, it may be that they should be subject to close court scrutiny.44 Similarly, the reasons which commend the use of least restrictive analysis with respect to shackling existence of a judicially assessable, discrete act, which entails a significant loss of individual liberty may be present in the context of such invasive, involuntary procedures. Since some of these procedures may not be present in the case at hand after it is remanded, we do not address the subject in any detail except to note the possibility of its existence. But, whenever unalterable interferences with bodily integrity place deprivations of liberty in issue, the law and not medicine is the ultimate decision-maker.

Once analysis transcends the groundlevel question of treatment vel non, however, and settles in the middle ground, legal questions of unconstitutional liberty deprivations begin to merge with medical judgments regarding the appropriateness of the regimen. Inasmuch as we are remanding, inter alia, because the district court's improper employment of Eighth Amendment scrutiny resulted in the exclusion of the plaintiff's proffered medical testimony, it is uncertain how the record will develop at trial. However the summaries of medical testimony, offered as proof out of the hearing of the jury, suggest that the trial court may need guidance in a range of situations, involving problems of both intrusive and inadequate treatment. We therefore find it in order to delineate the legal standards and jury charges applicable in the treatment realm.

Where the issue turns on which of two or more major treatment approaches is to be adopted, a "least intrusive" analysis may well be appropriate. If the staff decides upon a course of treatment which offers some promise of effectiveness without undue risk to the patient, such decision may well come within the ambit of least intrusive means. If the likelihood of serious side effects does exist, however, and substantial improvement is not reasonably expectable, then such judgments might well present a basis for finding that the course of treatment was not the least intrusive under the circumstances.45 In making constitutional judgments in such situations, the court must necessarily depend in large measure on medical evidence. Once the least intrusive regime has been selected, therefore, the application of a constitutional standard of "least intrusive alternative" to continuing treatment programs, which often involve qualitative medical determinations subject to daily, possibly hourly changes, would prove unworkable.46 The judiciary is not in a particularly advantageous position to determine which of two medications is less intrusive, nor especially competent in assessing present therapeutic benefits versus long-term consequences and side effects for each administration of a drug. Further, least restrictive assumes a direct correlation between restrictiveness and effectiveness, a presumption which can not be medically corroborated in every instance.47

Unfortunately, in attempting to craft the appropriate standard to apply to a charge of inadequacy of treatment the courts find themselves in a Scylla and Charybdis situation. On the one hand, the Supreme Court has cautioned that,

Although we acknowledge the fallibility of medical and psychiatric diagnosis, (citation omitted) we do not accept the notion that the shortcomings of specialists can always be avoided by shifting the decision from a trained specialist using the traditional tools of medical science to an untrained judge or administrative hearing officer.... the supposed protections of an adversary proceeding to determine the appropriateness of medical decisions for the commitment and treatment of mental and emotional illness may well be more illusory than real.

Parham v. J.R., 442 U.S. 584, 609, 99 S. Ct. 2493, (1979). On the other hand, the Supreme Court has noted that "The medical nature of the inquiry, however, does not justify dispensing with due process requirements." Vitek v. Jones, 445 U.S. 480, 499, 100 S. Ct. 1254, 1266, 63 L. Ed. 2d 552 (March 25, 1980), citing Addington v. Texas, 441 U.S. at 430, 99 S.Ct. at 1811. Although both these statements refer to initial commitment procedures for the mentally ill,48 their logic appears equally applicable to judicial review of treatment decisions within institutions for the mentally retarded.

In the present case Romeo was confined pursuant to § 406 of the Pennsylvania Mental Health & Mental Retardation Act of 1966, which provides that commitment is for "care and treatment."49 By basing Romeo's deprivation of liberty at least partially upon a promise of treatment, the state ineluctably has committed the community's resources to providing minimal treatment. While Romeo has an entitlement to some treatment,50 the evidence in the record, although somewhat contradictory, suggests not so much a total failure to treat as an inadequacy of treatment.51

Given this situation it would be unrealistic for us to ignore that courts by and large are not in as advantageous a position as the personnel at applicable institutions to make decisions relating to day-to-day or hour-to-hour treatment.52 Moreover, the evaluation of standards of adequacy and suitability in the psychiatric field is extremely difficult even for trained personnel, and certainly in the judicial context. Questions in this field, as the Court in Addington noted, often require not only a resolution of complicated factual issues, they turn on the very meaning of those facts a meaning that must be interpreted by expert psychiatrists and psychologists. 441 U.S. at 429, 99 S.Ct. at 1811.53 And unquestionably, the gap between available resources and those needed for ideal treatment programs is, unfortunately, considerable. Once minimum treatment needs are satisfied, courts should be cautious in requiring legislatures to allocate resources in a way that may well be more desirable, but not constitutionally mandated.54

Consequently, with respect to the claim regarding adequacy of treatment, we agree neither with the plaintiff, that he has a right, insofar as day-to-day decisions are concerned, to judicial review based on a constitutional standard of least intrusive,55 nor with the defendants that only deliberate indifference to the medical and psychological needs of Romeo constituted a violation of his right to treatment.56 It should be made clear to the jury that for the plaintiff to prevail it is necessary to find that an individual involuntarily confined in a facility for the mentally retarded did not receive a form of treatment that is regarded as acceptable for him in light of present medical or other scientific knowledge.57 The state should be permitted to adduce evidence of security concerns, administrative necessities and fiscal constraints in the course of explaining its adoption of the challenged program. Thus, if defendants can demonstrate a coherent relationship between a particular treatment program and a resident's needs, they would not be liable.58

The standard articulated here may appear, at first blush, to differ minimally from that propounded by the concurrence. However, the concurring opinion's concern with distinguishing the constitutional prohibition against a "substantial departure from accepted professional judgment" from a malpractice standard might lead to an approval of all conduct that is "not a sham or otherwise illegitimate." Infra at 181. We do not believe that such a test can adequately protect a retarded person's acknowledged constitutional right to treatment or habilitation.59 Nor do we believe that the judiciary lacks the tools to ascertain whether the constitutional minimum of medical acceptability proposed here is being fulfilled. While earlier courts struggled to weigh conflicting opinions and contending medical theories regarding severely retarded persons, Congress has recently facilitated a court's ability to determine whether a coherent relationship between a treatment program and a resident's needs exists. The enactment of the Developmentally Disabled Assistance and Bill of Rights Act of 1975 provides guidelines, albeit not constitutionally binding, which enhance rational court review.

Should the plaintiff's constitutional claims fail, the recent case of State of Maine, et al. v. Thiboutot, 448 U.S. 1, 12, 100 S. Ct. 2502, 2508, 65 L. Ed. 2d 555 (1980), may provide a remedy under 42 U.S.C. § 1983 for statutory claims grounded in the Developmentally Disabled Assistance and Bill of Rights Act. The potential availability of such a route for redressing infringements or deprivations of rights established by federal statutes further suggests that we are dealing not with matters of common law malpractice, but with fundamental interests of national import.

Although the varying standards set forth in this opinion may occasion concern that the issues presented to the jury might appear overly complex, we believe they are meaningful and carefully distinguish varying factual situations. It cannot be gainsaid that the act of shackling a resident is substantially different from the implementation of daily treatment decisions. Likewise, a failure to protect an individual from a sustained series of attacks cannot properly be equated with a failure to choose a particular technique for habilitation. Accordingly, special effort should be employed to ensure that juries can intelligently grasp the differences. Such an effort is not completely novel in the law. It takes place, for example, in areas of negligence law, immunity law, and admiralty law. We do not, therefore, share the concern expressed by the concurring opinion that district court judges or juries will be unable to understand these distinctions. Moreover, it is reasonable to expect that future plaintiffs will have grounds for raising only one or two of the claims analyzed here, thus further simplifying the difficult task. In any event, it would be a disservice to an injured person who claims to have been abused during a period of involuntary institutionalization to lump indiscriminately a variety of constitutional violations for the sake of beguiling simplicity. Such oversimplification, so handy in political debate, often lacks the precision necessary for resolving complicated legal questions.

Inasmuch as it is still unclear how the evidence will emerge at the retrial, we have set forth in an appendix what may provide a beginning point for a set of jury instructions in a matter of this type.60 Of course the specific instructions submitted to the jury must be carefully structured to reflect the actual record that is developed on remand.

Because we have already ruled that a new trial is required, we need not decide whether the other errors raised by plaintiff concerning the method in which the trial was conducted are themselves grounds for a new trial. However, we believe that a comment on the trial court's approval of dual representation for the defendants is in order. At a final pretrial conference, the attorney who had appeared for all the defendants, as a result of an insurance agreement, requested that both he and the deputy attorney general, who represented the defendants as state employees, be permitted to act as counsel during trial. The district court ruled in favor of this joint participation. In order to explain the presence of both attorneys, the court suggested that insurance counsel identify himself as the attorney for Dr. Youngberg, and that the deputy attorney general identify himself as counsel for defendants Matthew and Conley. The jury was so advised, and heard two sets of opening and closing statements, as well as double cross-examinations, although no separate legal interests or conflicts among defendants were asserted or surfaced.

On remand, the district court should make some ruling on this issue. If it determines that dual representation is permissible, it should set forth supporting reasons.61 Moreover, the practice of permitting counsel to make deliberate misstatements to the jury, even for the purpose of concealing the existence of liability insurance coverage, is of dubious propriety and should not be repeated.

In order to appreciate the full problem posed by this case, it is important to note that the Supreme Court has recognized a common law of qualified immunity protecting officials and employees from civil liability for acts performed in the discharge of duties imposed upon them. Such immunity is available even when constitutional rights have been transgressed, if the official or employee was not aware, or had no reason to be aware, of such rights. As the Supreme Court explained in Wood v. Strickland, liability for damages for every action which is found subsequently to have been violative of a person's constitutional rights and to have caused injury would unfairly impose upon administrators the burden of mistakes made in good faith in the course of exercising discretion within the scope of official duties. 420 U.S. 308 at 319, 95 S. Ct. 992 at 999, 43 L. Ed. 2d 214 (1975).

Because of the evolving nature of the rights in issue here, it may well be that the defendants did not know of them or had no reason to know of them governmental officials are not "charged with predicting the future course of constitutional law." Pierson v. Ray, 386 U.S. 547, 557, 87 S. Ct. 1213, 1219, 18 L. Ed. 2d 288 (1967). Institutional administrators are entitled to rely on traditional sources for the factual information incorporated in their policy decisions, and when faced with behavior causing or threatening disruption, have an obvious need for prompt action and judgments, which can only be based on existing knowledge. See Scheuer v. Rhodes, 416 U.S. 232, 246, 94 S. Ct. 1683, 1691, 40 L. Ed. 2d 90 (1974). Accordingly, when this matter is remanded for a new trial, the trial judge should instruct the jury regarding the possibility of immunity62 with the caveat that defendants' reasonable belief is to be judged at the time their actions were taken. The jury should further be charged that the defendants are not responsible for unforeseeable developments in the law.

The deplorable conditions at Pennhurst and the manifold problems and anguish that those conditions spawned have been addressed in Halderman v. Pennhurst. The present litigation deals with the more discrete problems facing individual patients and their relationships with officials and employees of the institution. More particularly, it relates to the constitutional rights possessed by such citizens and the responsibilities and duties of officials and employees who are claimed to violate such rights.

Unfortunately, these constitutional rights have not always been clearly perceived, and even when recognized the description of such rights, their articulation, and their application to particular factual patterns has only slowly emerged. Even today, the process is still in its early stages.

The difficulty of the task and it frequently is difficult to describe with exactitude the contours of developing constitutional protections is further complicated because the applicable medical disciplines are themselves still in their formative period, and because the resources available to the states are so limited. Nonetheless, where valid rights of citizens are at stake, the courts may not shirk the task of seeking to spell out the elements and details of such rights as carefully as possible.

To the extent that in the past we may have invoked inapplicable analogies, we should be forthright enough to admit the error and to seek to rectify it. To the extent that utilization of other concepts, though appealing, is found to be impractical, we should continue the quest for standards that are realistic and reflective of the changes that are still taking place in psychiatry, psychology, sociology and other related fields that so greatly affect the problems of the retarded.

At the same time we must be cautious in establishing rules that would visit serious financial liabilities on administrators, who, when they performed the acts in question, were not aware of the new standards of rights and responsibilities subsequently formulated by the courts or the new scientific advances that have been achieved to minister to persons who are mentally retarded.

The judgment of the district court will be vacated, and the case remanded for a new trial.

If you find that the plaintiff was shackled against his will, and defendants knew or had reason to know of such conduct, you may find the defendants liable unless they can offer compelling reasons for this action. If the defendants satisfactorily explain that it was necessary to shackle the plaintiff for his own protection or for treatment purposes, then they may not be liable. However, it must be demonstrated that the shackling was the least restrictive means of handling the resident that other, less severe measures had been tried or considered and found unworkable. Only compelling reasons relating to the treatment or the protection of the resident are permissible justifications for shackling.II. PROTECTION FROM ATTACKS

If you find that the plaintiff was harmed by a series of attacks, and the defendants knew or had reason to know of them, then you may hold the defendants liable. If the defendants knew or had reason to know of these attacks, then only if they provide substantial reasons which explain that the attacks occurred despite their attention will they be deemed to have fulfilled their duty to care for and to protect the plaintiff. Substantial explanations have to do with promoting the treatment or protection of the plaintiff. For example, if the staff considered it important, for therapeutic purposes, to have the resident mingling with others instead of confined to an isolated cell, even though some risk of injury existed, you may consider that to be a substantial explanation. But overcrowded conditions or inadequate staffing patterns may not justify the injuries.

A. If you find that the plaintiff was involuntarily committed for treatment, and no treatment was administered, and no compelling explanation for the lack of treatment was offered, you may hold the defendants liable.

B. If you find that the plaintiff has received some treatment, you must then determine whether the treatment is regarded as acceptable in the light of present medical or other scientific knowledge. If you find that it is not acceptable, you may find for the plaintiff. In addition, if the evidence does not demonstrate that there is a relationship between the treatment administered and the plaintiff's needs, even if the treatment is arguably regarded as acceptable in other situations, then you may find for the plaintiff. In deciding upon the adequacy of the treatment program adopted you may consider the defendants' explanations regarding security concerns, administrative necessities and fiscal constraints.

C. If you find that a selection of a mode of treatment subjected the plaintiff to significant deprivations of liberty, then you must go on and determine whether that decision provided for the least intrusive treatment available under the circumstances. If the defendants considered other alternatives and ascertained that the program adopted was the least intrusive available, then you should find the defendants not liable.

SEITZ, Chief Judge, concurring, with whom Judges ALDISERT, ROSENN, and GARTH join.

I agree with the majority that the judgment of the district court must be vacated and the case remanded for a new trial because of the exclusion of relevant expert testimony and the use of improper legal standards in the charge to the jury. I also agree with the majority that this case is governed by the due process clause of the fourteenth amendment and not by the eighth amendment. I write separately because of a pervasive disagreement with the majority with regard to the standards that should be employed in charging the jury on remand.1

Pennhurst is not unfamiliar to this court. Halderman v. Pennhurst State School & Hospital, 612 F.2d 84 (3d Cir. 1979) (in banc), cert. granted, 447 U.S. 904, 100 S. Ct. 2984, 64 L. Ed. 2d 853 (1980), was a class action on behalf of the residents of Pennhurst challenging the practice of institutionalizing the mentally retarded. Plaintiff Nicholas Romeo is a member of that class, and the defendants here are also defendants in Halderman. A majority of this court held in Halderman that federal and state statutory provisions granted the residents of Pennhurst a right to treatment in the least restrictive environment. See 612 F.2d at 104-07; Developmentally Disabled Assistance and Bill of Rights Act, 42 U.S.C. §§ 6001-6081 (1976); Mental Health & Mental Retardation Act of 1966, Pa.Stat.Ann. tit. 50, §§ 4101-4704 (Purdon 1969). The court remanded for an individualized determination of what type of environment, institutional or otherwise, best suited each class member's needs. The majority did not have to reach the constitutional issues presented by the plaintiffs.

This suit presents legal issues that take the next step beyond Halderman. The plaintiff alleges that while confined at Pennhurst he was deprived of his rights under the eighth and fourteenth amendments of the Constitution.2 The defendants are state officials who had policymaking and supervisory responsibilities at Pennhurst when the alleged constitutional violations occurred. They are charged by the plaintiff with failing to adopt policies and procedures that would protect the plaintiff from attacks by other residents and staff. The plaintiff also claims that the defendants violated his constitutional rights after his transfer to the Pennhurst hospital building by causing him to be physically restrained for long periods of time and by refusing to provide adequate treatment. The violations after the plaintiff's transfer to the hospital building were allegedly to punish him for filing this suit. Although the theory of the plaintiff's complaint is not entirely clear, the defendants are apparently charged with direct liability rather than vicarious liability for the conduct of other state employees supervised by them. The plaintiff's complaint requested injunctive relief as well as damages against the defendants in their official and individual capacities, but only the damage claims against the defendants in their individual capacities are raised on this appeal.

The majority establishes a multilevel series of standards to govern the plaintiff's claims. First, with respect to the protection claim, the majority holds that failure of the defendants to provide for the plaintiff's safety can be justified only by a showing of substantial necessity. Second, because physical restraint "raises a presumption of a punitive sanction," it can be justified only by a compelling necessity. In addition, the majority holds that the plaintiff is entitled to an instruction that the defendants must show that restraint was the "least restrictive method of dealing with the patient, in light of his problems and the surrounding environment." Third, the majority divides the treatment claim into three categories. If the jury finds that no treatment was administered, it may hold the defendants liable unless they can provide a compelling explanation for the lack of treatment. If some treatment was administered, the defendants will not be liable if that treatment was "acceptable in light of present medical or other scientific knowledge.... Thus, if defendants can demonstrate a coherent relationship between a particular treatment program and a resident's needs, they would not be liable." Finally, "least intrusive" analysis applies to the selection of a treatment approach if the jury finds that the approach subjected the plaintiff to "significant deprivations of liberty."

These standards are unnecessarily complex, and they undoubtedly will cause confusion. This complexity and confusion will affect both the ability of a jury to meaningfully assess liability and the ability of state officials like the defendants to effectively conform their conduct to these standards. The jury will be forced to make artificial distinctions between what are in reality related issues. Moreover, the jury must give essentially similar standards the distinctive meanings envisioned by the majority. For example, even if the jury can meaningfully differentiate between compelling necessity and substantial necessity, it is unlikely that it can do so in the manner contemplated by the majority, i. e., as a way "to distinguish between isolated incidents and inadvertent accidents, on the one hand, and persistent disregard of patients' needs, on the other."

The majority's multilevel standards also will inject complexity and confusion into the work of state officials who, like the defendants, have policymaking and supervisory responsibilities at institutions like Pennhurst. These officials will have to formulate policies and procedures to ensure that the institution's residents are (a) protected from attack unless some substantial necessity justifies the lack of protection, (b) left unrestrained unless some compelling necessity justifies their restraint, and (c) treated with various methods that may, depending on the circumstances, have to be based on accepted medical or other scientific knowledge or be the least intrusive treatment available.

Furthermore, in the unlikely event that state officials are able to formulate appropriate policies and procedures under the different standards, they still may be uncertain as to what standards govern their conduct. Treatment programs, for example, could be governed by more than one standard. Testimony at the trial indicated that some behavior modification programs utilize "timeout" principles involving periodic mechanical restraint of the limbs to eliminate aggressive behavior. It is not clear whether such a program has to be the least intrusive alternative or only be acceptable in light of present medical or other scientific knowledge.

This complexity and confusion is unnecessary. In some situations the development of multilevel standards may be unavoidable. I do not believe, however, that such standards are required in this case to accurately differentiate the factual and legal issues presented. As will be noted below, the plaintiff's claims are interrelated, and it is unrealistic to treat them as discrete questions. Protection, restraint, and treatment are not severable issues in the context of the institutionalized mentally retarded. I believe that a single standard can be established to protect the constitutional rights of committed persons while recognizing the legitimate interests of the state.

It is important to examine the relevant constitutional principles before formulating a standard to govern the issues presented in this case. Initially, I agree that conditions of confinement of the mentally retarded are subject to due process scrutiny. Government action that infringes liberty interests is subject to scrutiny under the due process clause and, at a minimum, it must be supported by a legitimate state interest. A mentally retarded person does not lose this protection merely because he is institutionalized.

This is not the end of the matter, however. Institutionalization by definition entails some restrictions on the personal liberty of committed individuals. As I noted in Halderman, I believe that "the state's willingness to provide the residents with such necessities as food, shelter, medical care, and supervision, for which the residents have no other source, forms an adequate basis for some state-imposed restrictions on their liberty." 612 F.2d at 125 (Seitz, C. J., dissenting). The questions that remain, of course, are the degree to which personal liberty can be infringed and the duties owed by the state to residents of its institutions.

The state may not confine a mentally retarded individual unless it has at least a legitimate purpose for doing so. Because the state can confine an individual only for certain purposes, the conditions of that confinement must bear a reasonable relationship to those purposes. As the majority notes, three state interests traditionally have been advanced to justify commitment of the mentally disabled: danger to others, danger to self, and the need for care or treatment. When analyzing these state interests it is tempting to speak of them as being analytically distinct. In reality, they represent a finely meshed interaction of state interests and patient needs. This is especially true in the context of postcommitment issues. For example, it is somewhat artificial to speak of danger to self as being distinct from the need for care or treatment. The very fact that an individual is dangerous to himself means that he needs care and treatment. Similarly, the need for a safe environment, which is raised by the plaintiff's protection claim, implicates all three interests. Some of the plaintiff's injuries are the result of retaliation by other patients for his attacks on them, and this retaliation has a double effect: it makes the plaintiff's environment physically unsafe and it tends to impede his maximum development. Although the three interests are not entirely coextensive, this interrelation must be kept in mind when employing legitimate state interest analysis.

With this view of state interests and patient needs in mind, I believe that the plaintiff has a constitutional right to minimally adequate care and treatment. The existence of a constitutional right to care and treatment is no longer a novel legal proposition. See, e. g., Donaldson v. O'Connor, 493 F.2d 507 (5th Cir. 1974), rev'd on other grounds and remanded, 422 U.S. 563, 95 S. Ct. 2486, 45 L. Ed. 2d 396 (1975); Rouse v. Cameron, 373 F.2d 451 (D.C.Cir.1966). Although the seminal right-to-treatment cases were concerned only with the mentally ill, recent cases have extended this right to the mentally retarded. See, e. g., Welsch v. Likins, 550 F.2d 1122 (8th Cir. 1977); Wyatt v. Aderholt, 503 F.2d 1305 (5th Cir. 1974).

It is important to remember, however, that insofar as a constitutional right to treatment is concerned, there are critical differences between the mentally ill and the mentally retarded. Cf. Kremens v. Bartley, 431 U.S. 119, 135-36, 97 S. Ct. 1709, 1718-19, 52 L. Ed. 2d 184 (1977) (in defining appropriate class in rule 23 action challenging state commitment statute, court must pay careful attention to differences between mentally ill and mentally retarded). For example, this court noted in Halderman that "(s)trictly speaking, since mental retardation is not a curable disability, the term 'treatment' is inappropriate. Rather, 'habilitation,' which refers to 'that education, training and care required by retarded individuals to reach their maximum development'... is the more appropriate term." 612 F.2d at 95 n.14 (citation omitted). It is in this sense that I use the term "treatment" in this opinion. In addition, profoundly retarded persons such as the plaintiff are generally unable to provide themselves with food, shelter, and clothing as well as basic protection from other persons and physical hazards. Thus, the danger rationales justifying commitment of the mentally retarded are often quite different from those justifying commitment of the mentally ill.

The state does not contest that it has placed the plaintiff in Pennhurst to provide basic care and treatment. Indeed, he has a right to treatment under state law, see Halderman, 612 F.2d at 100-103, and the fact that Pennhurst has programs and staff to treat patients is indicative of such a purpose. I believe that when the purpose for confining a mentally retarded person is to provide care and treatment, as is undoubtedly the case here, it violates the due process clause to fail to fulfill that purpose. With the above considerations in mind, I will now examine the contours of the constitutional right to care and treatment in the context of each of the plaintiff's claims.

The plaintiff's first claim is that the defendants violated his constitutional rights by not adopting policies and procedures that would protect him from attacks by other residents and Pennhurst employees. In its charge to the jury on this issue, the district court stated that the defendants could be held liable only if they were "deliberately indifferent to the medical and psychological needs of (the plaintiff)." The plaintiff argues that the district court erred in using this language. The phrase "deliberate indifference" derives from Estelle v. Gamble, 429 U.S. 97, 97 S. Ct. 285, 50 L. Ed. 2d 251 (1976), in which the Supreme Court held that deliberate indifference to the serious medical needs of prisoners "constitutes the 'unnecessary and wanton infliction of pain' proscribed by the Eighth Amendment." Id. at 104, 97 S.Ct. at 291 (quoting Gregg v. Georgia, 428 U.S. 153, 173, 96 S. Ct. 2909, 2925, 49 L. Ed. 2d 859 (1976) (joint opinion of Stewart, Powell, & Stevens, JJ.)). The question presented here is whether this eighth amendment standard is a correct statement of the duty imposed by the due process clause to care for and to treat the institutionalized mentally retarded.

The constitutional right to care and treatment means that the defendants must provide a "basically safe and humane living environment." Goodman v. Parwatikar, 570 F.2d 801, 804 (8th Cir. 1978). This duty includes reasonable protection from attacks by other residents and staff. At a minimum, the defendants have an affirmative obligation to discover the needs of mentally retarded patients for protection and to respond to those needs in an adequate manner. Because the deliberate indifference standard is not a proper method to implement these affirmative due process requirements, use of such language in the charge was reversible error.

The plaintiff contends that the correct legal standard is that the defendants must take reasonable steps to protect the plaintiff. This standard, however, is virtually indistinguishable from the standard that would apply in a state malpractice action. Some courts have analogized to common-law tort principles in adopting the standard of care for a section 1983 claim. See, e. g., Whirl v. Kern, 407 F.2d 781 (5th Cir.), cert. denied, 396 U.S. 901, 90 S. Ct. 210, 24 L. Ed. 2d 177 (1969). However, the Supreme Court has exhibited concern in an eighth amendment context about turning every state malpractice claim into a constitutional violation. See Estelle, 429 U.S. at 106, 97 S.Ct. at 292, ("deliberate indifference" standard used to distinguish constitutional violations from medical malpractice). Constitutionalization of state torts is equally a concern in due process analysis. See, e. g., Paul v. Davis, 424 U.S. 693, 96 S. Ct. 1155, 47 L. Ed. 2d 405 (1976); United States v. Delerme, 457 F.2d 156 (3d Cir. 1972). Adoption of the standard requested by the plaintiff would make section 1983 coextensive with malpractice claims when state action is present.

Distinguishing constitutional violations from ordinary malpractice claims is especially difficult in the present case because care and treatment often involve issues similar to those presented by malpractice actions. Nevertheless, I believe that this distinction should be maintained in order to confine the jury to constitutional deliberations. It is true that some malpractice claims may also rise to the level of a constitutional violation. This does not mean, however, that liability under section 1983 should be coextensive with liability in state malpractice actions. The fourteenth amendment was not intended to remedy every tort violation in which there is state involvement. Section 1983 itself places limitations on the constitutionalization of state tort claims by requiring that plaintiffs be deprived, under color of state law, of a right secured by the Constitution and laws of the United States. Strict adherence to these requirements prevents section 1983 from becoming a "font of tort law to be superimposed upon whatever systems may already be administered by the States." Paul v. Davis, 424 U.S. at 701, 96 S.Ct. at 1160.

Because federalism principles are implicated here, I believe that a further limitation is necessary. Under these circumstances I would hold that allegations of constitutional violations based only upon mere malpractice do not state a cause of action under section 1983. See Hampton v. Holmesburg Prison Officials, 546 F.2d 1077, 1081 (3d Cir. 1976) (neglect, carelessness, or malpractice without more does not establish constitutional violation under § 1983); Harper v. Cserr, 544 F.2d 1121, 1124 (1st Cir. 1976) (section 1983 would not afford relief for complaint alleging negligence or malpractice). This construction of the statute implements the objectives of section 1983 without encroaching on interests that are only a matter of state concern. It must be remembered that with respect to the institutionalization of the developmentally disabled, the Constitution only establishes minimum standards below which the state's conduct may not fall. Although common-law tort principles may provide a useful starting point in formulating the appropriate standard of care under section 1983, they cannot be determinative of the constitutional issue.

In developing a standard to assess whether the defendants have infringed the plaintiff's constitutional right to care and treatment, it is also important to keep in mind the nature of the conduct that will be evaluated by the jury. The question of how best to treat the mentally retarded is a hotly debated subject among experts, and the answers most likely will change over time with the acquisition of new knowledge. Although the statutes in Halderman required us to enter the debate, I believe that such a debate should not be elevated to constitutional proportions. A statute is relatively easy to change but the Constitution is not. In my view, the Constitution only requires that the courts make certain that professional judgment in fact was exercised. It is not appropriate for the courts to specify which of several professionally acceptable choices should have been made. Cf. Developments in the Law Civil Commitment of the Mentally Ill, 87 Harv.L.Rev. 1190, 1342-44 (1974) (discussing use of an administrative law model to ensure that court does not engage in evaluating various psychiatric theories).

Although I have indicated that the care and treatment of the institutionalized mentally retarded must be left largely to the appropriate professionals, some judicial scrutiny is essential if the right to care and treatment is not to become meaningless. As I noted previously, the defendants must discover the needs of the institution's residents and, if action is necessary, respond adequately to those needs. Once the defendants have taken action, or have chosen not to act, it must be determined whether their conduct satisfies the Constitution.

I would hold that the jury should be instructed that the defendants are liable if their conduct was such a substantial departure from accepted professional judgment, practice, or standards in the care and treatment of this plaintiff as to demonstrate that the defendants did not base their conduct on a professional judgment. This is not a malpractice standard. By "accepted professional judgment" I do not mean some standard employed by a reasonable expert or a majority of experts in the community, as state malpractice actions would require, but rather that the choice in question was not a sham or otherwise illegitimate. The jury is to decide only whether the defendants' conduct had some basis in accepted professional opinion. Furthermore, unlike state malpractice actions, a departure from accepted professional judgment must be substantial to give rise to liability. Although violations of the standard that I have developed would probably contravene state malpractice standards as well, this does not mean that the two standards are coextensive. The "substantial departure from accepted professional judgment" standard effectively distinguishes between conduct that violates the minimum requirements of the Constitution and conduct, such as ordinary malpractice, that does not.

Instructions consistent with the previous paragraph properly balance the plaintiff's constitutional right to minimally adequate care and treatment against the legitimate interests of the state. Thus, with regard to the plaintiff's claim that the defendants failed to adopt policies and procedures that would adequately protect him from attacks by other residents and staff, the jury should be charged in accordance with the "substantial departure from accepted professional judgment" standard. The district court's charge is inconsistent with this standard, and it improperly excluded expert testimony relevant to the above inquiry.

In his second claim, the plaintiff alleges that the defendants violated his constitutional rights by causing his freedom of movement to be restricted through the use of various mechanical restraints while he was confined in the Pennhurst hospital building.3 The majority holds that the least restrictive alternative standard applies to this claim. I believe that this standard does not represent what the Constitution requires as a minimum for a number of reasons. In Halderman this court held that the residents of Pennhurst have a right to treatment in the least restrictive environment under the Developmentally Disabled Assistance and Bill of Rights Act and the Pennsylvania Mental Health & Mental Retardation Act of 1966. However, Halderman rested solely on statutory grounds and the present case involves the Constitution.4 Furthermore, the least restrictive alternative standard is inappropriate here for a far more fundamental reason.

The major issue in Halderman was whether institutionalization was per se illegal. Although we referred to the living conditions at Pennhurst, the main issue in Halderman was where the members of the class should be treated. That question is not presented in this case. See note 2 supra. The only question here is whether, assuming institutionalization is legal, the actions taken within the institution satisfy the Constitution. Regardless of the validity of least restrictive alternative analysis in other facets of treatment of the mentally retarded, it simply has no applicability to conditions within an institution.

Implicit in the least restrictive alternative theory is the notion that the state's only interest in institutionalizing a person is treatment. See, e. g., Wyatt v. Aderholt, 503 F.2d 1305, 1313 (5th Cir. 1974) (state interests other than treatment are "trivial" compared to massive curtailment of liberty involved). But see Morales v. Turman, 562 F.2d 993, 998 (5th Cir.) (questioning this and other aspects of Wyatt), rev'd and remanded on other grounds, 430 U.S. 322, 97 S. Ct. 1189, 51 L. Ed. 2d 368 (1977) (per curiam). I believe, however, that this gives inadequate recognition to the subtle interrelation of the various state interests involved.

As already noted, the three state interests normally associated with confinement of the developmentally disabled danger to others, danger to self, and need for care or treatment overlap and mesh together to some extent in this context. This is especially true as to actions taken within the institution, where decisions about proper care or treatment often must be made on a day-to-day basis. The least restrictive alternative test gives inadequate recognition to this fact. Because the state's interests are rather complex, a particular mode of care or treatment may serve one state interest but not another. For example, physical restraints may serve short-run safety needs but not long-run habilitation needs. By focusing on the "best" care or treatment through a rather rigid view of the state interests, least restrictive alternative analysis tends to obscure the true nature of the state interests. Because the standard to be applied must give recognition to the interplay of the various interests, I believe the majority's test is an inappropriate standard for constitutional purposes. In short, even if least restrictive alternative analysis provides an appropriate framework for answering the question of where to place the mentally retarded, it is simply too rigid a tool once we leave that question and focus on conditions of care and habilitation within a particular institution.

Moreover, the least restrictive alternative standard is not easily transferable from the context in which it originated to the situation presented by this case. The standard evolved from the less drastic means analysis formulated by the Supreme Court in cases involving free speech. See, e. g., Shelton v. Tucker, 364 U.S. 479, 81 S. Ct. 247, 5 L. Ed. 2d 231 (1960). Since its first exposition of the less drastic means doctrine, the Supreme Court has been rather cautious in using the doctrine because some less drastic alternative almost always exists. If used without caution, the doctrine could invalidate almost any state action. See generally Illinois State Board of Elections v. Socialist Workers Party, 440 U.S. 173, 188-89, 99 S. Ct. 983, 992-93, 59 L. Ed. 2d 230 (1979) (Blackmun, J., concurring); Note, Less Drastic Means and the First Amendment, 78 Yale L.J. 464, 472 (1969).

These difficulties can be avoided where there is one, objectively ascertainable less drastic means. See Richardson, Freedom of Expression and the Function of Courts, 65 Harv.L.Rev. 1, 40 (1951). This single, objective alternative simply does not exist in the care and treatment of the mentally retarded because what is the best care or treatment at any given time may be a subject of great professional debate. Thus, judges and juries likely will not be able to make a single, objective determination. Furthermore, because a mentally retarded individual's needs change over time, the least restrictive alternative standard will in effect require continuing evaluation of the "best" method, not a single determination. Although Congress or the states may adopt this standard as part of a statutory scheme, such difficulties caution against use of the doctrine as a constitutional matter. Thus I disagree with the majority's application of the least restrictive alternative standard to the plaintiff's restraint claim.

The "substantial departure from accepted professional judgment" standard that I developed in relation to the protection claim should also apply to the restraint claim. This standard provides for judicial scrutiny yet gives the state leeway in which to accommodate the interests it has in confining the mentally retarded. At first glance, this standard may appear to be inappropriate in the context of a restraint claim. As I noted earlier, however, the state's interests are interrelated, and it is unrealistic to view each of the plaintiff's claims as separate and distinct. Mechanical restraints, when used properly in accord with preestablished procedures, can be an integral part of a treatment program.5 Furthermore, restraints may be essential to protect the safety of the patients.6 Whether this is the situation here is, of course, a question for the jury.

I believe that the jury should be instructed that physical restraint related to the provision of care or treatment violates the plaintiff's rights if the defendants' conduct was such a substantial departure from accepted professional judgment, practice, or standards as to demonstrate that the defendants did not base their conduct on a professional judgment. Restraint for the convenience of the staff or for punishment serving no legitimate purpose would be a substantial departure from accepted professional judgment and thus a violation of the plaintiff's constitutional right to care and treatment. Of course, other prerequisites to establishing the defendants' liability also would have to be met.

The plaintiff's final claim is that the defendants deprived him of his constitutional rights by not making available adequate treatment while he was confined in the Pennhurst hospital building. The plaintiff also appears to be asserting that he has a right to treatment in the least restrictive alternative. My discussion of the least restrictive alternative standard in the preceding section applies with equal force to the treatment claim. As a result, I agree with the majority's holding that least restrictive alternative analysis is inappropriate when assessing the adequacy of treatment already administered. I do not agree, however, that the standard articulated by the majority is a correct statement of what the Constitution requires, nor do I concur in the majority's promulgation of a three-tier framework for analyzing treatment claims.

The majority holds that when the jury finds that some treatment has been administered, it should determine whether the treatment is regarded as acceptable in light of present medical or other scientific knowledge. This standard is in effect the same standard that would apply in a state malpractice action. As noted previously, it is of vital importance to maintain the distinction between common-law tort actions and constitutional decisions. The "substantial departure from accepted professional judgment" standard preserves this distinction while accommodating the concerns of both the institutionalized mentally retarded and state officials.

The majority also divides treatment claims into three categories and formulates different standards of review for each category. I believe, however, that such distinctions are artificial and will lead to unnecessary complexity and confusion. See part II supra. The jury should be instructed to apply the "substantial departure from accepted professional judgment" standard to all aspects of the treatment claim. Inaction by the defendants or the use of restrictive procedures in treatment can, of course, be taken into consideration by the jury in determining whether professional judgment has been exercised.

The lamentable conditions that exist at Pennhurst have been detailed elsewhere. See Halderman, 612 F.2d 84. The institution is unsanitary and understaffed, and many of its residents are subjected to violence and enforced inactivity. I stand second to no one in condemning these conditions, and, as I emphasized in my dissent in Halderman, I believe that the federal judiciary has a duty to ensure that these conditions are eliminated. However, our duty in this case is only to establish minimum constitutional standards to govern the individual claims of a Pennhurst resident against certain Pennhurst officials. I would hold that the Constitution requires the district court to instruct the jury that the defendants' conduct comports with due process if it is not a substantial departure from accepted professional judgment. Because the majority's decision is inconsistent with this standard, I cannot join the opinion of the court.

ALDISERT, Circuit Judge, with whom Judge GARTH joins, concurring.

Although I join the concurring opinion of Chief Judge Seitz in all respects, I write separately to emphasize the two important differences, one conceptual and one pragmatic, that divide this court.

The concurring judges approach this troublesome problem in the grand tradition of the common law, precisely distinguishing between the elements of a due process deprivation and the elements of the state tort of medical malpractice. Chief Judge Seitz emphasizes with elegance and profound jurisprudential integrity that not all state law torts amount to due process deprivations. He contends that in the factual complex presented here, a prima facie case of a due process deprivation is established only by proof of "such a substantial departure from accepted professional judgment, practice, or standards ... as to demonstrate that the defendants did not base their conduct on a professional judgment," or, in other words, that the treatment technique employed was "a sham or otherwise illegitimate." Con. op. of Seitz, C. J., at page 178.

In the common law tradition, courts resolve disputes by examining the rules laid down by prior decisions. These are rules of law in the narrow sense, "precepts attaching a definite detailed legal consequence to a definite, detailed state of facts,"1 or "fairly concrete guides for decision geared to narrow categories of behavior and prescribing narrow patterns of conduct."2 The common law "creeps from point to point, testing each step"3 and grows slowly by gradual accretion from the resolution of specific problems. Holmes noted that the great growth of the common law came about incrementally.4 Courts fashion principles from a number of rules of decision, in a process characterized by experimentation. Rules of case law are treated not as final truths, "but as working hypotheses, continually retested in those great laboratories of the law, the courts of justice."5 The common law tradition has been described as a method "of reaching what instinctively seem the right result in a series of cases, and only later (if at all) enunciating the principle that explains the pattern a sort of connect-the-dots exercise."6

Chief Judge Seitz adheres to this tradition in two respects. First, he sets forth a detailed legal consequence for the detailed set of facts in this case. By contrast, the majority's formulation is cast in terms of grandi astrazioni, predicating liability on notions of "substantial necessity," "compelling necessity," and "least intrusive means." These abstractions, dubbed in classroom jargon as constitutional law "buzz words," have the unfortunate capacity to mean all things to all people.

Second, Chief Judge Seitz develops the law incrementally, explaining how a respectable body of state tort law, built on thousands of fact situations resolved by courts over several centuries, is immediately available to the plaintiffs. If the plaintiffs prove egregious conduct constituting gross negligence and demonstrate that the treatment or lack of treatment may be fairly characterized as a sham, the plaintiffs would have available additional relief to cure the constitutional deprivation. Thus, he recognizes that the plaintiffs carry separate arrows in their quiver: state malpractice relief for certain levels of conduct falling below standards of reasonable medical practice, and federal relief under 42 U.S.C. § 1983 if the conduct is sufficiently egregious to violate the fourteenth amendment.

By approaching the case from the common law perspective, Chief Judge Seitz recognizes a tenet recently articulated by one of America's great law schools, to which I subscribe completely:

The United States, more than any other country, has been a country of law, laws, and lawyers; nowhere else is so much of the ferment of society expressed in terms of a Constitution.... We have come, however, to see that all the great issues of the day are not concentrated on constitutional law ... but that contemporary problems embrace and are informed by a variety of public law fields, from tax and administrative law to health, labor and penal law. We have come to see, also, that the traditional private law subjects contracts, torts, and property, for instance are more and more intertwined with considerations of public law and public interest; and further, that much of the traditional learning in these subjects is not revealed truth, but a reflection of values of society at given periods of time.7

Rather than creating overriding constitutional precepts that obfuscate the role of tort law in correcting institutional abuses, he relies on the Anglo-American tort law experience for a more principled resolution of this controversy.

Although the majority purport to recognize and preserve the distinction between claims under the Constitution and claims under state tort law, their decision actually blurs that distinction. By formulating a standard for constitutional deprivation low enough to be asserted in virtually all malpractice actions arising out of the treatment of the institutionalized retarded, the majority place a substantial burden on lay juries, who must avoid confusing the elements of these related claims. Notwithstanding my high regard for lay jurors, I doubt that medical professionals, much less laymen, can foresee the ramifications of the majority's due process standards.

In addition, by promulgating standards for due process claims that are facially almost indistinguishable from malpractice allegations, the majority have effectively extended federal jurisdiction into all malpractice claims arising in state institutions. Although diversity is currently required to litigate state tort claims in the federal courts, future plaintiffs will be able to assert, undoubtedly in good faith, that the state treatment had no "coherent relationship" to their needs, that they were restrained without "compelling necessity," or that they were left unprotected without "substantial necessity." These plaintiffs will be able to assert federal jurisdiction under 28 U.S.C. § 1343(3), and append their claims of malpractice. I do not doubt that the federal courts of this circuit will soon be repositories for all personal injury claims arising in state institutions for the mentally retarded. Indeed, that appears to be the majority's intent, for they advance the rather peculiar notion that these situations, in which a person has been committed to a state institution by a state process, subject to state control through state employees, present issues not primarily of state concern, but "fundamental issues of national import." Maj. op. at page 170. This expansion of jurisdiction deserves greater consideration than just one case.

The majority's abandonment of incremental decisionmaking in favor of promulgation of broad standards is not only methodologically offensive, but lacks utility for the groups most affected by this decision. As the preceding concurring opinion notes, the complexity and confusion of the majority's formulation "will affect both the ability of a jury to meaningfully assess liability and the ability of state officials like the defendants to effectively conform their conduct to these standards." Con. op. of Seitz, C. J., at page 174. But the confusion will not start at the door of the jury room. In making decisions on motions for summary judgment or directed verdict, the district judges will be grasping at ropes of sand when they attempt to perceive the current wisdom of "substantial necessity" in one category of behavior or "compelling necessity" in another. And to predict what a given panel of this court will do on review would be akin to interpreting the speech of Pythia after she chewed the laurel leaves at Delphi.

But many persons besides juries, judges, and legal philosophers must participate in this divining process. There are staff physicians and, more important, staff attendants. Staff attendants, even more than physicians, are responsible for the hour-by-hour operation of state mental institutions, and their activity has now become strictly regulated by amorphous constitutional law tenets. We can assume that these attendants have neither the intellectual acumen nor the constitutional law insight possessed by Anthony Amsterdam, Gerald Gunther, Archibald Cox, or Laurence Tribe. Yet the majority formulation now requires them to apply these new constitutional precepts in deciding, for example, whether to employ "soft restraints," metal shackles, or a leather muff to control a disruptive patient. See con. op. of Seitz, C. J., at page 179 n.3. As a result, staff attendants as well as staff physicians will be compelled to seek legal counsel. Their attorneys will then prophesy "what the courts will do in fact...."8 The lawyers, even specialists in medical malpractice, public or hospital administration, or casualty insurance underwriting, will be hard pressed to formulate guidelines.

One obvious situation in which guidelines will be difficult to formulate comes to mind. The plaintiff here complained of seventy separate injuries. We can assume that some were self-inflicted; some were inflicted by fellow patients, retaliating from the plaintiff's aggressive behavior. The majority formulation makes it a violation of the United States Constitution to shackle one patient improperly, and a separate violation if an unrestrained patient attacks another patient. In the real world environment of a hospital ward the staff attendant must decide which constitutional right deserves protection. The decision must be made amid the cry and tumult of a mental hospital, not in the cloister of the Third Circuit law library, nor in the conference room of nine circuit judges who are divided five to four on a difficult problem.

It took centuries for the common law to move from thirteenth century notions of criminal trespass to the landmark decisions establishing trespass on the case, now modern day negligence.9 It took the French from July, 1790, to March, 1804, to formulate the French Civil Code,10 and the Germans from 1873 to 1896 to fashion the Burgerliches Gesetzbuch (BGB).11 More recently, it took fifteen years of incremental development to produce strict liability for defective products.12 Attorneys attempting to advise clients by predicting the effect of today's decision on specific fact patterns will be deprived of the body of law leading to the broad standards promulgated by the majority. Their ability to predict will suffer, and the persons most affected will be those who seek legal advice.

The majority's purpose is admirable. They recognize that the disposition of this case will presage an expansion of litigation involving the constitutional rights of the institutionalized mentally ill. In an attempt to aid the lower courts in dealing with the anticipated cases, the majority formulate general standards for evaluating these claims. But their approach does not accord with the common law adjudicatory tradition of proceeding gradually and empirically. The direction and instructions presented are thus founded not on experience, but on speculation. It is not surprising that this attempt to achieve predictability in disregard of the common law tradition achieves just the opposite result. All persons who must rely on these standards institutional staffs, physicians, their attorneys, district judges, and subsequent panels of this court will struggle to determine the context in which the claim arose,13 and then to apply the appropriate standard for that context. The resulting confusion will undermine the majority's attempt to formulate universal precepts on a limited record. I predict that this court's current inexperience in this area will become increasingly apparent in subsequent cases.

The majority have failed to recognize that every new case is an experiment, and that if the accepted rule that seems applicable yields a result felt to be unjust, under the common law tradition, the rule is reconsidered. "(I)f a rule continues to work injustice, it will eventually be reformulated."14 As Chief Judge Seitz explains, however, "(a) statute is relatively easy to change but the Constitution is not." Con. op. of Seitz, C. J., at page 178.

It has been almost twenty years since I faced a client across a law office desk. But were I to be placed in that position today, my advice would probably go like this: if you are contemplating a position as an attendant in a mental hospital, seek another job; there is simply too much unpredictability in the law governing your conduct. If you are a physician, make certain that the state's malpractice insurance policy includes a clause protecting you from the new "constitutional torts" manufactured today by the Third Circuit Court of Appeals. If you are an insurance carrier, set your premiums high, anticipating fully and completely the open-ended notions of "substantial" and "compelling" necessity and "least intrusive treatment." If you are a governor or state legislator, cut back and retrench institutional programs for the mentally retarded, because the potential for lawsuits has now increased geometrically, with corresponding demands on the tax base to support the costs. If you are a parent with a retarded child, I simply feel sorry for you. The sincere effort to provide financial relief for personal injuries suffered by retarded persons has been accomplished at an enormous social cost. In striking the balance between how seriously the complainants are being hurt and how much it will cost to afford them more relief than is presently available under state tort law, the balance has been struck so vaguely that state institutional programs, already diminishing, will be curtailed further. Given the recent political currents, there simply are not, and will not be, sufficient tax dollars, state or federal, to give the defendants adequate financial protection to meet the illusory standards promulgated today.

GARTH, Circuit Judge, with whom Judge ALDISERT joins, concurring.

I join in Chief Judge Seitz's concurring opinion. I write separately, however, to express my views with respect to a troubling feature of the majority opinion. As Chief Judge Seitz has recognized in note 1 of his concurring opinion, this case does not present claims involving nonreversible surgery or the administration of antipsychotic drugs. The majority concedes as much. Nonetheless the majority includes both of these procedures in its discussion (Maj. Op. pages 165, 166). That discussion is then apparently related to the jury charge which appears as III C in the Appendix (Maj. Op. page 173) requiring as a constitutional measure the application of a "least intrusive treatment" standard where "significant deprivations of liberty" occur.

The problem with including this discussion and this charge in this case is apparent. It is crystal clear that the Plaintiff here had neither alleged nor suffered from non-reversible surgery. Nor did he allege or suffer from being medicated with a powerful anti-psychotic drug. Thus, the entire discussion in the majority opinion which refers to these two conditions is gratuitous and constitutes no more than dictum. Moreover, while it may be contended that this discussion of surgery and drugs was only to illustrate the thesis that various standards are needed to accommodate different levels of medical treatment, by referring to procedures outside the Romeo record and crafting jury instructions which govern these procedures, the majority has trenched directly upon a case which does concern the administration of anti-psychotic drugs and which is pending before this Court. That case, Rennie v. Klein, Nos. 79-2576 and 79-2577, was argued before this Court on April 22, 1980. The filing of the panel's decision has been withheld by a majority vote of the full Court, pending the filing of the present en banc opinion.

Rennie deals with the issue of the involuntary administration of anti-psychotic drugs to mentally ill patients (as distinguished from mentally retarded patients, the focus of the Romeo case). In Rennie, the trial of this issue produced an extensive record which centered wholly on the administration of anti-psychotic drugs. That record, and the issues arising from it, were fully briefed and argued by the parties in Rennie before a panel of this Court. The decision in Rennie will therefore meet head-on the very issue of the requisite constitutional standards involved in the administration of anti-psychotic drugs which the Romeo majority by its equivocal discussion and purported adoption of a "least intrusive" standard seeks to influence, if not to establish, here. This has been sought to be accomplished notwithstanding the absence of an appropriate record in Romeo and without the benefit of analysis, argument, or briefing of the anti-psychotic drug issue by the parties in Romeo.

Thus, the multi-level series of standards adopted by the majority has now additionally resulted in bald dictum which purports to promulgate a legal standard involving the administration of anti-psychotic drugs a standard that is wholly inappropriate for resolution in this case and one that may very well be rejected or modified by the holding of this Court in Rennie v. Klein, supra, the case which presents that very issue.