I. INTRODUCTION
John Ballay was born in Albania in 1933, where he was educated, served in the military and worked until the mid 1950’s. He then entered Yugoslavia and was soon imprisoned for six years as a result of his militant anticommunist activities. Following his release he entered Austria, where he was employed as a tailor. He journeyed to the United States in 1966 “because everybody knows that the United States are [sic] the country of freedom and democracy,” 1 and became a citizen in 1970. Between 1966 and 1970 he occupied the position of machine operator at a factory in New York. Subsequently, from March 7 to June 7, 1971, he was employed at a club in New York City, apparently as a porter. Mr. Ballay has no known criminal record nor had he any known record of mental illness prior to 1971.
*60On January 19, 1971, Mr. Ballay appeared at the United States Capitol and claimed he was a Senator from Illinois. He was promptly committed to Saint Elizabeths Hospital upon the certified recommendation of a member of the hospital staff, where he remained until his discharge on March 5 of that year. On June 7 he visited the White House claiming that he was a Senator from Illinois and asking to see Tricia Nixon concerning her forthcoming marriage. The result was identical and he was committed until June 25. Finally, he again arrived at the White House gate on June 29, 1971, claiming to be a Senator from Illinois, the husband of Tricia Nixon, and seeking an audience with the President. The resulting civil commitment is the subject of this appeal.
After his June 29 appearance at the White House, Mr. Ballay was taken into custody and committed to Saint Elizabeths Hospital for emergency observation on the basis of an application filed by a Secret Service agent. See 21 D.C. Code § 521 (1967). He was then examined by an undisclosed staff psychiatrist for an undisclosed period of time, after which a written petition for an order authorizing continued hospitalization was filed with the United States District Court for the District of Columbia. See 21 D.C.Code § 523 (1967). On June 30 an order was entered permitting his institutionalization for an additional seven days for the purpose of continued observation and diagnosis. See 21 D.C. Code § 524 (1967). Ballay was thereafter detained pursuant to 21 D.C.Code § 528 (1967) at the discretion of the administrator of the hospital pending a hearing before the Commission on Mental Health.
Following a hearing held approximately one month later, the Commission recommended that the court commit Mr. Ballay for institutional care. Ballay asserted his right to a jury trial, 21 D.C. Code § 545 (1967), was found by the jury to fall within the statutory proscription, and was committed to Saint Elizabeths Hospital. An alleged constitutional error in that trial is the subject of this appeal.
The issue presented, while perhaps abstruse in resolution, is prosaic in statement. John Ballay was alleged to be “mentally ill and, because of that illness, . likely to injure himself or other persons if allowed to remain at liberty . . . ,”2 The jui’y was instructed that it must be convinced by a preponderance of the evidence that both elements were present, a standard of proof which has been consistently applied by force of case law.3 The question is whether appellant was deprived of due process 4 of law because the jury did not determine, beyond a reasonable doubt, that he was mentally ill and consequently dangerous.
The process accorded in any adversary proceeding reflects the interests at stake. In the present case the paramount interest is liberty, since the individual who is civilly committed faces restrictions which may exceed in length those imposed in most circumstances on *61the criminal or juvenile delinquent. Our deliberation therefore focuses on competing interests in an attempt to determine whether any may offset the immense individual interests involved. Focusing precisely on the state interest is a difficult task, however, because the statutes which address the enormous problem of mental illness broadly reflect dual motives, each of which may permit or require distinct procedures if considered separately. The first and dominant objective involves society’s concern with antisocial conduct. This leads inexorably to analogy with the criminal system, not only because there are certain similarities in objective, but primarily because the resulting restriction of liberty has assumed a significant and visible role in the creation of inhibitions to the state’s overzealous or mistaken application of that power. On virtually every plane of comparison the civil model presents an equally compelling plea for a stringent burden of proof. The justifications for institutional confinement as a means of implementing state interests —in traditional terms retribution, rehabilitation, deterrence and protection— are only partially applicable to civil commitment. The evidence which serves as a prerequisite to hospitalization remains uncertain, particularly with respect to predictions of dangerousness, and the ultimate decision may therefore unduly reflect clinical, rather than the appropriate legal and community, considerations.
Inextricably intertwined in both the statute and its legislative history is a second state interest involving its role as parens patriae. While viscerally a more persuasive rationale in terms of offsetting the individual’s loss of liberty, the argument largely dissolves upon closer inspection. No distinction is made between the statutory standards permitting institutionalization where dangerous to society and where dangerous to self in terms of the individual or in terms of the differing process which distinct interests may require. Moreover, the latter standard itself sweeps in varied and complex classes who represent different interests and it also presents equally perplexing definitional problems. Recognizing again the immense individual interests involved, it is questionable whether a rather significant margin of error should be tolerated regardless of the rationale, particularly since a more demanding burden of proof is by its nature largely neutral in its affect on relevant state policies. It is more appropriately characterized as a particularly suitable means of reducing the risk of factual errors which may be engendered by the statute or by the difficulties inherent in the disciplines associated with mental illness. Finally, we cannot help but recognize the stigma which unfortunately still accompanies a finding of mental illness.
“Stone walls do not a prison make, nor iron bars a cage.” 5 We align ourselves with those courts 6 that have held that proof of mental illness and dangerousness in involuntary civil commitment proceedings must be beyond a reasonable doubt.7
*62II. MOOTNESS
Ballay was discharged from Saint Elizabeths Hospital on November 4, 1971. The government contends that the present appeal is therefore moot. We think not. Appellant has now been civilly committed on three separate occasions, and each confinement was less than five months in duration. In light of his fixed delusions and the testimony indicating the likelihood of their persistence, the prospect of his again being subjected to commitment proceedings is not so remote as to bar the present challenge. A similar case is Friend v. United States, 128 U.S.App.D.C. 323, 388 F.2d 579 (1967), where an inmate at Saint Elizabeths appealed from a district court ruling which revoked his conditional release. While the appeal was pending, a subsequent conditional release was issued. After considering testimony similar to that presented here concerning the likelihood of future mental problems, the court found the appeal to be the subject of a continuing controversy and not moot. See also Matthews v. Hardy, 137 U.S.App.D.C. 39, 420 F.2d 607 (1969), cert. denied, 397 U.S. 1010, 90 S.Ct. 1231, 25 L.Ed.2d 423 (1970), and Women Strike for Peace v. Hickel, 137 U.S.App.D.C. 29, 420 F.2d 597 (1969). Moreover, we cannot be oblivious to the importance of the constitutional issue posed nor the number of persons who are affected. Having recognized additionally that appellant was released on each occasion in far less time than necessary to perfect an appeal, we heed with care the words of the Supreme Court in So. Pac. Terminal Co. v. ICC, 219 U.S. 498, 515, 31 S.Ct. 279, 283, 55 L.Ed. 310 (1911), that consideration of issues of public importance “ought not to be, as they might be, defeated, by short term orders, capable of repetition, yet evading review . . . .’’See also Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705, 35 L.Ed.2d 147 (1973); Moore v. Ogilvie, 394 U.S. 814, 89 S.Ct. 1493, 23 L.Ed.2d 1 (1969); United States v. W. T. Grant Co., 345 U.S. 629, 632-633, 73 S.Ct. 894, 97 L.Ed. 1303 (1953); Walling v. Helmerich & Payne, 323 U.S. 37, 43, 65 S.Ct. 11, 89 L.Ed. 29 (1944); and Alton & So. Ry. v. International Ass'n of Mach. & A. W., 150 U.S.App.D.C. 36, 463 F.2d 872, 878-879 (1972).
There is yet another independent reason why the present appeal is not moot —the collateral consequences of being adjudged mentally ill remain to plague appellant. We recently had occasion to consider whether the standard applied in criminal cases, that a “case is moot only if it is shown that there is no possibility that any collateral legal consequence will be imposed on the basis of the challenged conviction,” Sibron v. New York, 392 U.S. 40, 57, 88 S.Ct. 1889, 1900, 20 L.Ed.2d 917 (1968) (emphasis added), is applicable to contested civil commitment adjudications. We answered in the affirmative relying upon the multitude of legal disabilities radiating from the label “mentally incompetent.” Justin v. Jacobs, 145 U.S.App.D.C. 355, 449 F.2d 1017, 1018-1020 (1971).8 Cf. Hudson v. Hardy, 137 U.S.App.D.C. 366, 424 F.2d 854 (1970). For example, while the commitment stands on the record, the party may face state constitutional and statutory restrictions on his voting rights;9 restrictions on his right to *63serve on a federal jury;10 restrictions on his ability to obtain a drivers license;11 and limitations on his access to a gun license.12 Moreover, it is still commonly held that an adjudication of mental incompetency gives rise to a rebuttable presumption of continued incompetency.13 In many jurisdictions such a presumption may, in turn, affect the right to dispose of property, to execute contracts and to perform many similar and commonplace functions. Indeed, such an adjudication, while not always crippling, is certainly always an ominous presence in any interaction between the individual and the legal system. Such evidence will frequently be revived to attack the capacity of a trial witness. Depending upon the diagnosis, it may be admissible for impeachment purposes. Indeed, even in a criminal trial it may be available to attack the character of a defendant if he has put character in issue.14 Most significantly, records of commitments to a mental institution will certainly be used in any subsequent proceedings for civil commitment,15 a factor which may well have been influential in the present case.
*64We fully reject the government’s argument that any collateral consequences which might emanate from the present commitment would remain, even were we to reverse, because of the two prior commitments. Initially, we note that with respect to the proceedings leading to the earlier commitments there is no indication that there were jury trials, that any oral testimony was presented, or even that Mr. Ballay was present.16 Thus the instant case first presents for appellant the aura of a jury determination. We reject the argument for a more fundamental reason, however, which was aptly summarized by the Court in Sibron v. New York, supra, 392 U.S. at 56-57, 88 S.Ct. at 1899 (1968) when an identical argument was faced:
[W]e see no relevance in the fact that Sibron is a multiple offender. . A judge or jury faced with a question of character, like a sentencing judge, may be inclined to forgive or at least discount a limited number of minor transgressions .... It is impossible for this Court to say at what point the number of convictions on a man’s record renders his reputation irredeemable. And even if we believed that an individual had reached that point, it would be impossible for us to say that he had no interest in beginning the process of redemption with the particular case sought to be adjudicated. [I]t is far better to eliminate the source of a potential legal disability than to require the citizen to suffer the possibly unjustified consequences of the disability itself for an indefinite period of time before he can secure adjudication ....
(Emphasis added; footnotes omitted).
III. THE STANDARD OF PROOF
A. Introduction
The constitutional guarantee of procedural due process is not in any sense absolute in application but rather is relative to the interests at stake and the corresponding rights, constitutional or otherwise, of the respective parties. Unfortunately the rights and interests of human beings sought to be civilly committed have been considered all to infrequently. Senator Ervin, when convening a series of hearings on the constitutional rights of the mentally ill, commented about the deficiency:
Certainly the loss of freedom, of property, and of civil and personal rights solely because of mental illness is a process which should disturb every American concerned with the blessings of liberty. It is tempting to say that the problems of the mentally ill and their families are of no concern to the rest of the population.17
Later, referring to previous Congressional hearings, he stated that “the testimony we heard in 196118 supported the charge that this was one of the most neglected areas of American law — neglected by most private citizens, by the courts, by state legislatures and generally by politicians.”19 An enlightened attitude is not unique to legislators. A recent American Bar Foundation study begins modestly by stating that “[o]ne of the most troublesome problems in the field of law concerns the legal status of *65mentally disabled persons.” 20 The ambivalence alluded to is particularly surprising when statistics and predictions of the last decade have so strikingly portrayed the magnitude of the problem. It has been estimated that one out of every ten 21 or twelve children born in this country will at one time in his life be treated in a mental institution, and an even larger portion of our population will be subject to some form of mental care.22 Analyzed somewhat differently, one in three American families is likely to be disrupted by the commitment of one of its members.23 The absolute figures are equally striking. In 1966 there were approximately 580,000 resident patients in public and private24 mental hospitals and an additional 557,000 receiving care through outpatient clinics. Nearly 922,000 were admitted to the inpatient clinics and nearly 630,000 to the outpatient,25 a substantial majority of which were involuntary.26 This approximates if not exceeds the number of criminals sentenced and institutionalized in the United States during the same period.27
The legal problems of those sought to be adjudged mentally ill are not in any sense confined to procedural matters. In Jackson v. Indiana, 406 U.S. 715, 92 S.Ct. 1845, 32 L.Ed.2d 435 (1972), while considering the due process rights of an individual found incompetent to stand trial and summarily committed until his "sanity” returned, the Court reflected on the “broad power to commit persons found to be mentally ill” which has traditionally been exercised:
Considering the number of persons affected, it is perhaps remarkable that the substantive constitutional limitations on this power have not been more frequently litigated.
406 U.S. at 737, 92 S.Ct. at 1857 (footnotes omitted). Indeed, it may not be totally inaccurate to observe that the recent surge of interest in civil commitment28 may occasionally focus on procedure to the ultimate detriment of substance.29 Although we are not now *66called upon to resolve any such problems, a brief examination of the relevant substantive provisions is unavoidable if we are to accurately discern the interests reflected in the present legislation as written and in its practical operation, see Jackson v. Indiana, supra, 406 U.S. at 737-738, 92 S.Ct. 1845, in order to relate them to the significant interests at stake for the individual.
B. Analytical Framework
The analysis subscribed to by the Supreme Court in Morrissey v. Brewer, 408 U.S. 471, 92 S.Ct. 2593, 33 L.Ed.2d 484 (1972), provides the framework for our deliberation. Petitioners had challenged in a habeas corpus proceeding the revocation of their paroles without a hearing. The issues as characterized by the Court were first, whether due process applied to the parole system and second, if so, what process was due. There can no longer be any doubt that the nature of the interests involved when a person sought to be involuntarily committed faces an indeterminate and, consequently, potentially permanent loss of liberty and privacy accompanied by the loss of substantial civil rights (the loss of which frequently continues even if his liberty is restored) 30 is “one within the contemplation of the ‘liberty and property’ language of the Fourteenth Amendment.” Morrissey v. Brewer, supra, 408 U.S. at 481, 92 S.Ct. at 2600. See Jackson v. Indiana, supra; Murel v. Baltimore City Criminal Court, 407 U.S. 355, 358-365, 92 S.Ct. 2091, 32 L.Ed.2d 791 (1972) (Douglas, J., dissenting); Williams v. Robinson, 139 U.S.App.D.C. 204, 432 F.2d 637 (1970); Heryford v. Parker, 396 F.2d 393 (10th Cir. 1968); and Lessard v. Schmidt, 349 F.Supp. 1078 (E.D.Wis., 1972, No. 71-C-602). Cf. Specht v. Patterson, 386 U.S. 605, 87 S.Ct. 1209, 18 L.Ed.2d 326 (1967); In re Barnard, 147 U.S.App.D.C. 302, 455 F.2d 1370 (1971); Rouse v. Cameron, 125 U.S.App.D.C. 366, 373 F.2d 451 (1966); and In re Gault, 387 U.S. 1, 87 S.Ct. 1428, 18 L.Ed.2d 527 (1967). Hence we focus our attention on the second part of the Morrissey analysis which addresses what process is due.
[D]ue process is flexible and calls for such procedural protections as the particular situation demands. “[ Consideration of what procedures due process may require under any given set of circumstances must begin with a determination of the precise nature of the government function involved as well as of the private interest that has been affected by governmental action.”
408 U.S. at 481, 92 S.Ct. at 2600. The Court noted that the state, having already found the parolee guilty of a crime against the people which justified imposing extensive restrictions upon his liberty, recognized that his premature release on parole posed a risk that he would engage in additional antisocial conduct. The state therefore had an overwhelming interest in being able to incarcerate the individual without the burden of a full adversary criminal trial if he had, in fact, breached the conditions of his parole. On the other side of the coin was the liberty of the parolee, albeit conditional, which was of obvious value to him as well as to society.31 Finally, society has an interest in basic fairness which serves to enhance the *67chance of rehabilitation by avoiding adverse reaction to arbitrariness. Rejecting the argument that parole is so totally discretionary that some form of hearing would be administratively intolerable, the Court concluded that there could be no state interest in revoking parole without some basic informal procedural safeguards.32
C. Application of the Proof Beyond a Reasonable Doubt Standard to Involuntary Civil Commitment
The case of John Ballay is much more convincing in several respects than that of petitioners in Morrissey. At the heart of both, of course, are society’s tolerance for error in the factfinding process and the liberty of the individuals involved. The latter, however, is the fulcrum on which the paramount distinction rests: Ballay stands to lose the substantial liberty which the constitution ensures for all citizens, not the “conditional” or restricted liberty afforded the parolee. A related consideration involves the system by which restraints are justified. For the parolee, it has already been determined, beyond a reasonable doubt, that he committed specific acts which trigger societal sanctions. “Given the previous conviction and the proper imposition of conditions, the state has an overwhelming interest in being able to return the individual to imprisonment without the burden of a new adversary criminal trial if in fact he has failed to abide by the conditions of his parole.” Morrissey v. Brewer, supra, 408 U.S. at 483, 92 S.Ct. at 2601, (emphasis added). In the present case, on the other hand, we are examining that very trigger, one which is far from shorn in the criminal system but, quite the contrary, one which has historically been scrutinized meticulously and squeezed gingerly and only in compliance with carefully prescribed standards. Ballay has never been convicted of a crime nor was he ever apprised that certain conduct on his part might lead to confinement. The state’s interest is therefore not “overwhelming” in the sense that it was in Morrissey. Moreover, the state is not faced with “the burden of a new adversary trial” or any repetition of efforts as it is in the parole system. Finally, the due process right involved in the present case, the standard of proof beyond a reasonable doubt, see In re Winship, 397 U.S. 358, 90 S.Ct. 1068, 25 L.Ed.2d 368 (1970), belies even the unconvincing administrative burden argument since it presents no additional burden in the sense that the preparation of notices and disclosure of evidence entail additional paperwork, or that rights of confrontation and cross-examination significantly expand the time required for a hearing. It seeks merely to assure convincing evidence before an individual is deprived of “an interest of transcending value” — his liberty. Speiser v. Randall, 357 U.S. 513, 525, 78 S.Ct. 1332, 2 L.Ed.2d 1460 (1958). See also Humphrey v. Cady, 405 U.S. 504, 509, 92 S.Ct. 1048, 31 L.Ed.2d 394 (1972).
1. State Interest
We proceed now to an analysis of the interests involved when the state seeks to involuntarily commit one alleged to be mentally ill. We turn first to the statutory language as an indication of its purpose. 21 D.C.Code § 545(b) (1967) provides in pertinent part:
If the court or jury finds that the person is mentally ill and, because of that illness, is likely to injure himself *68or other persons if allowed to remain at liberty, the court may order his hospitalization for an indeterminate period, or order any other alternative course of treatment which the court believes will be in the best interests of the person or of the public.
The statute thus suggests two entirely different interests, making no distinction in terms of verdict, record for appeal or placement of the accused. Notably absent is a third possible criteria which addresses a third, but related interest.
To the extent that a person who is mentally ill and, because of that illness, likely to injure others may be hospitalized for an indefinite period or, alternatively, treated in any fashion which the court deems to be in the best interests of the public, the civil system is most felicitously analyzed by comparison with the criminal model. While the two obviously differ in some respects, they possess one overwhelming similarity: the result of successful prosecution may be loss of liberty, frequently in its most absolute sense through confinement. The criminal and civil commitment systems are most accurately juxtaposed on two distinct levels, one involving the decision to' convict or commit, and the other involving what the state may and should do once that decision is reached. For the moment we shall direct our attention to the latter consideration.33 Generally speaking, the goals of the criminal system are accomplished through incarceration since it presumably deters, rehabilitates, and physically removes.34 A fourth consideration, retribution, while “no longer the dominant objective of the criminal law,” Williams v. New York, 337 U.S. 241, 248, 69 S.Ct. 1079, 1084, 93 L.Ed. 1337 (1949), can be of no relevance whatsoever to involuntary civil commitment because no act deserving of retribution has occurred.35 Similarly, deterrence would seem at best to fill a modified role in the civil commitment system.36 While it is true that individuals sought to be committed may have “meaningful elements of responsibility,” cf. United States v. Brown, 155 U.S.App.D.C. 402, *69478 F.2d 606 (1973, No. 24,646), which would permit them to react to the institutionalization of others or to prior institutionalizations themselves, the very definition of “mental illness” 37 suggests that many would not be responsive.38 To the extent that the goal is to rehabilitate or treat the individual, there is theoretically no difference between the criminal and civil commitment models. Furthermore, rehabilitation cannot be viewed narrowly as a beneficent goal that works only one way. Certainly it is in the interest of the individual if it permits him to return a free man capable of functioning in society; but it is also in the interest of society because it reduces the risk of future antisocial conduct, relieves a financial burden, and contributes to society ill a more abstract sense by restoring a member to normal and useful life within the law. See Morrissey v. Brewer, supra, 408 U.S. at 484, 92 S.Ct. 2593.
The final justification, society’s desire to remove dangerous persons from the community, is perhaps the most convincing in terms of extending additional safeguards to civil commitment proceedings. In the criminal model, an individual can be “removed” only after it has been unequivocally shown that he voluntarily engaged in antisocial conduct proscribed by statute so as to warrant the imposition of criminal sanctions. Without further analysis this represents a significant distinction, since the civil inmate may be committed because predictably dangerous. He is not even afforded the benefit, fundamental as it may seem, of knowing what conduct may lead to his confinement. A more telling examination reveals only more incongruity. Even though the safeguards recognized-above are now implicit in the criminal system, legislatures and courts, unsatisfied, have been relentless guardians of individual rights, interjecting additional bulwarks in a system which effectuates such a substantial state interference. It is safe to say that no other process in our recent history has exacted such scrutiny. It becomes quite difficult in this light to justify the minimal protections available to a would be committee who may be restrained similarly if “dangerous,” particularly where it is apparent that the traditional justifications for confinement are partially inapposite. Finally, we note in this strain that for the criminal the desire of society to punish for conduct deemed unacceptable is characteristically tempered by determinate statutory sentences, see Jackson v. Indiana, supra, 406 U.S. at 737, 92 S.Ct. 1845, unlike the civil commitment model which permits indeterminate confinement.
The statute alternatively provides for commitment of an individual who is mentally ill and, because of that illness, likely to injure himself. A person falling within that circumscription may also be hospitalized for an indeterminate period or the court may order any other alternative course of treatment which it believes will be in the best interests of the person. Since this individual poses no danger to society, society’s interference must be justified on the basis of the state’s status as parens patriae. It is clearly recognized that the state may *70act in this capacity and that relaxed procedures may be justified in certain circumstances, see, e. g., Kent v. United States, 383 U.S. 541, 554-556, 86 S.Ct. 1045, 16 L.Ed.2d 84 (1966), and, indeed, it has been suggested that a duty to so act may occasionally exist. Unlike the British model, however, it is not the historical basis for intervention in the United States. It was not until the mid or late nineteenth century that therapy took its place beside detention in the American model.39 Of even more recent vintage is the realization that substantial deprivation of constitutional rights justified largely on behalf of the individual must be accompanied by a corresponding right to treatment. See, e. g., In re Curry, 147 U.S.App.D.C. 28, 452 F.2d 1360 (1971).40 Without some form of treatment the state justification for acting as parens patriae becomes a nullity. Yet even without considering the issue upon which the parens patriae rationale must either stand or fall — whether meaningful treatment is available, see Kent v. United States, supra, 383 U.S. at 555, 86 S.Ct. 1045 and In re Curry, supra—we note several problems in the current statute which seem to undercut the apparent justification. Consider the individual who is untreatable in the present state of medical science. Perhaps all that can be done for him under institutional tutelage is to care for his physical needs and/or subdue him through use of drugs.41 Unless the right to treatment be interpreted to include ineffective treatment, an anomaly in itself, the parens patriae rationale would seemingly fail and indeterminate institutionalization would necessarily be *71a purely custodial function and justifiable only by other considerations. See Jackson v. Indiana, supra, 406 U.S. at 737-738, 92 S.Ct. 1845; but cf. Robinson v. California, 370 U.S. 660, 82 S.Ct. 1417, 8 L.Ed.2d 758 (1962). Yet such an individual can presumably be committed under the statute in question and with minimal procedural safeguards.42 The problem again stems from a statutory failure to isolate the interests sought to be served or, stated conversely, by failure to accurately identify the class of persons at which the statute is directed. Another example referred to frequently is that of an old man living alone in the country who, although typically lucid, is subject to severe fits of depression with accompanying suicidal tendencies. It may be accurately predicted that he will ultimately take his own life under the influence of such fits, yet he is well aware of that and chooses to spend the remainder of his life at home and free.43 It would seem that this class of individuals represents interests in addition to those considered earlier, cf. Roe v. Wade, 410 U.S. 113, 93 S.Ct. 705, 35 L. Ed.2d 147 (1973) yet it, too, may fall within the present statute. We note also the ambiguity in the phrase “likely to injure himself.’’ Literally this encompasses injury of an active sort ranging from intentional acts such as suicide, to purely unintentional acts, such as ignorantly crossing the street against the light, to purely passive injury, such as disease which results from inability to secure gainful employment and furnish the minimum nutritional and hygenic requirements of life. Again we find ourselves straying into different, yet related societal interests, certainly of equal importance though for different reasons. Cf. Goldberg v. Kelly, 397 U.S. 254, 90 S.Ct. 1011, 25 L.Ed.2d 287 (1970).
In attempting to discern and analyze the state interests involved and the primary zones of legislative concern it is, of course, implicit that we consult the history of the relevant legislation. The most recent comprehensive legislation enacted in 1964, does not refute the preceding analysis. The goals of the new legislation were broadly summarized by Senator Ervin, a co-sponsor and principal draftsman, before the Senate:
[O]ur concern has been with hospitalization procedures, with the protection of the rights of patients after, as well as before, they enter the hospital, and with the encouragement of voluntary admissions so that a problem of serious national scope may the sooner be eradicated. Our concern has been to assure that when an individual is deprived of his liberty because he is mentally ill, he will receive appropriate attention and the treatment necessary to restore him to his place in society.
110 Cong.Rec. 21346 (1964).44 The primary thrust of the legislation, to secure *72civil and constitutional rights for involuntary civil committees,45 is certainly not disturbed — indeed, is fostered — by our holding today. Indeed, it is likely that the legislature anticipated judicial assistance in effectuating these goals, particularly in areas, such as the burden of proof, which have traditionally been judicial. See Woodby v. Immigration Service, 385 U.S. 276, 286, 87 S.Ct. 483, 17 L.Ed.2d 362 (1966). Consider, for example, the following response of Senator Ervin to Dr. Davis, then vice-president of the National Association of State Mental Health Program Directors, during testimony by the latter:
I think there is some difficulty from a legislative standpoint in getting the legislative body to take too big a dose of legislation at a time. Also, if you can take such revision in small doses, your doses will have a little bit more therapeutic value from both the standpoint of the patient and also from the standpoint of legislation.
1963 Senate Hearings, supra note 8, at 111.
More specifically, the objectives which absorbed the bulk of legislative energy —the encouragement of voluntary commitment, the separation of commitment from legal incompetency, and the availability of prompt and meaningful treatment to those committed — are largely untouched by our decision. They are more properly considered to be parallel and complimentary lines of development, and in most senses not contradictory.
There are two items of legislative history which more directly illuminate our consideration of state interests. The first concerns the frequent conflict between legal and medical testimony involving the possible therapeutic disadvantages of formal and complicated commitment proceedings, a problem which will be temporarily deferred. The second involves the suggestion frequently tendered that a third criteria be added as a ground for civil commitment: a provision for hospitalization of persons who, because of mental illness, lack sufficient capacity to make responsible decisions with respect to hospitalization.46 The suggestion was rejected, and several potential grounds for so doing are conceivable. It may be the legislature concluded that sufficient treatment guarantees were not currently available to justify such a massive curtailment of rights solely on behalf of the individual. This interpretation would favor substantial procedural safeguards for the individual committed under the statute as enacted, even where he was found to be a danger only to himself. Another possible explanation was summarized by Dr. Davis during Senate hearings when explaining why New Jersey had recently *73excluded a similar provision from its legislation :
Our legislation . . . does not specifically state this on the basis that our commission felt that this very condition in itself constituted a peril to the patient, and when the physician determined that the individual lacked this sufficient capacity [to make a responsible decision], this could be the basis for the physician feeling that unless this individual were hospitalized he would be dangerous to himself.
1963 Senate Hearings, supra note 8, at 106. If this were the basis for rejecting the proposal it would add little to our deliberation since the interests involved have already been reviewed in the context of involuntary commitment on the grounds of dangerousness to self.47 The most convincing explanation in light of the legislative history, however, is that Congress hoped to encourage individuals in this third category to seek voluntary treatment.48 One implication of the latter interpretation is that, while the legislature was concerned with the care and treatment of the benign mentally ill, its primary concern was with those who, because of their illness, posed a threat to society. On balance, the thrust may well have been preventative more than therapeutic.
2. Analysis of the Standard Itself
The process due depends upon the relative interests involved and we have seen that two entirely distinct state interests are involved in the involuntary civil commitment system. It also depends upon the nature of the procedure itself and the function it performs. For example, the right to counsel may be perceived to fulfill different functions in different proceedings or even at different stages of the same proceeding. Broadly stated, the standard of proof reflects the risk of winning or losing a given adversary proceeding or, stated differently, the certainty with which the party bearing the burden of proof must convince the factfinder. “There is always in litigation a margin of error in representing error in factfinding, which both parties must take into account.” Speiser v. Randall, 357 U.S. 513, 525, 78 S.Ct. 1332, 1342, 2 L.Ed.2d 1460 (1958). We find the preponderance of the evidence standard applied to most civil proceedings where the stakes are frequently economic and where “we view it as no more serious in general for there to be an erroneous verdict in the defendant’s favor than for there to be an erroneous verdict in the plaintiff’s favor.” In re Winship, 397 U.S. 358, 371, 90 S.Ct. 1068, 1076, 25 L.Ed.2d 368 (1970) (Harlan, J., concurring). On the other hand, in situations where the various interests of society are pitted against restrictions on the liberty of the individual, a more demanding standard is frequently imposed, such as proof by clear, unequivocal and convincing evidence. See, e. g., Woodby v. Immigration Service, 385 U.S. 276, 87 S.Ct. 483, 17 L.Ed.2d 362 (1966) and Chaunt v. United States, 364 U.S. 350, 81 S.Ct. 147, 5 L.Ed.2d 120 (1960). Cf. Rosenbloom v. Metromedia, Inc., 403 U.S. 29, 50-52, 91 S.Ct. 1811, 29 L.Ed.2d 296 (1971). Finally, where the interests of society are pitted against an individual who “has at stake an interest of transcending value — as a criminal defendant his liberty — this margin of error is reduced as to him by the process of placing on the other party the burden of persuading the factfinder at the conclusion of the trial of his guilt beyond a reasonable doubt.” Speiser v. Randall, supra, 357 U.S. at 525-526, 78 S.Ct. at 1342. This result is actuated in the latter case, as eloquently expressed by Mr. Justice Harlan, because “we do not view the social disutility of convicting an innocent man as equivalent to the disutility of acquitting someone who is guilty. . . . [This conclusion is] bottomed on a fundamental value determination of our society that it is far *74worse to convict an innocent man than to let a guilty man go free:” In re Winship, supra, 397 U.S. at 372, 90 S.Ct. at 1076. Closer inspection of the standard reveals several implications which favor its use in involuntary civil commitment proceedings.
First, “a society that values the good name and freedom of every individual” has a substantial and fundamental interest in assuring accuracy for reasons applicable to involuntary civil commitment :
[T]he reasonable-doubt standard is indispensable to command the respect and confidence of the community in applications of the criminal law. It is critical that the moral force of the criminal law not be diluted by a standard of proof that leaves people in doubt whether innocent men are being condemned. It is also important in our free society that every individual going about his ordinary affairs have confidence that his government cannot adjudge him guilty of a criminal offense without convincing a proper factfinder of his guilt with utmost certainty.
In re Winship, supra, 397 U.S. at 363-364, 90 S.Ct. at 1072.
Second, we note that imposition of a higher standard of proof is unlikely to have any discernible effect upon the proceedings themselves, see In re Winship, supra, 397 U.S. at 365-366, 90 S.Ct. 1068, since the proceedings are already similar in many respects to a criminal trial, the person sought to be committed having the right to counsel49 and to trial by jury,50 among others. While it may result occasionally in an expanded hearing, where the government chooses to muster additional evidence, “[t]he constitutional guarantee demands no less if the search for truth is not [to] be sacrificed to administrative speed and convenience.” Menechino v. Warden, 27 N.Y.2d 376, 318 N.Y.S.2d 449, 267 N.E.2d 238, 242 (N.Y.1971). The lack of significant physical impact caused by a revised standard of proof is important for another reason. Many medical experts are of the view that substantial procedural safeguards may be detrimental to the patient. Merely being subjected to the hearing, they argue, is a traumatic experience that may aggravate the condition of certain mental illnesses. The time involved in the procedures impedes and delays treatment. Finally, a full adversary proceeding similar to a criminal trial forces ' many to believe they are being incarcerated and stigmatized rather than hospitalized.51 The trauma argument can be of no avail where an individual is sought to be committed because dangerous to others in light of our earlier conclusion that the promise of treatment can serve no more adequately as a justification for the relevant interests of society than it does in the criminal model. Thus, we focus our attention on its application where the individual is to be committed because dangerous to himself on the ground that society can treat him.
Delay in treatment is now a minimal problem because the act permits temporary retention and treatment prior to hearing before a judge or jury52 The trauma that may unfortunately result is attributable to the public exposure, the testimony presented, the witnesses and the general adversary character of the hearing, and would seem to be wholly unaffected by the burden of proof. The individual’s perception of the proceeding *75as identical to a criminal trial is indeed unfortunate but the problem is fundamental to this and analogous opinions. See, e. g., In re Winship, supra. Moreover, this argument presupposes that an individual will ultimately be committed, the very proposition that we now reappraise. Tempering the problem is the fact that one of the primary purposes of the new legislation was to encourage voluntary admissions,53 and exemplary provisions were enacted to that end, including a provision directed at the situation where an individual is unwilling for one reason or another to commit himself, but is willing to submit when the initiative is taken by another. See 21 D.C.Code § 513 (1967) (“Hospitalization of nonprotesting persons”). We also note that the choice of avoiding much of the potential trauma accompanying a heated involuntary commitment proceeding remains in the control of the individual.54 Finally, we reiterate that our concern is with possible involuntary hospitalization of an individual who is not mentally ill and/or dangerous.
The mental condition of one whose mind is so deranged as to require imprisonment for his own and others’ good is indeed pitiable. But the mental attitude of one who is falsely found insane and relegated to life imprisonment is beyond conception. No greater cruelty can be committed in the name of the law.
5 J. Wigmore, Evidence § 1400 (3d ed. 1940).55 This consideration certainly weighs heavily against the setbacks occasionally fostered by a traumatic adversary proceeding.
The third consideration involves the very nature of the evidence presented and it is here that the positive impact of a higher standard of proof on the proceedings is perhaps most clearly visible. First, the factfinder must be convinced that the individual is “mentally ill”— i. e. that he is inflicted with a “psychosis or other disease which substantially impairs” his mental helath. 21 D.C. Code § 501 (1967).
One does not have to echo the scepticism uttered by Brian, C. J., in the fifteenth century, that “the devil himself knoweth not the mind of men” to appreciate how vast a darkness still envelopes man’s understanding of man’s mind. Sanity and insanity are concepts of incertitude. They are given varying and conflicting content at the same time and from time to time by specialists in the field. Naturally there has always been conflict between the psychological views absorbed by law and the contradictory' views of students of mental health at a particular time.
Leland v. Oregon, 343 U.S. 790, 803, 72 S.Ct. 1002, 1009, 96 L.Ed. 1302 (1952) (Frankfurter and Black, JJ., dissenting). This is not to suggest that the *76problems engendered by such evidence are not susceptible to proof, particularly by analogy to state of mind in the criminal law. See Murel v. Baltimore City Criminal Court, supra, 407 U.S. at 364, 92 S.Ct. 2091 (Douglas, J., dissenting). However, in the criminal model the factfinder frequently has the benefit of presumptions and inferences to assist him in determining state of mind. He is also faced, in most instances, with much more concrete evidence of conduct from which he may derive mens rea. The absence of these factors in commitment proceedings is presumably offset by the presence of expert testimony. Unfortunately, even in light of great recent advances in disciplines such as psychiatry, and even acknowledging this to be the best diagnostic and predictive device currently available, it is far from satisfactory. “The only certain thing that can be said about the present state of knowledge and therapy regarding mental disease is that science has not reached finality of judgment . . . . ” Greenwood v. United States, 350 U.S. 366, 375, 76 S.Ct. 410, 415, 100 L.Ed. 412 (1956).56 That “mental illness” is by no means a static concept but rather one of degree of deviation from the “norm” is illustrated by the very ambiguity in the definition which embraces “a psychosis or other disease which substantially impairs mental health.” 57 Finally, we are forced to consider the practical limitations which have hindered psychiatric testimony. Although three psychiatrists testified at Ballay’s judicial hearing following his three month hospitalization pending trial, one had been with him for a total period of less than two hours, and another for a total of merely one hour.58 This is not particularly surprising in light of the limited personnel available.59
Even more problematic is proof of dangerousness to self or others. Remaining largely undefined, application of these concepts by judge and jury may unduly reflect clinical definitions and conclusions rather than the appropriate judicial exegetics and community values. See United States v. Ashe, 155 U.S.App.D.C. 457, 463-465, 478 F.2d 661, 667-669. Cf. Washington v. United States, 129 U.S.App.D.C. 29, 390 F.2d 444 (1967). Since most “mental illness” is defined and measured in terms of behavioral deviance, to classify and label it as a basis for subsequent testimony predictive of future behavior presents somewhat of an anomaly in itself.60 It is at least questionable in many cases exactly how much this differs from a prediction *77that 80-90% of the individuals featuring specified socio-economic traits and backgrounds are destined to commit crimes, then using this profile to rehabilitativeIy incarcerate, presumably an unacceptable procedure both morally and constitutionally.61
There is little reason to suspect that persons classified as “mentally ill” tend to be more dangerous to society than others. A study of 5,000 patients discharged from New York State mental hospitals over a five and one-half year period showed that “ ‘patients with no record of prior arrest have a strikingly low rate of arrest after release. . Their overall rate is less than V12 that of the general population, and the rate for each separate offense is also far lower, especially for more serious charges.’ ”62 The fact is that predictions of dangerous behavior, by psychiatrists or others, have never been characterized by a high degree of accuracy. See Murel v. Baltimore City Criminal Court, supra, 407 U.S. at 364-65 n. 2, 92 S.Ct. 2091 (Douglas, J., dissenting). That this phenomenon applies equally to “dangerous to self” is amply illustrated by the testimony in the present ease. The first psychiatrist to testify for the government, upon being probed by counsel as to why Ballay would be dangerous to himself if released, responded:
He will be existing like vagabond type. He will be wandering from state to state, town to town, picking up anything in the trash that he will eat or something.63
Mr. Ballay cannot hold any job. He cannot provide for himself, and he will just be eating out of maybe the trash cans or something like that.64
Similar questions presented to the second government psychiatrist produced the following responses:
[Tjhere is a more than reasonable likelihood that the delusions that he had which brought him to Washington and to the Capitol and to the White House gate on three occasions, will return and that he may respond appropriately to those delusions or false beliefs.
[Counsel]: What do you mean by respond appropriately?
[A.] If a person believes he is a U. S. Senator, he will act like one.65
[Q.] How will he be a danger to others if he is released ?
[A.] Only if his false beliefs lead him to some antisocial act. I cannot be specific in that.66
This was in sharp contrast to the other evidence presented, which showed that Ballay was cooperative and very helpful in the ward, attested to by, inter alia, his unaccompanied ground privileges at the hospital, that he had never committed a crime in his life nor could it be shown that he ever committed even an act of violence. Finally, a psychiatrist testifying in his behalf indicated that while Ballay apparently nurtured a belief which was in sharp contrast with reality, it was “essentially a benign belief . . . that is not violent in nature. . . . Persons of this kind . . . are no more likely than any one else in the population to be dangerous ,”67 Neverthless, Ballay was *78hospitalized for an indefinite term. While a more rigorous standard of proof may not allay infirmities in substantive statutory elements it certainly may, and the reasonable doubt standard is designed particularly to, partially offset them by reducing the risk of factual error.
3. The Interest of the Individual
All that remains to be seen is whether there is some individual interest which distinguishes civil commitment from incarceration sufficiently to justify assuming a greater risk of wrongful deprivation of liberty in the latter. One argument occasionally raised is that the individual who is committed may benefit from institutionalization while the criminal will seldom do so.68 While this argument may have some merit in other areas, such as the exclusionary rule where the party released may have fallen within the statutory proscription, it has little bearing on the possible mistaken commitment of one who is not “mentally ill.” Moreover, offsetting this argument is the equally persuasive argument that institutionalization may have devastating side effects upon those committed, particularly where they were not initially in need of treatment.69
The second prong of the theory, that unlike the criminal model treatment will benefit those who are mentally ill but are mistakenly diagnosed as dangerous, remains to be considered. We first heed the warning of a prominent commentator in the criminal field:
Measures which subject individuals to the substantial and involuntary deprivation of their liberty are essentially punitive in character, and this reality is not altered by the facts that the motivations that prompt incarceration are to provide therapy or otherwise contribute to the person’s well-being or reform. As such, these measures must be closely scrutinized to insure that power is being applied consistently with those values of the community that justify interferences with liberty for only the most clear and compelling reasons.
Allen, Criminal Justice, Legal Values and the Rehabilitative Ideal, 50 J.Crim.L.C. & P.S. 226, 230 (1959) (emphasis added).70 This conforms to our earlier observation that the dominant objective of the current legislation appears to be the protection of society. Even ignoring such reasoning, an analysis of In re Winship, 397 U.S. 358, 90 S.Ct. 1068, 25 L.Ed.2d 368 (1970) dispels any lingering doubt that the argument must fail.
In Winship the Court held that where a juvenile is charged in a juvenile proceeding with an act that would constitute a crime if committed by an adult, proof beyond a reasonable doubt is among the essentials of due process and fair treatment required at the adjudicatory stage of proceedings. Rejecting a beneficent, parens patriae philosophy *79nearly identical to that espoused above, the Court found that the reasonable doubt standard was singularly neutral in terms of defeating the state policies at work in juvenile proceedings, as it is here, and was additionally persuaded by the interests of immense importance at stake for the accused: loss of his liberty through institutional confinement and the stigma of a finding that he violated the criminal law. As we have highlighted throughout, the loss of liberty — the interest of “transcending value” — 71 is obviously as great for those civilly committed as for the criminal or juvenile delinquent. Indeed, it may be greater in the former since the statute provides for •indefinite commitment. The only question is whether the “stigma” associated with involuntary civil commitment is as severe as the stigma of finding that an individual committed a crime. Even accepting recent medical advances, current studies clearly indicate the fallacy of contending that most people view mental illness as a disease similar to any physical ailment of the body.72 At best, an enlightened minority has been persuaded to accept this view. Presumably, however, that same enlightened minority would also recognize that a juvenile is not to be viewed with disdain and considered incorrigible for an indiscretion during his youth. Besides the social ostracism involved in a finding of mental illness, the very livelihood of the individual may be at stake, a bitter irony for persons such as B allay who were hospitalized to begin with partially on the basis of testimony indicating that the danger he posed to himself was an inability to support himself through gainful employment.73 The situation is aggravated since the discharged patient not only must cope with the stigma of having once been hospitalized, but he must often continue to cope with the “mental illness” label itself. Of the 384 patients judicially hospitalized at Saint Elizabeths in 1969 who were subsequently *80dismissed, only 3 were discharged as “recovered.” The bulk of the remainder were discharged with labels such as “improved” or “mentally ill — not a danger.” 74
Stigma is, of course, a double-edged blade, and the patient’s perception of the phenomenon is also important. Even in the best of institutions the patient forcefully committed suddenly faces the regimented routine of ward life and daily confrontation with state employees, however capable, rather than family and friends.75 These and other factors often cause him to demean himself and to magnify social ostracism.76 The very fact that the new legislation has served to convince only one-half those receiving treatment to do so voluntarily suggests that archaic images remain.77
In Winshvp, the Court concluded that while the consequences of being adjudged a juvenile delinquent were not identical to being adjudged a criminal, the differences were not sufficient to support a distinction in the standard of proof. This was despite the fact that, unlike involuntary civil commitment, being adjudged delinquent did not deprive the child of his civil rights nor did the statute, which called for confidentiality, expose him to the stigma of a public hearing. We cannot help but conclude that the forcefully committed civil patient has at stake interests of equivalent proportions.
For the reasons hereinbefore stated, we reverse on the ground that the jury was not instructed to find, and did not find, beyond a reasonable doubt that John Ballay was both mentally ill and dangerous.
So ordered.