United States Ex Rel. Martin v. Strasburg

513 F. Supp. 691 (1981)

UNITED STATES of America ex rel. Gregory MARTIN, Luis Rosario, Kenneth Morgan et al., Petitioners,
v.
Paul STRASBURG, as Commissioner, New York City Department of Juvenile Justice, Respondent.

No. 77 Civ. 6176 (RLC).

United States District Court, S. D. New York.

April 17, 1981.

Martin Guggenheim, Bruce J. Ennis, New York City, American Civil Liberties Union, Charles Schinitsky, Charles A. Hollander, Janet R. Fink, Brooklyn, N. Y., for petitioners.

Robert Abrams, Atty. Gen. of the State of N. Y., New York City, pro se; Judith A. Gordon, Kathleen Gill Miller, Asst. Attys. Gen., New York City, of counsel; Mary Ann B. Orenstein, Legal Asst.

OPINION

ROBERT L. CARTER, District Judge.

I

Statement of the Case

This habeas corpus class action proceeding is being brought on behalf of a class of all juveniles who are being held or who will be held before these proceedings are concluded in pretrial detention under N.Y. Family Court Act § 739(a)(ii) (McKinney) ("the Act"). Petitioners seek a declaratory judgment that § 739(a)(ii) violates the due process and equal protection clauses of the 14th Amendment. The case was certified as a class action in an unpublished opinion dated April 3, 1978.

In New York persons between the ages of 7 and 16 accused of various acts which would be a crime if committed by an adult are subject to the exclusive jurisdiction of the family court to be prosecuted as juvenile delinquents.^[1] In the exercise of its exclusive jurisdiction over juvenile delinquents, the family court is authorized under the Act to subject an alleged delinquent to pretrial detention prior to a probable cause or fact finding determination if it determines that "there is a serious risk that he may before the return date do an act which if committed by an adult would constitute a crime." § 739(ii). The youth may also be detained because of the substantial probability that he will not appear on the return date 739(a)(i), but this provision is not involved or challenged in this litigation.^[2]

A trial on the merits occurred in March, 1980. Edwin Rodriguez, a 16 year old, Juan Santiago, a 15 year old, and Jerome Basnight, a 17 year old, representatives of the class, testified in person. In addition, the files of their proceedings in the family court along with the § 739 case histories of 33 other purported members of the class were introduced in evidence. One of the case histories concerns a Vincent Harris who was not detained and accordingly is not properly a member of the class. The Attorney General contends that Victor Melendez is not properly a member of the class since he was detained pursuant to § 739(a)(i). A reading of the files does not support that contention, however. At the § 739 hearing of Melendez, the judge paraphrased the language of both § 739(a)(i) and § 739(a)(ii) as grounds for Melendez' detention. Accordingly, we will count him as properly included in the class. However, Daryl Larkin, whose case history is included in the class was detained because, in the judge's view, there was "a grave chance" that he would not return to court. (Exhibit 9A). That is the basis for detention under § 739(a)(i). In his brief the attorney general states that Larkin was detained under both § 739(a)(i) and (ii), but I read the transcript differently. Larkin, therefore, will not be considered a member of the class. Thus, the case histories which will be studied are those of 3 named petitioners and 31 other members of the class.

In addition there was testimony concerning the family court process by Steven Hiltz, an attorney for 9 years in the juvenile rights division of the Legal Aid Society, and attorney-in-charge of its Manhattan office from 1976-1979, Hubert Benjamin, supervisor of probation in Bronx Family Court since 1968 and a probation officer for 30 years, and Judge Cesar Quinones, of the Family Court of the City of New York since 1970, who testified concerning § 739 hearings from a judge's perspective.

Michael Bigley, director of detention services of the New York State Division for Youth and Ralph E. Kelly, deputy commissioner of operations of New York City Department of Juvenile Justice, testified concerning secure and non-secure facilities and the functions of the state and city agencies in the housing of juveniles remanded after a § 739 hearing. Professor Leslie Wilkins, professor of criminal justice at the State University of New York at Albany, and Dr. Lynwood David Zinn, Director of Child Psychology at Michael Reese Hospital in Chicago, Illinois, and clinical assistant professor at the University of Chicago, testified as experts for petitioners concerning the harmful effects pretrial detention inflicts on the young and the arbitrariness and untrustworthiness of predictions that a juvenile will commit a crime in the future. Dr. Wilkins questioned whether any reasoned predictive determination to that effect could be made by a family court judge on the basis of an intake interview and stated that the judge's subjective prognosis founded on such data would be roughly only 4% better than pure chance. Finally, Judge Margaret Driscoll of the Supreme Court of Connecticut and with 18 years on the bench testified about national juvenile standards and Connecticut procedures in respect of the pretrial detention of juveniles.

II

The § 739 Case Histories of the Named Plaintiffs and 31 Other Members of the Class

Edwin Rodriguez was arrested on December 12, 1977, and charged with arson and reckless endangerment (that he with other youths attempted to start a fire in a subway station). Rodriguez was 14 at the time and had no prior arrest record. He had a § 739 hearing on December 13, 1977, before Judge Ferrara. Probation recommended detention. He was detained on the grounds that he was charged with a serious act that occurred at 12:30 A.M., and he was not going to school. A fact finding hearing^[3] was scheduled for December 15, but was adjourned to December 16. At the December 16 hearing, Rodriguez entered a plea to reckless endangerment and was released. He had been detained for a total of 3 days. At the dispositional hearing^[4] on June 23, 1978, Rodriguez was given 2 years probation and family counselling. There have been no subsequent court contacts. Rodriguez, as indicated, testified in person at the trial, and his § 739 case history is set out in Exhibit 12.

Juan Santiago, 12 years old with no prior court contact, was arrested on April 11, 1978, and charged with first degree assault on a 71 year old woman. The offense was alleged to have occurred at 6:00 P.M. on the street and to have resulted in injury to the woman. He was taken to Spofford (a secure facility)^[5] where he was held until his § 739 hearing on April 12, before Judge Matthews. The court ordered him detained because of the seriousness of the offense and the condition of the alleged victim. A probable cause hearing was set for April 17. He had been detained for 6 days at the time of the probable cause hearing. Probable cause was found and the boy was detained at Spofford until his fact finding hearing on April 24, 1978. As a result of that hearing, he was adjudicated a delinquent and was ordered detained further at Spofford for psychological testing and investigation. At his dispositional hearing on May 19, 1978, he was released to the custody of his mother on 2 years probation. Since his release, he has had no further contact with the court. (Exhibit 13)

Jerome Basnight, 14 years old, was arrested on October 14, 1977, and charged with first degree robbery and criminal possession of a weapon. He is alleged to have acted with 2 others to steal a pocketbook at gunpoint. He was released to the custody of his mother by the police. A § 739 hearing was held on October 20, 1977, before Judge Ferrara, 6 days after the event. The boy was ascertained to be registered in school, to be a resident of Staten Island and to have no prior criminal record. The judge detained him because the matter was "quite serious," and the boy seemed involved in bad company. A probable cause hearing was held on October 24. He had been detained 4 days at the time of the probable cause hearing. Probable cause was found, and he remained in custody until a fact finding hearing on November 1. He was adjudicated a delinquent at that hearing on the robbery charge, and disposition was reserved. At the dispositional hearing on November 29, 1977, Basnight was placed on probation for one year under the supervision of the County of Richmond. (Exhibit 10) Basnight testified at trial that he was registered in McKee Vocational High School in Staten Island, and that because of the time spent in detention, he lacked the necessary points to qualify for auto shop training because he could not make up the deficiency.

Gregory Martin, 14 years old, was arrested on December 13, 1977, charged with robbery and assault with criminal possession of a weapon for striking another youngster with a loaded revolver and stealing his jacket and sneakers. A § 739 hearing was held on December 14, before Judge Ferrara. Martin had no prior record. Citing the lateness of the hour when the crime occurred, that the boy had given the police a false address and the loaded weapon, the judge ordered Martin detained. At the December 19 probable cause hearing, probable cause was established. He had been detained 6 days before this hearing. The fact finding hearing held December 27-29, adjudicated Martin a delinquent and guilty on all counts. At the February 14, 1978 dispositional hearing, he was given 2 years probation. (Exhibit 11)

Luis Rosario, 14 years old, together with 4 others, was charged with an attempt on March 2, 1979, to rob two men. It was alleged that a gun was used and that Luis and his companions beat the two men about the head with sticks. Luis was released. A § 739 hearing was held before Judge Roache on March 15. At that time Luis had another delinquency petition pending for knifing a student, and two prior petitions had been adjusted. The court ordered him detained, noting his prior court appearances. Probable cause was found on March 21. On April 11, the boy was released to his father, and the case was terminated on September 25, 1979, without adjustment. (Exhibit 2)

Kenneth Morgan, 14 years old, was arrested on March 25, 1978, and charged with attempted robbery and threatening a 14 year old girl and her brother. A § 739 hearing was held on March 27, before Judge Moskoff. Kenneth was then on release status on another robbery charge. He was ordered detained. A probable cause hearing was apparently set for March 30, but continued until April 4, where it seems to have been combined with a fact finding hearing. He was found guilty of harassment and petty larceny. Morgan was detained 10 days before the April 4 probable cause continued hearing, and 8 days subsequent to the § 739 determination. On July 5, at a dispositional hearing, he was ordered placed with the Department of Social Services for 18 months. (Exhibit 3)

Phil Hanna, 13, and Calvin Jenkins, 15, were arrested and charged with robbery and criminal trespass on August 23, 1977, for breaking into a store and damaging equipment. A § 739 hearing was held before Judge Rigler on August 24. Hanna had had prior court contacts and Jenkins, with 2 delinquency petitions pending, was on probation. Both were ordered detained. At the August 29 probable cause hearing, Jenkins and Hanna were paroled. On October 21, 1977, their cases were dismissed for lack of prosecution. They were held in pretrial detention for 6 days before the probable cause determination. (Exhibit 4)

James Price, 13 years old, was arrested on September 11, 1978, for attempting to pull a necklace from a man's neck and was charged with robbery, assault and grand larceny. A § 739 hearing was held before Judge Ferrara on September 12. Price had had prior court contacts, but the previous charges had not been pressed. He was ordered detained, the court citing his recent court contacts. A probable cause hearing was set for September 15, and at that hearing Price admitted to grand larceny and was placed in a non-secure facility. He absconded on September 27. At the time of the probable cause hearing Price had been detained for 3 days. (Exhibit 5)

Johnny McArthur, age 15, was arrested on October 12, 1977, for pointing a loaded, cocked automatic at a 13 year old. He was charged with criminal possession of a dangerous weapon and menacing and released by the police to the custody of his mother. A § 739 hearing was held on November 28, 1977, six weeks after his arrest, before Judge Heller. He was ordered detained, the judge stating it was his policy to remand those caught with loaded .45 caliber revolvers. The probable cause hearing was scheduled for November 30, 1977, but was adjourned, and he was released to the custody of his mother. As of November 30, he had been detained for 2 days. On January 10, 1978, the matter was dismissed without prejudice. (Exhibit 6)

Christopher Cox, 13, Glen Maloney, 13, and Wade Forde, 13, were arrested on October 13, 1976, along with Clarence Smith, 13, and Alex Michael, 15, and charged with grabbing and threatening people with a weapon in an attempt to take their bus passes. They were released to the custody of their parents. A § 739 hearing was held on November 12, before Judge Roache. Smith, who had retained private counsel, was paroled as was Michael. The court ordered Cox, Maloney and Forde detained despite the intake probation officer's recommendation of parole and the fact that the three had had no prior court contact. Forde was paroled to his parent's custody on November 15, and at the November 23 fact finding hearing he was adjudicated a delinquent. Forde was placed on probation for 2 years. Cox was placed on probation on January 13, 1977, but it was subsequently revoked. Forde appears to have been held in custody for 3 days before being paroled to his mother's care. The case history does not contain any further information. (Exhibit 7)

Francisco Ramos, 15, was arrested on March 4 for breaking into a store and stealing goods and charged with burglary, petty larceny, criminal mischief and criminal possession of stolen goods. He was released to the custody of his parents. A § 739 hearing was held on March 17, 1978, before Judge Roache. He had a record of 9 prior arrests but all had resulted in adjustments and dismissals. Probation recommended release, but Judge Roache ordered him detained on the ground that prior arrests showed a likelihood that Ramos would commit a crime before the return date. A probable cause hearing was set for March 21, 1978, but the matter was continued. The final disposition of the proceeding was a discontinuance. Ramos was apparently in custody by order of the court for some 4 days until the matter was discontinued. (Exhibit 8)

Jose Cruz, 14, was charged with attempted murder and assault based on a March 7, 1977 incident in which Cruz allegedly pushed a 77 year old man down the stairs, beat him about the head and threatened him. At a § 739 hearing on March 25, before Judge Rigler, he was ordered detained in view of the severity of the charge. Probable cause was found on March 28, at which time he had been detained 3 days. The attempted murder charge was dismissed, and he was paroled in his mother's custody on April 1. On May 25, the assault charges were reduced, and the case was adjourned in contemplation of dismissal. (Exhibit 11)

Edwin Ascencio, 14, was arrested on December 12, 1977, with Edwin Rodriguez, and charged with arson and reckless endangerment for allegedly starting a fire in a subway. A § 739 hearing was held on December 13, 1977, before Judge Ferrara. Ascencio had a petition pending and had not been attending school. The judge ordered detention, citing the seriousness of the allegations as a basis. A probable cause hearing was held on December 16, 1977. Ascencio admitted to committing the offense and was released to his family. He had been held in custody for 4 days, 3 of them under order of the court at the time of the probable cause hearing. At the dispositional hearing on January 26, 1978, he was placed on 2 years probation. (Exhibit 12)

James Peebles, 15 years old, was arrested and charged with arson for allegedly setting fire on July 4, 1978, outside the door of the complainant's apartment while the complainant and his family were inside. A § 739 hearing was held on July 5, before Judge Quinones. The boy was ordered detained, the judge citing the seriousness of the crime as the basis for his action. At the July 10 probable cause hearing Peebles admitted to reckless endangerment. At that time he had been in custody for 6 days. Detention was continued for psychological evaluation. On July 20, 1978, he was released to his mother. Parole status was revoked in December, 1978, and at the dispositional hearing he was placed with the Division for Youth. (Exhibit 14)

Jeffrey McCain, 15, was arrested on May 23, 1978. A § 739 hearing was held on May 24, 1978, before Judge Doran. The boy had been arrested on robbery charges on 3 prior occasions that same month, and petitions for these offenses were pending. He was ordered detained. A probable cause hearing was set for May 28, 1978, but was postponed. The matter was eventually dismissed because of lack of proof beyond a reasonable doubt to support the charges. McCain was held in custody for a least 4 days before the scheduled May 28 probable cause hearing. The record does not disclose the date of the boy's release. (Exhibit 15)

Carl Harris, 15, and Michael Taylor, 14, were arrested on March 5, 1977, for possession of stolen property and released. A § 739 hearing was held 12 days later on March 17, before Judge Rigler. In view of prior court contacts, the court ordered both detained. On March 21, 1977, both were adjudicated delinquent. By that time they had each spent 4 days in detention. At the dispositional hearing on May 16, 1977, Carl was placed on probation and Michael with the Commissioner of Social Services. (Exhibit 16)

Victor Melendez, 15, was arrested on July 11, 1977, for breaking and entering a school building at night and causing $1,500 in damages. A § 739 hearing was held on July 13, 1977, before Judge Gartenstein. This was Melendez's fourth arrest since March, 1977. He was ordered detained. A probable cause hearing was set for July 19, but was adjourned and was finally held on August 12, 1977, at which time Melendez was adjudicated a delinquent. On December 16, 1977, he was placed on one year's probation. It is unclear from the record what part of the period from July 11 to August 12, Melendez spent in detention.^[6] (Exhibit 17)

Tyrone Parson, 15, was arrested on November 6, 1976, for promoting gambling and for possession of a gambling device and released. At a § 739 hearing held on December 1, he was ordered detained. Seven prior court petitions were then pending. At the probable cause hearing on December 6, the petition was dismissed because the offense alleged did not come within the provisions of the penal law. He had been detained for 5 days between his § 739 and probable cause hearings. (Exhibit 18)

Geraldo Delgado, 15, was arrested on March 4, 1976, for stabbing and killing another boy. A § 739 hearing was held on March 5, 1976, before Judge Matthews. He ordered the boy detained because of the seriousness of the crime. Probable cause was found on March 10. Delgado had been detained for 5 days between his § 739 hearing and probable cause determination. Delgado was adjudicated a delinquent on April 4, 1976, and psychological tests were ordered. On July 7, 1976, he was released to his mother's custody and placed on one year's probation subject to treatment at Lincoln Hospital Adolescent Treatment Center. (Exhibit 19)

Tony Gomez, 15, was arrested on October 26, 1978, for allegedly assaulting and robbing a 77 year old woman and charged with grand larceny, robbery, assault and possession of stolen property. A § 739 hearing was held on October 27, 1977, before Judge Ferrara. He was ordered detained. Three prior petitions for robbery and attempted rape had been terminated without adjustment. A probable cause hearing was scheduled for October 31, but was adjourned. Gomez was released on a writ of habeas corpus on November 10, 1977. At the time he had been held in custody without a probable cause adjudication for 15 days, 14 of those days were pursuant to the court's order under § 739(a)(ii). On December 7, 1977, he was adjudicated a delinquent and given a one year suspended sentence. (Exhibit 20)

Maurice Dunaway, 13, was arrested on July 22, 1976, for theft of subway bus transfers from a Transit Authority transfer box. A § 739 hearing on July 23, 1976, was held before Judge Roache. The boy was ordered detained due to alleged threats against the complainant, disobedience at home and prior court contact. Another petition for a similar offense was outstanding. A probable cause hearing was set for July 27, but was adjourned to August 3. On the latter date Dunaway was adjudicated a delinquent. He had been detained for 11 days under court order at the time of his probable cause hearing on August 3. Thereafter he remained in custody until October 12, 1976, and then was released to his mother while placement was explored. On April 20, 1977, he was placed with the Division for Youth for 18 months. (Exhibit 21)

Daniel Nelson, 15, was arrested on April 19, 1979, charged with petit larceny for allegedly stealing a radio valued at $85, and released to the custody of his parents. A § 739 hearing was held on May 3, 1979, before Judge Roache. At the time, Nelson was on probation on a prior charge and was ordered detained. The probable cause hearing was held on May 6, and he was adjudicated a delinquent. At the time of his probable cause hearing he had been detained for 3 days. On December 12, he was placed with the Commissioner of the Department of Social Services for 18 months. (Exhibit 22)

Daniel Gregoire, 15, was arrested on August 23, 1977, on charges of burglary and criminal possession of stolen property. A § 739 hearing was held on September 20, 1977, before Judge Roache. Six prior delinquency petitions against this juvenile had been dismissed. His father told the judge that he could not control the boy and did not want him home. He was ordered detained and remained in custody until September 23, 1977, when he was paroled to his brother. Probable cause was found on September 29, and at the dispositional hearing he was placed in St. Vincent's for 18 months. Gregoire had spent a total of 3 days under court ordered detention. (Exhibit 23)

Carlos Lopez, 13, was arrested on August 30, 1977, for breaking and entering a store at 4:00 A.M., and was charged with burglary, petit larceny and criminal mischief. A § 739 hearing was held on August 31, 1977, before Judge Rigler. He was ordered detained because of family problems and five pending petitions. A probable cause hearing was set for September 6, 1977. On September 9, he absconded, was returned and absconded again on September 22, 1977, and again was returned. On November 10, 1977, the case was adjusted in contemplation of dismissal. It is difficult to ascertain the precise number of days this juvenile was detained, but it seems to have been at least 7 days before the probable cause hearing. (Exhibit 24)

Kevin West, 14, was arrested on May 31, 1971, and charged with grand larceny, jostling and resisting arrest, for taking property from a 71 year old woman and striking the arresting officer. The § 739 hearing was held on June 1, 1979, before Judge Matthews. Several outstanding petitions were then pending. West was ordered detained. On June 5, he was adjudicated a delinquent and placed with the Division for Youth for 12 months. His time in detention under court order before his probable cause hearing was 4 days. (Exhibit 25)

Carlos Fargas, 14, was arrested on April 24, 1979, for threatening to strike the complaining witness with a broomstick. He was charged with robbery, grand larceny, unlawful imprisonment and possession of stolen property. He was released pending a § 739 hearing which was held on June 6, before Judge Pollard. The failure of his parents to accompany him to court meant a lack of supervision to the judge and created a likelihood of further delinquent acts. For these articulated reasons he was detained. On June 11, he was paroled to the custody of his parents. He had been in pretrial detention under court order for 5 days. (Exhibit 26)

Jody Allen, 12, was arrested on May 25, 1979, for beating an elderly woman who had to be hospitalized. There had been seven previous court contacts but no findings. He was charged with assault and attempted robbery and released. The § 739 hearing was held on June 1, before Judge Pitaro. He was ordered remanded. The complaining witness did not prosecute the matter. Allen was released to the custody of his grandmother on June 4. He had been detained under court order for 3 days before being released. The petition was withdrawn on June 25. (Exhibit 27)

Emmett Jacobs, 14, was arrested for stealing a watch on May 30, 1979, and charged with robbery and grand larceny. A § 739 hearing was held on May 31, before Judge Leddy. Three prior petitions had been filed, but no findings had resulted. Noting his prior court contacts, the court ordered Jacobs detained. At the probable cause hearing on June 4, Jacobs admitted to petit larceny. At final disposition he was placed with the Division for Youth for 12 months. He had been detained 4 days before his probable cause adjudication. (Exhibit 28)

James Ancrum, 15, was arrested on August 16, 1979, for attempting to take a necklace off a man's neck. The man was allegedly beaten and thrown down a flight of stairs. A § 739 hearing was held on August 21, before Judge Ferrara after an order was issued removing the case from criminal court. Probation recommended release, but Ancrum was ordered detained, in view of the lateness of the hour when the act was allegedly committed. On August 24, probable cause was found. He was released to his mother's custody, and the petition was withdrawn. He was detained for 3 days under court order before the probable cause hearing. (Exhibit 35)

Douglas McLaughlin, 14 years old, was arrested on June 29, 1978, for burglarizing an apartment. Apparently he was on parole between the incident and the § 739 hearing on July 7. He had been arrested once previously, though no petition had been filed. The § 739 hearing was held before Judge Quinones, who ordered the boy remanded due to a statement by the boy's mother that he was beyond her control. A fact finding hearing was set for July 11, then adjourned until July 13, at which time McLaughlin was released to the custody of his mother. He was detained under court order for 6 days before the probable cause hearing. The petition was ordered dismissed on January 12, 1979. (Exhibit 43)

III

Juvenile Delinquency Proceedings

The threshold step in the court process in juvenile delinquency proceedings occurs at probation intake in family court. The juvenile may be brought to the intake part from temporary detention or in police custody, in which case he comes to the court in handcuffs and under police escort. The juvenile may have been released to the custody of his parent or guardian, with instructions to appear in court at an appointed time. N.Y. Family Court Act § 741 (McKinney). In these cases the juvenile appears on his own at probation intake, usually accompanied by a parent or relative. If the complainant appears, and it is the court's understanding that, more often than not, the complainant is a police officer, the case is assigned to a probation intake officer. If no complainant appears, a new date is set, and if after several continuances the complainant still does not put in an appearance, the non-felony proceedings are terminated. If a felony, the case is referred to the district attorney.

The intake probation officer must hold either a master's degree in social work or in sociology and related fields or such a bachelor's degree, plus two years of paid-for experience in social work and related fields. Many of the probation officers currently on duty in probation intake possess only the lower academic credentials augmented by relevant work experience.

The probation officer to whom the case is assigned interviews the police officer, the juvenile and parent or guardian, if present. Typically, the civilian complainant signs a statement at the police station and does not appear at probation intake. The juvenile is not represented by counsel. The purpose served by these interviews is to enable the probation officer to ascertain the nature of the offense charged, to obtain the juvenile and the police version of what took place, to assess the juvenile's role and personal involvement in the offense and to secure some background facts on the child. The juvenile is not subjected to any professional psychiatric or psychological examination unless mental health problems are patently evident. The entire interview process takes between 10 and 40 minutes. After the interviews are concluded, the officer checks to ascertain whether the juvenile has a history of prior involvement with the court. He then decides whether the case should be adjusted, that is, disposed of without court intervention. He may adjust the case, leaving it open to see whether the juvenile lives up to the conditions imposed with a view to having the case ultimately dismissed. Whether the case is adjusted or referred to court intake for further proceedings is determined based on criteria set forth in Family Court rules and the regulations in the New York State Department of Probation. N.Y. Family Court Act § 734(a) (McKinney); Family Court Rules, §§ 2507.3-2507.6.

If the officer decides that the case must be processed through court, the case is sent to court intake for a petition to be drawn and prosecution by the district attorney or corporation counsel. N.Y. Family Court Act § 731(1) (McKinney). When the probation officer decides to refer the matter to court intake for the drafting of a petition and further proceedings, he must accompany the reference with a recommendation that the juvenile either be remanded or paroled. This recommendation is based on the nature of the case, whether there is a prior record, the attitude of the juvenile and of the parent.^[7] There do not appear to be any governing criteria which must be followed by the probation officer in choosing between proposing detention and parole, nor do psychological or psychiatric factors appear to be considerations involved in the probation officer's recommendations.

Although the court was advised at trial that the corporation counsel is increasingly demanding that petitions be based on the statement of a complainant with personal knowledge of what took place, we were nonetheless advised that petitions based solely on information and belief are still being processed.

If in custody, the juvenile must be brought before a judge for a § 739 hearing within 24 hours. The judge may order him held for an additional 24 hours before holding the § 739 hearing. If the hearing results in pretrial detention, a fact finding hearing, which is equivalent to a trial on the merits, must occur within 3 to 6 days thereafter. N.Y. Family Court Act § 747 (McKinney). In some cases fact finding may be put off for good cause as long as 14 days after the § 739 proceeding, but if such a postponement occurs, there must be a probable cause hearing within 72 hours of the § 739 determination. N.Y. Family Court Act § 748 (McKinney).

The § 739 hearing takes place at court intake and is similar to an arraignment in an adult case. The juvenile is brought before the judge and assigned counsel. The charges are read, and the juvenile is advised of his rights. N.Y. Family Court Act § 739 (McKinney). In addition to the juvenile and assigned counsel, the prosecutor, court liaison officer and probably a parent or guardian are also present. The court liaison officer is the probation officer on duty at court intake. He is not the probation intake officer who interviewed the juvenile and whose recommendation to remand or parole the judge will act upon. There is an official stenographic transcript of this proceeding, but no inquiry is made into the non-hearsay allegations of the truth of the charges, and no testimony is taken under oath. After the charges are read and the juvenile is advised of his rights, the judge asks for the recommendation from probation intake. This is read to him by the court liaison officer. The judge may make a few inquiries of defense counsel, prosecution and court liaison officer. Then he announces his decision whether to order detention or parole the juvenile. He usually gives some indication on the record of his reasons for ordering pretrial detention.

Each judge follows his own individual approach to this determination. Judge Cesar Quinones, the only judge of family court to testify, stated that his determinations are based on the petition, prior record and the probation officer's recommendation. He also considers whatever facts are brought out by defense counsel, but appointed counsel (the usual case) is assigned at court intake when the case is called and has very little opportunity to learn much pertinent information about the juvenile. Background on the child is not available at the court intake unless there has been prior court contact, and if the latter, this is a negative factor. Often there is no one present with personal knowledge of what happened.

Judge Quinones stated that he considers allegations of violence against persons more serious than property infractions; that he looks to the seriousness of the charges, whether anyone was hurt, whether the acts occurred in early morning which would be indicative of lack of parental supervision. He testified that a § 739 determination to detain the juvenile is not intended to benefit the child, but to protect society. At court intake, Judge Quinones testified, approximately 30 cases are processed in a typical 9-5 day, thus each proceeding is of 5-15 minutes in duration. The great majority of juveniles ordered detained are blacks and Puerto Rican. No uniform court or judicial criteria, yardstick or guidelines govern a § 739 disposition. In deciding whether the juveniles should be detained, each judge must rely on his own subjective judgment, based on the limited information available to him at court intake and whatever personal standards he himself has developed in exercising his discretionary authority under the statute. (Tr. 435-513).

At the fact finding hearing, a determination is made whether the juvenile is to be adjudged delinquent because he is found to have committed the offenses charged, or, as in the typical case, a lesser offense than that originally charged. Where a probable cause hearing precedes fact finding, there must be a sufficient prosecutorial presentation to warrant a finding of probable cause that the juvenile committed the offense charged, and, as the case files show, these probable cause and fact finding proceedings often result in an admission by the juvenile that he committed a lesser offense. If the fact finding hearing results in an adjudication of delinquency, the juvenile has a dispositional hearing, at which time the judge determines what form of punishment or discipline is warranted. The dispositional hearing determines whether the juvenile is to be sent to a state facility, secure or unsecure, or released on probation. N.Y. Family Court Act § 753 (McKinney). Probation prepares a report for the judge at disposition which is compiled after a full investigation of the juvenile in school, community and home and a study of relevant intake records.

In 1979, 58.2% of the juveniles held in pretrial detention^[8] were accused of crimes against property and other non-violent crimes; 47.1% were accused of crimes against persons and other violent crimes. In 1977, 18,447 juvenile delinquency petitions were disposed of. Pretrial detention was ordered in 3,546 instances.^[9] At final disposition 47 juveniles were subjected to some form of restricted placement and 594 were placed in Division for Youth under Title 2, and 667 were placed with the Division under Title 3.^[10] The latest published statistics are for the year 1979. There were 17,880 juvenile delinquency petitions filed in the state.^[11] Pretrial detention was ordered in 4,290 cases.^[12] Of the petitions filed, 932 were dismissed at fact finding;^[13] 23 were dismissed for other reasons and 292 at the dispositional hearing;^[14] 731 were dismissed for failure to prosecute;^[15] 846 were dismissed without prejudice;^[16] 4,882 were dismissed in furtherance of justice,^[17] and 1,617 were dismissed for other reasons;^[18] 1,456 petitions were withdrawn;^[19] 19 were discharged with warnings;^[20] in 725 cases judgment was suspended;^[21] 3,511 were placed on probation without placement;^[22] 17 were placed on probation and placed in private homes;^[23] 760 were placed with the Commissioner of Social Services;^[24] 412 were placed with the Division for Youth under Title 2;^[25] 594 were placed with the Division under Title 3;^[26] 2 were placed in restrictive placement for 5 years, 2 for 3 years and 70 in other forms of restricted placement.^[27]

IV

The Parties' Contentions

Petitioners contend that the statute is unconstitutional on its face and as applied in that it offends the due process and equal protection guaranties of the 14th Amendment. Petitioners argue that the subjective prediction or prognosis of the imminence of future misconduct which § 739(a)(ii) authorizes as the basis for the pretrial detention of juveniles is a vague, arbitrary and capricious standard and that no rationally based prediction or reasoned determination is possible under the statutory scheme. They insist that the statute is overbroad and violates equal protection constraints in authorizing pretrial detention of juveniles, since they are thereby treated differently from adults, and there is neither a rational basis pertinent to the differentiation between the two groups, nor any compelling state justification to warrant imposition of the restrictions imposed on juveniles under the statutory scheme. Petitioners also argue that the burden the statutory scheme imposes falls more heavily on non-white minorities and the poor than on whites and the more affluent, but they stop short of making a specific claim of a denial of equal protection on that basis.

The attorney general argues pro se that the statutory scheme is constitutionally and legislatively permissible; that all judicial and administrative decision making in the criminal justice system has the same potential for inaccuracy in predicting future acts, as does the statute involved here, but nonetheless the judge's subjective determinations as to bail, parole, and length of sentence have been accepted as appropriate exercises of judicial discretion free of constitutional infirmities; that the statute does not result in random detention; that pretrial detention does not constitute punishment without trial and does not deny juveniles any due process or equal protection rights, and in any event, the compelling interest of the state in reducing the risk of pretrial crimes by juveniles justifies the difference in the treatment of juveniles and adults and validates the statute.

V

Determination

A

The issues raised in this litigation are difficult and complex. Petitioners contend that the statutory scheme is at war with the guarantee of equal protection since juveniles are subjected to pretrial detention while adults are not. Both parties argue that the appropriate test to measure the constitutional validity of § 739(a)(ii) under the equal protection clause is whether the state has demonstrated a compelling, overriding state interest justifying the legislation, and no feasible, less drastic measures are appropriate to accomplish the state's objective.

In People ex rel. Wayburn v. Schupf, 39 N.Y.2d 682, 385 N.Y.S.2d 518, 350 N.E.2d 906 (1976), the New York Court of Appeals upheld the Act against an equal protection challenge on the ground that the statute served a compelling state interest. That interest as articulated by the court was the juveniles' emotional and intellectual immaturity, necessitating that they not be held to the more onerous adult standards of responsibility for their conduct. Moreover, the court reasoned that immaturity and lack of comprehension caused juveniles to view criminal conduct less seriously than adults. Accordingly, procedures designed to prevent further criminal acts by juveniles accused of delinquency were held to be justified to protect the public.

If the parties are, and the state court was, correct that compelling justification and unavailability of any feasible less drastic alternatives are the criteria to measure the statute's validity under the equal protection clause, my task would be considerably eased. § 739 clearly cannot withstand a constitutional challenge when the equal protection test of strict scrutiny is applied.

The record made at trial reveals that roughly 3,546 juvenile delinquents were subjected to pretrial detention in 1977. The figures on final disposition of all juvenile delinquencies show that 667 were placed in training schools of the Division for Youth and 47 others were accorded some other form of restricted placement. Assuming that the above 714 juveniles were all among those subjected to pretrial detention (an assumption not necessarily correct), the remaining 2,832 cases resulted in probation to the custody of the parent, guardian or placement in a foster home, or placement in a youth camp.

The 1979 figures paint the same picture. Some 17,880 juvenile delinquency petitions were filed. 4,290 juveniles were held in pretrial detention; 670 were placed with the Commissioner of Social Services; 412 with the Division for Youth under Title 2; 594 with the Division for Youth under Title 3, and 77 were accorded some form of restricted placement. (See ante, pp. 703-704). Assuming again that all of these 1,753 were subjected to pretrial detention, the result is that more than half such cases resulted in some form of dismissal or probation. Under any yardstick when at final disposition the overwhelming majority of juveniles are paroled and sent home, the claim that there is a compelling need to subject them to pretrial detention becomes hard to credit.^[28]

Sixteen juveniles in the case file histories which make up the record in this case were on parole from 5 to over 30 days between the time of the commission of the offense and their § 739 hearing, with no evidence that in the interim any further crimes had been committed. Yet they were remanded. In some cases there had been no prior court contact, and parole was recommended. Nonetheless, detention was ordered. These instances show a clear lack of any compelling justification in the application of the statute. If a juvenile has been on parole a month, two weeks or 5 days without committing another crime, unless family courts are assumed to be sorcerers who can foretell future events, it is difficult to understand the rationale for depriving him of his freedom because of the likelihood that he will commit a crime within 3 to 6 days. Such decisions, under any circumstances would seem to constitute abuse of discretion.

It is, of course, clear that neither the 14th Amendment nor the Bill of Rights' strictures protect only adults. In re Gault, 387 U.S. 1, 87 S. Ct. 1428, 18 L. Ed. 2d 527 (1967). Moreover, the mere recitation of benign, compensatory purposes is not an automatic shield which protects against inquiry into the actual impact of a statutory scheme, Weinberger v. Wiesenfeld, 420 U.S. 636, 648, 95 S. Ct. 1225, 43 L. Ed. 2d 514 (1975), and in its application demonstrated on this record, there has been no articulation of the parens patriae thesis as justification for exercise of the court's authority under § 739(a)(ii).

Remand has been deemed warranted in this case because of the seriousness of the crime, because of prior court contact, because the court always orders detention in certain cases, because such is warranted to punish the juvenile or, as Judge Quinones testified, to protect the public. Yet restraints of minors which might be considered impermissible if applied to adults have been justified as furthering an appropriate state interest, see e. g., Planned Parenthood of Central Missouri v. Danforth, 428 U.S. 52, 102-103, 96 S. Ct. 2831, 2856-2857, 49 L. Ed. 2d 788 (1976) (Stevens, J., dissenting), and parens patriae, while not an unlimited justification, Kent v. United States, 383 U.S. 541, 555, 86 S. Ct. 1045, 1054, 16 L. Ed. 2d 84 (1966), has been the underlying validating basis for a state's special treatment of the young. The high minded purposes of the juvenile justice system and the designation of proceedings involving juveniles as civil, rather than criminal, serves a benign social function, but "serious questions [arise] as to whether the actual performance measures well enough against theoretical purpose to make tolerable the immunity of the process from the reach of constitutional guarantees applicable to adults." Kent v. United States, supra at 555, 86 S.Ct. at 1054.

Yet, in no case has the United States Supreme Court subjected a juvenile classification to the strict scrutiny reserved for classifications based on race, Loving v. Virginia, 388 U.S. 1, 11, 87 S. Ct. 1817, 1823, 18 L. Ed. 2d 1010 (1967); McLaughlin v. Florida, 379 U.S. 184, 191-2, 85 S. Ct. 283, 287-289, 13 L. Ed. 2d 222 (1964), alienage, Oyama v. California, 332 U.S. 633, 664-66, 68 S. Ct. 269, 284-85, 92 L. Ed. 249 (1948); Korematsu v. United States, 323 U.S. 214, 65 S. Ct. 193, 89 L. Ed. 194 (1944); Hirabayashi v. United States, 320 U.S. 81, 63 S. Ct. 1375, 87 L. Ed. 1774 (1943), or where the right or interest interfered with is characterized as fundamental, Zablocki v. Redhail, 434 U.S. 374, 383, 98 S. Ct. 673, 679, 54 L. Ed. 2d 618 (1978), or struck down the differentiation between the treatment of juveniles and adults because the criteria applicable to the above classification were not met.

This provision seems to lack the high minded purpose of protecting the juvenile which is the objective generally cited to justify deviations from adult standards in the juvenile justice system. Rather, the purpose seems to be to protect society from the juvenile. No articulated rationale has been set forth on this record which would warrant disadvantaging the young on this ground. Nonetheless, I regard the equal protection challenge to the statute as insubstantial.

The right of a state to differentiate between adults and minors is too ingrained in the fabric of the law for such classifications at this late date to be held to the test of strict scrutiny. Even though pretrial detention interferes with the juveniles' personal liberty, that alone would not seem to warrant application of the strict scrutiny test. See e. g. Bell v. Wolfish, 441 U.S. 520, 99 S. Ct. 1861, 60 L. Ed. 2d 447 (1979). The more appropriate inquiry is whether § 739(a)(ii) can be justified on grounds of rationality, see Weinburger v. Salfi, 422 U.S. 749, 95 S. Ct. 2457, 45 L. Ed. 2d 522 (1975), and under this less rigorous yardstick, the statutory scheme would qualify as meeting the constitutional requirements of equal protection. Equal protection guarantees deal with invidiously discriminatory classifications and differentiations. This statute does differentiate between youth and adults but to paraphrase Justice Powell, concurring in Zablocki v. Redhail, 434 U.S. 374, 399, 98 S. Ct. 673, 688, 54 L. Ed. 2d 618 (1978), to subject § 739 to a compelling state purpose inquiry would cast doubt on the network of laws that the states have devised to deal with problems of the young — delinquency, neglect and care. Accordingly, I turn to what I regard as a more substantial challenge to the statute.

B

The statutory scheme curtails the freedom of one presumptively innocent, albeit a juvenile, on the prediction or hunch of a judge that unless held in custody he will commit another crime within 3 to 6 days. The more fundamental consideration, therefore, is whether on its face or as applied, § 739(a)(ii) meets due process standards. There is little doubt that due process is a requisite to the constitutional validity of proceedings which may result in the curtailment of a juvenile's freedom, since all "individuals possess a liberty interest in being free from physical restraint." Greenholtz v. Inmates of Nebraska Penal & Correctional Complex, 442 U.S. 1, 33, 99 S. Ct. 2100, 2117, 60 L. Ed. 2d 668 (1979) (Marshall, J. dissenting). While the limits of the liberty interest protected from deprivation without due process have never been precisely defined, it has "always been thought to encompass freedom from bodily harm and punishment. See Rochin v. California, 342 U.S. 165, 72 S. Ct. 205, 96 L. Ed. 183 (1952). It is fundamental that a state cannot hold and physically punish an individual except in accordance with due process of law." Ingraham v. Wright, 430 U.S. 651, 673-74, 97 S. Ct. 1401, 1413-14, 51 L. Ed. 2d 711 (1977).

Juvenile proceedings which may result in confinement to state institutions because of alleged misconduct must have the essential ingredients of fair treatment. In re Winship, 397 U.S. 358, 359, 90 S. Ct. 1068, 1070, 25 L. Ed. 2d 368 (1970); In re Gault, 387 U.S. 1, 18, 19, 87 S. Ct. 1428, 1438-39, 18 L. Ed. 2d 527 (1967); In the Matter of W & S, 19 N.Y.2d 55, 277 N.Y.S.2d 675, 224 N.E.2d 102 (1966). Application of due process standards may not be defeated because of the labels attached by the state to the pertinent proceedings, In re Gault, supra, 387 U.S. at 50-51, 87 S.Ct. at 1455-56; In re Winship, supra, 397 U.S. at 366, 90 S.Ct. at 1074, and a "claim that the traditional requirements of due process are applicable in the context of pretrial detention" has not been foreclosed. Gerstein v. Pugh, 420 U.S. 103, 127, 95 S. Ct. 854, 870, 43 L. Ed. 2d 54 (1975) (Stewart, J. concurring).

The predictive § 739 determination is made before any proceeding, either adversarial or neutral, has found probable cause or rendered a determination on the merits that the juvenile committed the delinquent act charged. The Act violates due process requirements because: (1) it gives the judge a license to act arbitrarily and capriciously in a § 739 prediction of the likelihood of future criminal conduct which cannot result from a reasoned determination, (2) pretrial detention without a prior adjudication of probable cause is, itself, a per se violation of due process, and (3) in addition, constitutes punishment that is constitutionally impermissible under the due process clause.

1. The first defect is perhaps the most pernicious. The judge is empowered to make a prediction about the probability of an individual committing a crime if released. No guidelines for making that determination are set out in the statute, and none has been adopted by the court. The judge's determination is moored to no concrete or reasonably determinable yardsticks. We were advised at trial that each judge utilizes his own personal standards. Accordingly, there can be no uniform application of § 739, since each judge's subjective views and biases must necessarily govern. The whole process is riddled with subjectivity and caprice and confers upon the judge "a license for arbitrary procedure." Kent v. United States, supra, 383 U.S. at 553, 86 S.Ct. at 1053.

A judge is not an expert criminologist and even these professionals who have studied more about an individual's criminal propensities than the rest of us confess, as we shall see later, an inability to perform the task § 739 imposes on the judge. Thus, "[a] serious question of procedural due process is raised by this feature of standardless discretion, particularly in light of the hazards of prediction..." Zablocki v. Redhail, 434 U.S. 374, 402 n.4, 98 S. Ct. 673, 689, 54 L. Ed. 2d 618 (1978) (Powell, J. concurring). Moreover, the judge is accorded total and unbridled discretion since effective review is not possible, see People ex rel. Wayburn v. Schupf, supra.^[29]

Although the probation intake officers are required to follow established criteria in determining whether to adjust a case or pursue it through the courts, they make their recommendations whether to release or remand the juvenile based on "whether they feel a child may be a calculated risk if returned to the community." (Tr. 161). Thus, as the study undertaken by Professor Wilkins referred to in his trial testimony demonstrated, juveniles under such circumstances may have their liberty curtailed because they possess "characteristics that annoy" the parole officer or judge, or because they may be viewed as "uncooperative." (Tr. 317). It is clear that the judge decides on pretrial detention for a variety of reasons— as a means of protecting the community, as the policy of the judge to remand (Exhibit 6a, p.3), as an express punitive device (Exhibit 42, p.11), or because of the serious nature of the charge. (Tr. 465, Exhibit 13a, p.3), among others.

Probation intake cannot undertake a thorough investigation of the youth. The probation officer learns what delinquency is charged. This may be only marginally helpful because as a general rule the original charge is considerably reduced or modified after a probable cause or fact finding hearing.^[30] He evaluates the case on the facts presented by the police officer and the juvenile. He ascertains whether the juvenile has had any prior court contacts and the nature of those contacts. He is not equipped to make any sophisticated or in depth evaluation of the charges or the child, particularly in light of time constraints. If the family court guidelines require him to process the case through the court, he must recommend, on the limited information provided, whether the juvenile should be detained or paroled.

While the time of probation intake is limited, there is considerably more time available than the judge may allocate. The judge has no opportunity to interview the juvenile. Whatever he hears in opposing remand comes from counsel for the juvenile. If counsel is court appointed, and this is the typical case, he takes on responsibility for the proceedings only moments before convincing reasons must be presented to the court for not ordering pretrial detention. The judge has roughly 5 to 15 minutes to determine whether there is the likelihood that the juvenile would commit another crime before the return date if released.

At trial, plaintiff's expert, Professor Leslie Wilkins, stated that he would be surprised if recommendations based on intake interviews were better than chance and assessed the judge's subjective prognosis about the probability of future crime as only 4% better than chance — virtually wholly unpredictable. He testified that no method had yet been devised which could predict with any acceptable degree of accuracy that a juvenile shall commit a crime, particularly the commission of an offense in a short space of time, as the judge must do in making his § 739 decision.

A review of the literature supports Professor Wilkins' observations and shows that no diagnostic tools have as yet been devised which enable even the most highly trained criminologists to predict reliably which juveniles will engage in violent crime.

As one of the leading scholars on the issue has put it:

[t]he ability to predict which juveniles will engage in violent crime, either as adolescents or as adults, is very poor. The conclusion ... that "there has been no successful attempt to identify, within ... offender groups, a subclass whose members have a greater than even chance of engaging again in an assaultive act" is as true for juveniles as it is for adults. It holds regardless of how well trained the person making the prediction is — or how well programmed the computer — and how much information on the individual is provided.

J. Monahan, The Prediction of Violent Behavior in Juveniles, 10-11 (paper presented at National Symposium on the Serious Juvenile Offender in Minneapolis, Sept. 19-20, 1977) (emphasis deleted), quoted in Feld, Reference of Juvenile Offenders for Adult Prosecution: The Legislative Alternative to Asking Unanswerable Questions, 62 Minn.L. Rev. 515, 541-42 (1978).

Moreover, not only does it appear that one cannot predict dangerousness with an acceptable degree of accuracy, but, to the extent that dangerousness can be predicted at all, there is a substantial problem of overprediction, that is, to identify persons potentially dangerous who, if subsequently released, would engage in no further violent or even criminal behavior. If a true positive is a person, who is predicted to and in fact engages in such behavior, then a false positive would be someone predicted to be dangerous but who would not engage in violent or criminal behavior. Thus, overprediction can be expressed in terms of the ratio of false positives to true positives.

Monahan also found that

violence is vastly overpredicted whether simple behavioral indicators are used or sophisticated multivariate analyses are employed, and whether psychological tests are administered or thorough psychiatric examinations are performed.

J. Monahan, supra, quoted in Schlesinger, The Prediction of Dangerousness in Juveniles: A Replication, 24 Crime & Delinquency 40, 47 (1978).

The literature on the prediction of dangerousness, distinguishes two methods for anticipating future behavior: clinical prediction,^[31] on the one hand, and statistical or actuarial prediction,^[32] on the other.

A clinical prediction entails a clinician reviewing whatever information is deemed relevant and making a predictive judgment on the basis of professional training and intuition. Clinical prediction requires an integration of available information about the individual in order to develop "some psychological hypothesis regarding the structure and dynamics of this particular individual."

Feld, supra at 543 (quoting P. Meehl, Clinical Versus Statistical Prediction 4 (1954)). Clinical methods rely on the use of trained psychiatric or social science staff, attempting a "wide band procedure in which a broad range of information about the individual is gathered from sources such as interviews, social history and projective testing." Koerin, Violent Crimes, Prediction and Control, 24 Crime & Delinquency 49, 51 (1978). Fashioning the hypothesis about the particular individual is frequently said to be done in terms of a diagnostic frame of reference — a scheme of important personal variables that figure in the individual's personality and some sense of the dynamics of the personality. See e. g. Kozol, Boucher and Garofalo, The Diagnosis and Treatment of Dangerousness, 18 Crime & Delinquency 371, 384-86 (1972).

Actuarial or statistical prediction, on the other hand

entails the development of probability relationships between predictor variables, such as age and prior offenses, and the behavior to be predicted — violence. It requires an examination of the individual only to determine the presence of the predictor variables.

Feld, supra at 543. In statistical prediction, the individual is, in effect, classified in terms of the presence or absence of the predictor variables. Given the classification, the frequencies of behaviors can be concluded for persons belonging to the same class. Id.

The family court judge's evaluation of a juvenile bears a superficial resemblance to both methods in the sense that it lacks the refinements of each. The judge's assessment could not properly be called a clinical prediction, unless it could be supposed that the judge's experience on the bench and his judgment of human nature substitutes for a clinician's training; his confrontation with the accused juvenile substitutes for a diagnostic interview and examination; and the juvenile's arrest record substitutes for a patient's history. Nor is the evaluation by the judge in any way comparable to a rigorous statistical prediction. In comparison to the large number of variables used in the studies of statistical prediction, apparently very little data is made available to the judge. And, what is presented is crude and unrefined. There is no guarantee that significant information is available to him, or, if it is there, that it is perspicuous. Finally, there is no use of follow up studies to refine and reformulate the statistical prediction method that, implicitly, the family court judge might be said to be using. Accordingly, he would continue to rely on useless or even incorrect data and assign incorrect weight to the various variables utilized.

The literature generally agrees that no reliable method of predicting dangerousness, whether clinical or actuarial in nature, exists at this time. It follows a fortiori that the family court judge's opinion, lacking the refinements of each sort of predictive method, is also unreliable. Further, even to the extent that the literature on each type of method is sometimes optimistic about the prospect of developing a reliable clinical or statistical method, nothing can properly be inferred about the judge's evaluation of juveniles, since the indications of optimism reflect methodological improvements which are not utilized in the judge's determinations.

Thus, it is clear that juveniles who are subjected to § 739 detention have their freedom curtailed by judgments that are untrustworthy and uninformed and without the requisite rationality which due process mandates.

In Jurek v. Texas, 428 U.S. 262, 275-76, 96 S. Ct. 2950, 2957-58, 49 L. Ed. 2d 929 (1976), an attack on a Texas statute, requiring a jury at the sentencing proceedings following conviction for a capital offense to consider whether the evidence established beyond a reasonable doubt that the defendant would commit criminal acts of violence, that would constitute a continuing threat to society, was rejected in an opinion by Justices Stewart, Powell and Stevens. They said:

It is, of course, not easy to predict future behavior. The fact that such a determination is difficult, however, does not mean that it cannot be made. Indeed, prediction of future criminal conduct is an essential element in many of the decisions rendered throughout our criminal justice system. The decision whether to admit a defendant to bail, for instance, must often turn on a judge's prediction of the defendant's future conduct. And any sentencing authority must predict a convicted person's probable future conduct when it engages in the process of determining what punishment to impose.... The task that a Texas jury must perform in answering the statutory question at issue is thus basically no different from the task performed countless times every day throughout the American system of criminal justice. What is essential is that the jury have before it all possible relevant information about the individual defendant whose fate it must determine. Texas law clearly assures that all such evidence will be adduced.

(footnote citations omitted)

The attorney general argues to the same effect, but the views expressed by the Justices in Jurek and the argument of the attorney general that courts are constantly engaged in predicting future conduct afford no basis for validating a § 739 determination.

Denial of bail because of a risk of non-appearance is usually made after a hearing which reveals that the defendant is in good health, has no roots in the community, no family ties, is not employed and owns no property, has possible access to financial resources and no recognized attachments which would assure his appearance at trial. A sentence is imposed after full investigation of the defendant's background. The Texas jury in Jurek, as in bail and sentencing cases, had all possible relevant information about the defendant, and it had heard evidence adduced at trial and found him guilty beyond a reasonable doubt of an offense for which the defendant could be sentenced to death.

The family court judge in making a § 739 decision knows nothing about the juvenile if he is a first time offender. The judge, of course, has the charges before him, but he also should know that the charges will surely be reduced at probable cause or fact finding. If the juvenile has had prior court contact, the judge knows that he has before him a recidivist, but at that point he does not know the extent of the juvenile's personal involvement, the extent to which the charges have been inflated and whether probable cause will be found. He must operate on intuition and his personal predilections, the antithesis of reasoned action.

Arbitrary and capricious adjudications that may deprive a juvenile of his freedom because of a feeling that a probation officer and judge have that the juvenile will commit a crime if released — a feeling that is intuitive rather than reasoned — are fundamentally offensive to a concept of ordered liberty in a free society.

Moreover, even aside from the innate inability to support a § 739 order of pretrial detention as a rational or reasoned judgment, the record is replete with § 739 determinations that constitutes arbitrary and capricious action in its crudest and rawest form. Whether the juvenile was a first offender with no prior conduct, whether the court was advised that the juvenile was an obedient son or was needed at home, whether probation intake recommended parole, the case histories in this record disclose that it was not unusual for the court to discount these considerations and order remand based on a 5 to 15 minute evaluation. See e. g., transcript of proceedings in case file histories of Carlos Fargas, Johnny McArthur, Christopher Cox, Glen Maloney, Wade Forde, Francisco Ramos, Jerome Basnight, Jose Cruz.

What seems to be the most patently arbitrary action, however, occurs in those cases in which the juvenile is paroled to the custody of his parents or guardians after the incident, has been home on parole ranging in the case histories in this record from roughly 5 days to over a month prior to the § 739 hearing, and even though there is no showing of any other infraction while on parole, the judge nonetheless orders remand purportedly because there is a risk that the juvenile will commit a new crime before he is tried on the instant offense. Whatever meager rationale may exist in other situations to warrant ordering the juvenile confined pursuant to a § 739 determination, remand is wholly unjustified and unsupportable in these circumstances, and if the determination could be tested on review, it would surely be set aside as a clear abuse of discretion.

The examples in this record are such as to indicate that a § 739 remand is not unusual even in such cases. For example Carlos Fargas was arrested on April 24, 1979, and paroled to the custody of his parents. His § 739 hearing occurred on June 6, some 43 days later. His mother was sick in the hospital. Nonetheless, he was ordered remanded. Similarly, Luis and Sammy Rosario were charged with involvement in a delinquent act which occurred on March 2, 1978. Both were on parole from March 2 to March 15 when their § 739 hearing took place. Both were remanded. Johnny McArthur was involved in delinquency on October 12, 1977. He was on parole from that date until his § 739 hearing on November 28, 1977. His mother spoke highly of Johnny, but he was remanded. Christopher Cox, Glen Maloney and Wade Forde, along with Clarence Smith and Alex Michael, were involved in an incident on October 13, 1976, and paroled to the custody of their parents. At the § 739 hearing on November 12, in response to a court inquiry, the police officer stated that the boys had not given him any trouble. Smith, who had private counsel, and Michael were paroled. Cox, Maloney and Forde, however, were remanded. There is nothing in the record to indicate the reason for the difference in treatment. Cox who was 13 was remanded and Michael, who was 15, was paroled. Probation intake had recommended parole for all. None had had any prior record.

Francisco Ramos was charged with a delinquency on March 4, 1978, and was on parole from March 4 to his § 739 hearing on March 17. While this was his 9th arrest and he was on parole on other matters, he had been reporting regularly. Parole was recommended. He was remanded. Jerome Basnight was free on parole for 6 days after his delinquent act, from October 14, 1977, until October 20, 1977. At his § 739 hearing on the latter date, despite this being his first court contact, he was remanded. Jerome Cruz was on parole from March 7, 1977, until his § 739 hearing on March 25, 1977. This was his first court contact, but he was remanded.

Carl Harris and Michael Taylor were free on parole from March 5, 1977, to March 17, 1977; Tyrone Parson was free on parole from November 6, 1976, to December 1, 1976; Daniel Nelson was free on parole from April 19, 1977, to May 3, 1977; and Daniel Gregoire was free on parole from August 23, 1977, to September 20, 1977. Each was remanded. It is conceded that these juveniles had prior court contacts. In the case of Parson, there had been 7 such contacts, and in the case of Gregoire, 6 such prior contacts, five of which had been dismissed. While evidence of prior delinquency may make remand seem less egregious than the other listed examples, arbitrary use of § 739 power is nonetheless clearly manifest when a juvenile comes to court days, weeks or more than one month after the charged incident without any showing of any intervening criminal act and is nonetheless remanded because of the court's professed belief that he will commit a crime before the return date. Such determination under those circumstances is both irrational and unreasoned. A statutory scheme that permits an individual's freedom to be restricted pursuant to that kind of irrational behavior must be condemned under the due process clause. See Kent v. United States, supra; Board of Regents v. Roth, 408 U.S. 564, 572, 92 S. Ct. 2701, 2706, 33 L. Ed. 2d 548 (1972).

2. A juvenile subjected to pretrial detention has not as yet had a "judicial determination of probable cause which the Fourth Amendment requires as a prerequisite to extended restraint of liberty following arrest." Gerstein v. Pugh, 420 U.S. 103, 114, 95 S. Ct. 854, 863, 43 L. Ed. 2d 54 (1975); Bell v. Wolfish, 441 U.S. 520, 526, 99 S. Ct. 1861, 1867, 60 L. Ed. 2d 447 (1979). When a juvenile is arrested and held in custody he must be taken before a family court judge within 24 hours for a § 739 hearing. The judges may postpone the hearing for another 24 hours. No probable cause determination is made before the § 739 hearing. The juvenile at this stage is without counsel until he appears for the § 739 hearing. In some instances he is before the court on a delinquency petition filed on information and belief. That is, the charges in the petition are not verified by someone with direct knowledge of what took place. If ordered remanded, a fact finding hearing must occur 3-6 days thereafter. That hearing may be delayed for as long as 14 days after the § 739 hearing but in such a case a probable cause hearing must be held within 72 hours of the § 739 hearing. Accordingly a juvenile may be held in pretrial detention under a § 739 determination for as long as 5 days prior to any proceedings being held, neutral or adversary, pursuant to which a judicial determination of probable cause is made.

The burdens of pretrial detention are substantial to impose on a presumptively innocent man, even when there is probable cause to believe he has committed a crime. Baker v. McCollan, 443 U.S. 137, 153, 99 S. Ct. 2689, 2699, 61 L. Ed. 2d 433 (1979) (Stevens, J., dissenting). To allow a state to impose these burdens on a juvenile before the state has taken the initial validating step of establishing its right to restrict the individual's liberty, in order to insure his presence at trial, would seem to be at war with accepted concepts of due process of law.

While not every disability inflicted during pretrial restraint amounts to punishment in the constitutional sense, Bell v. Wolfish, supra, 441 U.S. at 537, 99 S.Ct. at 1872, and there is clearly a distinction between punitive measures, imposition of which the constitution forbids prior to an adjudication of guilt and regulatory restraints that may be permissible, id.; Kennedy v. Mendoza-Martinez, 372 U.S. 144, 165-167, 186, 83 S. Ct. 554, 565-567, 576, 9 L. Ed. 2d 644 (1963), the Act at issue here licenses a judge of the family court, acting on an intuitive hunch that the juvenile is probably guilty, to treat an adolescent like a convict until his innocence is established. In a society riddled with violence and fearful of crime, this approach may seem appropriate. Whatever current social imperatives might lead us to wish to deal with the problem of juvenile delinquency on the above basis, the presumption of innocence is ingrained in the fabric of our fundamental law, and the prerogatives § 739 permit contravene guaranties and rights we have long considered fundamental. See e. g., Ingraham v. Wright. Pretrial adult detainees held because they could not meet the terms of bail have the presumption of innocence even though there has been a probable cause determination. Such detention has been held to be regulatory, not punitive, see Bell v. Wolfish, but juveniles held in pretrial detention present a stronger case of innocence than the adult. There has been no probable cause determination as to them, and "under the Due Process Clause a detainee may not be punished prior to an adjudication of guilt in accordance with due process of law." Bell v. Wolfish, supra, 441 U.S. at 535, 99 S.Ct. at 1872.

3. When a court pronounces the judgment that a youth must be incarcerated under § 739 because he or she is likely to commit additional crimes if released, the act of incarceration constitutes punishment which cannot constitutionally be imposed prior to an adjudication of guilt. Such a decision to incarcerate offends the due process clause of the 14th Amendment because it is imbued with what have long been recognized as the three essential attributes of punishment: it inflicts a deprivation of a constitutionally protected liberty or property interest; it is officially imposed rather than being the spontaneous act of an errant official, see Hernandez v. Lattimore, 612 F.2d 61 (2d Cir. 1979); and it stigmatizes the youthful detainee in a way that associates him with criminal behavior.

Whether a deprivation of liberty or property constitutes "punishment" has often turned on the criminal or non-criminal nature of the behavior for which it is imposed. Compare Trop v. Dulles, 356 U.S. 86, 78 S. Ct. 590, 2 L. Ed. 2d 630 (1958) (loss of citizenship for desertion from armed forces held to be punishment) with Perez v. Brownell, 356 U.S. 44, 78 S. Ct. 568, 2 L. Ed. 2d 603 (1958) (loss of citizenship for voting in foreign election held not to be punishment). See also Lipke v. Lederer, 259 U.S. 557, 42 S. Ct. 549, 66 L. Ed. 1061 (1922); United States v. LaFranca, 282 U.S. 568, 51 S. Ct. 278, 75 L. Ed. 551 (1931). In the latter two cases the regulatory or penal character of a tax depended on its "involving the idea of punishment for infraction of the law." 282 U.S. at 572, 51 S.Ct. at 280. Similarly, in Kennedy v. Mendoza-Martinez, supra, the automatic forfeiture of citizenship for draft evaders was held to constitute punishment.

However, "punishment" is not limited to deprivations based on having committed a specific, identifiable crime. It is sufficient if the person is stigmatized as disreputable in a manner closely analogous to the process that occurs upon a formal finding of criminal guilt. In United States v. Lovett, 328 U.S. 303, 66 S. Ct. 1073, 90 L. Ed. 1252 (1946), the termination of employment for government employees "found guilty of disloyalty" was characterized as punishment notwithstanding the absence of any specific law making disloyalty a crime. 328 U.S. at 316, 66 S.Ct. at 1079. A similar process of stigmatization occurs with juveniles who are incarcerated because of anticipated imminent criminal behavior notwithstanding the court's inability to predict or specify what crimes they will commit.

The sharp distinction between pretrial detention of juveniles under § 739 and pretrial detention of adults for failure to meet bail is apparent when we consider the criteria traditionally used to distinguish punitive from regulatory impositions:

Whether the sanction involves an affirmative disability or restraint, whether it has historically been regarded as a punishment, whether it comes into play only on a finding of scienter, whether its operation will promote the traditional aims of punishment — retribution and deterrence, whether the behavior to which it applies is already a crime, whether an alternative purpose to which it may rationally be connected is assignable to it, and whether it appears excessive in relation to the alternative purpose assigned...

Kennedy v. Mendoza-Martinez, supra, 372 U.S. at 168-69, 83 S.Ct. at 567-68; also see Bell v. Wolfish, supra, 441 U.S. at 537-538, 99 S.Ct. at 1873. Bail is set for adults for the specific purpose of assuring the defendant's appearance at trial, and as such it may not be "excessive." U.S.Const. 8th Amendment. Excessiveness is determined in relation to this specific purpose. A court may not impose higher bail than appears "reasonably necessary" to effectuate the justifying reason for the restraint. Stack v. Boyle, 342 U.S. 1, 72 S. Ct. 1, 96 L. Ed. 3 (1951).

In contrast to the rather specific delimited purpose for which bail is set, the purposes of § 739 — to protect society from juveniles' criminal conduct — are not sufficiently distinguishable from those of the juvenile justice system as a whole to constitute an "alternative purpose." Moreover, even if they did constitute such, the inability of trial judges to predict which juveniles will commit crimes, see discussion ante pp. 707-714 et seq., means that any given decision to detain cannot "rationally be connected" with the purpose behind it. Thirdly, the stigmatizing of juveniles as prospective or habitual criminals prior to trial is such a departure from basic principles of due process that it "appears excessive" even in relation to the goal of protecting society.

The state cannot escape the obviously punitive nature of detention under § 739 by labelling the statute "preventive" and "regulatory," since "even a clear legislative classification of a statute as `non-penal' would not alter the fundamental nature of a plainly penal statute." Trop v. Dulles, supra, 356 U.S. at 95, 78 S.Ct. at 595. If incarceration under the statute were based exclusively on a founded suspicion that the juvenile had committed crimes, no one would doubt that this constituted punishment. To hold that it is not "punishment" when based on a vague suspicion that the juvenile may commit future crimes would render the applicability of the due process clause in inverse proportion to the arbitrariness of governmental decision-making. The absurdity of such a result is perhaps what inspired Chief Justice Warren to insist that a deprivation did not cease to be punitive because its professed goals were preventive:

It would be archaic to limit the definition of "punishment" to "retribution." Punishment serves several purposes: retributive, rehabilitative, deterrent — and preventive. One of the reasons society imprisons those convicted of crimes is to keep them from inflicting future harm, but that does not make imprisonment any the less punishment.

United States v. Brown, 381 U.S. 437, 458, 85 S. Ct. 1707, 1720, 14 L. Ed. 2d 484 (1965). When detention is imposed solely on grounds which are tantamount to a stigma of criminality, this is "punishment" regardless of the label attached, and regardless of whether it is based on past or predicted future acts.

A statutory scheme which empowers the state to have juveniles incarcerated for as long as 5 days without the state having established a justification for their being held surely constitutes a punitive measure offensive to due process requirements.

Accordingly, I hold that placing a juvenile under disability of restraint pursuant to the arbitrary and subjective evaluation of a judge before the state has been required to show that his being held, even for regulatory reasons, is punitive and in violation of due process.

For these reasons I find § 739(a)(ii), both on its face and as applied, contravenes the rights of petitioners and the class they represent to due process of law as guaranteed by the 14th Amendment to the Constitution of the United States.

IT IS SO ORDERED.

NOTES

[1] For most serious crimes such as murder and arson, members of this group 13 years old and older may be prosecuted as juvenile offenders in the criminal court. N.Y. Penal Law § 10.00(18) (McKinney). This subclass, however, does not concern us.

[2] N.Y. Family Court Act § 739 (McKinney) provides:

(a) After the filing of a petition under section seven hundred thirty-one or seven hundred thirty-two, the court in its discretion may release the respondent or direct his detention. In exercising its discretion under this section, the court shall not direct detention unless it finds that unless the respondent is detained:

(i) there is a substantial probability that he will not appear in court on the return date; or

(ii) there is a serious risk that he may before the return date do an act which if committed by an adult would constitute a crime.

(b) Unless the respondent waives a determination that probable cause exists to believe that he is a juvenile delinquent or a person in need of supervision, no detention under this section may last more than three days (i) unless the court finds, pursuant to the evidentiary standards applicable to a hearing on a felony complaint in a criminal court, that such probable cause exists, or (ii) unless special circumstances exist, in which cases such detention may be extended not more than an additional three days exclusive of Saturdays, Sundays and public holidays.

[3] N.Y. Family Court Act § 742 (McKinney) defines "fact finding hearing" in a case of alleged juvenile delinquency as one "to determine whether the respondent did the act or acts alleged in the petition which, if done by an adult, would constitute a crime." N.Y. Family Court Act § 744 (McKinney) requires evidence to be limited to the competent, material and relevant and a finding that "respondent did the acts or acts alleged must be based on a preponderance of the evidence."

[4] N.Y. Family Court Act § 743 (McKinney) defines a dispositional hearing in cases of alleged juvenile delinquency as one "to determine whether the respondent requires supervision, treatment or confinement."

[5] There are two types of detention facilities — secure and non-secure. A secure facility is defined as one "characterized by physically restricting construction, hardware and procedures." N.Y. Family Court Act § 712(d) (McKinney). A non-secure facility is "characterized by the absence of physically restricting construction, hardware and procedures." N.Y. Family Court Act § 712(e) (McKinney). The secure facility has a more authoritarian atmosphere. The juveniles are subjected to stripsearches, wear institutional clothing and follow institutional regimen. At Spofford, which is a secure facility, some juveniles who have had dispositional determinations and were awaiting placement (long term care) comingle with those in pretrial detention (short term care).

[6] If he was in fact held in custody from July 13 to August 12 before the probable cause hearing, the statutory time constraints appear to have been breached. See N.Y. Family Court Act § 739(b) (McKinney).

[7] In certain designated felony cases, the officer makes no recommendation about release or parole but merely summarizes the record in the case. (Tr. 72).

[8] There was considerable testimony concerning the type of facility to which a juvenile is remanded for detention prior to dispositional hearing. Michael Bigley, director of detention services section, New York State Division for Youth, testified that detention is short term care which may last through dispositional stage pending final placement. Long term care commences at placement pursuant to a dispositional determination. He testified that in 1978, in New York City there were 7,362 initial admissions (of both juvenile delinquents and juvenile offenders) to secure facilities and 50% of the juvenile delinquents admitted to those facilities were there for 3 days or less (Tr. 206-07). In 1979, there were 1,216 admissions of juvenile delinquents to non-secure facilities (Tr. 211-12). In 1979, a total of 8,557 juveniles were admitted to secure facilities; 2,428 whites, 4,501 blacks and 1,386 hispanics. (Tr. 222). The testimony appears to be at odds with the final figures set out in the Report of Chief Administrator (Tables 79, 80, 81, 82) for the calendar year 1979, see text pp. 21-22. If we include all of those detained adding those detained before petition to the figures given for those detained after petition, and include placement with the Commissioner of Social Services, Title 2 and Title 3 placements with the Division for Youth as placement in secure facilities, the total 4,783 does not coincide with 8,557, the figure given by Mr. Bigley at trial.

[9] Table 81 and Table 82, pp. 162-165, 1977 Annual Report for 1977 of the Administrative Board of the Judicial Conference and Office of Court Administration of the State of New York (1978), show that 347 juveniles were detained before the filing of the petition, and 3,546 were detained between the filing of the petition and disposition. If the latter figure does not include the former, as the text assumes, the number of juveniles subject to pretrial detention would be 3,793.

[10] Tables 83 and 84, pp. 166-169. N.Y. Executive Law § 502 et. seq. (McKinney), authorizes the Division for Youth to establish youth centers — camps, residences or other buildings for the care, treatment, education, rehabilitation and guidance of youth between 15 to 18 years of age. The program may include training and service in conservation, civil defense, disaster relief, etc. Youths assigned there may be there only part time. It is to such centers that Title 2 placement refers.

Title 3 placement is provided for in N.Y. Executive Law § 510 et. seq. (McKinney), and these are training schools operated by the Division for Youth for juvenile delinquents under the age of 16 at the time the delinquent act is committed and under 17 at the time of placement. These schools are authorized to offer general rehabilitation programs.

[11] Tables 79 and 80, pp. 86-87, Report of the Chief Administrator of the Courts for the Calendar Year 1979, State of New York (1980).

[12] Ibid.

[13] Tables 81 and 82, pp. 88-89, Report of Chief Administrator of the Courts, op. cit. supra note 11.

[14] Ibid.

[15] Ibid.

[16] Ibid.

[17] Ibid.

[18] Ibid.

[19] Ibid.

[20] Ibid.

[21] Ibid.

[22] Ibid.

[23] Ibid.

[24] Ibid.

[25] Ibid.

[26] Ibid.

[27] Ibid.

[28] I admire the candor of Michael Bigley, who, on being asked whether he would consider pretrial detention probably unwise if the ultimate disposition was parole to the youth's own home or to a foster home, replied that he would not say it was probably unwise but that it was possibly unwise. (Tr. 261).

[29] The short span of pretrial detention makes effective review impossible. Detention is over before the appellate review can be obtained. In People ex rel. Wayburn v. Schupf, and, in this case, motions to dismiss on grounds of mootness were denied because to hold otherwise would have meant insulating § 739 from constitutional challenge. Nonetheless, in reality no juvenile ordered into pretrial detention can test the validity of that adjudication while he is being held in custody pursuant thereto.

[30] Justice Stevens dissenting in Baker v. McCollan, 443 U.S. 137, 154 n. 13, 99 S. Ct. 2689, 2700 n. 13, 61 L. Ed. 2d 433 and accompanying test (1979), pointed to a study of the District of Columbia, Brosi, A Cross-City Comparison of Felony Processing (1979) which showed that 49% of those arrested are never tried at all, with all charges being dropped before trial, and another study, Kamisar, LaFave and Israel, Modern Criminal Procedure (1974) which states that nationally 40% of all adults arrested are released without charges being filed.

[31] CLINICAL PREDICTION

In the face of the literature about clinical predictions, it would be impossible to be sanguine about relying on clinical methods. Feld, supra at 543, asserts that, "In view of the uncertainties and inconsistencies typically associated with social science research, the clear-cut superiority of actuarial methods over clinical methods is startling." Other researchers, Hermann Mannheim and Leslie Wilkins reported that their use of statistical prediction techniques proved to be twice as accurate as clinician's "intuitive" judgment. See Koerin, supra at 53, and see Tr. 314, and Wilkins, passim, Tr. 314-384. In light of the universal pessimism concerning the prospects of making accurate and discrete predictions of future criminal behavior using statistical methods, and the latter's recognized "clear-cut" superiority over clinical methods, the clinician must be disheartened. Schlesinger, supra, noted the disparity between the general claim that prediction of violence was futile, on the one hand, and the claims of 9 studies, on the other, purporting to make such predictions of violent behavior in children. Employing 30 factors identified amongst the 9 studies, Schlesinger attempted to predict violent behavior in 122 children, based on their presentence evaluations from family court in California. He found a very low incidence of violent behavior among the juveniles — only 5.7%; no significant relationship among the predictor variables, clinician's recommendations and criterion behaviors; and no significant relationship between predictions of dangerousness and subsequent violent behavior. Cocozza & Steadman, The Failure of Psychiatric Predictions of Dangerousness: Clear and Convincing Evidence, 29 Rutgers L.Rev. 1084, 1094-1101 (1976), reported on their study which followed up criminal defendants who had been found incompetent to stand trial in New York City. These defendants had been evaluated by psychiatrists to determine the defendant's dangerousness, although no definition of that term was provided. The difference in assaultive behavior between those diagnosed as dangerous and those not so diagnosed was found not to be statistically significant. The authors concluded that the study presented clear and convincing evidence of the clinicians' inability to predict assaultive behavior. Id. at 1098-99. See also Dix, Clinical Evaluation of the Dangerousness of "Normal" Criminal Defendants, 66 Va.L.Rev. 523, 542-44 (1980) (reviewing Cocozza and Steadman study). Examination of leading studies is said to confirm the inadequacy of evidence of the accuracy of clinical assessments. Id. at 544.

Those studies that indicate that it may be possible to develop reliable clinical methods of prediction also indicate how unlikely it would be for the family court judge's evaluations to be reliable. In The Clinical Prediction of Dangerousness, 24 Crime & Delinquency 28 (1978), Cohen, Groth and Siegel reviewed the literature criticizing clinical predictions and discussed two newer studies (Kozol, Boucher and Garofalo studying male sexual offenders over a ten-year period, and Hodges studying dangerous offenders over a three-year period) and concluded that these studies show "considerable improvement over the statistical studies... The two studies demonstrate, we believe, that a conclusion about our capability of predicting dangerousness is premature." Id. at 35.

In The Diagnosis and Treatment of Dangerousness, 18 Crime & Delinquency 371 (1972), Kozol, Boucher and Garofalo reported on a ten-year study of almost 600 convicted male offenders, most of whom had committed a sex crime, compounded in some cases by extreme violence. On the basis of the initial diagnosis, the authors concluded that more than half of the subject offenders were not dangerous. Of these, 8.6% of those who were released subsequently committed "serious assaultive crimes." After receiving treatment for an average period of 43 months, 82 additional patients were recommended for release. Of these, 6.1% subsequently committed serious assaultive crimes. Another 18 patients were released against the clinicians' advice after treatment for an average period of 30 months with a recidivism rate of 27.8%.

The Hodges study, reported in Cohen, et al., at 35, followed up on 447 dangerous offenders for three years, and found that of those recommended by the clinical staff for incarceration who were not incarcerated by the court, 81% subsequently committed another offense (not necessarily a violent crime). Of those committed to the treatment center but later released against staff recommendations, 71% committed a new crime, but of those committed and later released on the staff's recommendations, only 37% were later arrested for a new crime. These newer studies do not show that a family court judge could make reliable predictions. Rather, by highlighting what is required for a reliable clinical prediction to be made, they show why the judge could not be expected to make reliable evaluations along clinical lines. As a preliminary matter, each of these newer studies ranged over a population that was in some respect selected before the clinicians began their work. The offenders studied were a small portion of the total inmate population. Thus, it is unclear how well these results could be generalized to the situation of the family court. In particular, since, on actuarial grounds, the fact that these individuals studied had been convicted makes it more likely that they will be "dangerous." The same cannot be said of the juveniles who were detained, or who risk detention, before trial. Moreover, each of the studies involves an initial classification by the clinical staff on the basis of which a substantial number of offenders are recommended for release. (In the Kozol, et al. study, supra, over half were so released.) Even this initial diagnosis is, in terms of its depth, detail and sophistication well beyond what could be hoped for in the family court setting. In Kozol, supra, the initial diagnosis "is based on clinical examinations, psychological tests, and a meticulous reconstruction of the life history [of the offender] elicited from multiple sources — the patient himself; his family, friends, neighbors, teachers, and employers; and court, correctional, and mental hospital records..." 18 Crime & Delinquency at 383. Similarly, Cohen, et al., describe the norm for clinical predictions:

In evaluating an offender, the clinician bases his or her prediction on an extended period of study, makes use of a variety of professional personnel, and consults diverse clinical, behavioral, and social sources of data. Typically, this involves a minimum of a sixty-day period.

18 Crime & Delinquency at 30-31.

Finally, these studies are apparently open ended in their definition of recidivism. As was noted ante, there is a wide discrepancy between the violence that may be indicated in general and that may be indicated for the relevant period of pretrial detention. There is no basis for supposing that a family court judge could predict reliably the juveniles who could be dangerous within the pretrial period.

[32] STATISTICAL PREDICTION

In reporting on both the current developments in statistical prediction and their own study, Wenk, Robison and Smith stated "The present state of the art holds little promise for the development of a prediction instrument that would warrant implementation in actual preventive or correctional programs." Can Violence Be Predicted?, 18 Crime & Delinquency 393, 401 (1972). While an expectation of increasing statistical sophistication might lead one to suppose that, in the fullness of time, a reliable predictive method could be achieved, there is no basis on which to suppose that the family court judge's evaluations could be expected to succeed where other attempts have failed.

For a number of reasons, the study reported in Preventive Detention: An Empirical Analysis, 6 Harv.Civ.Rights Civ.Lib.Rev. 289 (1971) ("Harvard study") is particularly telling. This study examined the District of Columbia Court Reform and Criminal Procedure Act of 1970, Pub. L.No.91-358, 84 Stat. 473 (July 29, 1970) ("preventive detention legislation"), which provides for the "preventive pretrial detention of arrested persons felt to pose a threat to the safety of the community." More specifically, under this law

any defendant charged with a "dangerous" crime, with obstructing justice, or with a "violent" crime if certain conditions are met, may be held for pretrial detention hearing, at which a judicial officer is to determine whether any form of release can satisfactorily protect the community.

Id. at 303-04 (footnotes omitted). If the officer does not so conclude, he may order the defendant held for up to sixty days. In deciding which criminal defendants pose so great a danger to the community that they should be detained without bail, the statute charges the judicial officer to

take into account such matters as [1] the nature and circumstances of the offense charged, [2] the weight of the evidence against such person, [3] his family ties, [4] employment, [5] financial resources, [6] character and mental conditions, [7] past conduct, [8] length of residence in the community, [9] record of convictions, and [10] any record of appearances at court proceedings, flight to avoid prosecution, or failure to appear at court proceedings.

Id. at 309. The statute does not assign a relative importance to these various factors.

The Harvard study tested the reliability of two numerical dangerousness scales, each of which combines in one index the particular defendant's score on all of the variables. The first scale (DS-1) was fashioned by assigning weights to the variables on the basis of a subjective assessment of their importance. The second scale (DS-2) "was constructed using a statistical technique, readjusting the weights of the variables on the basis of their actual correlations to recidivism in the sample." Id. at 310. Each scale could be used by selecting a cut-off score, above which the individual would be detained. For DS-2, no cut-off point existed for which more recidivists than non-recidivists would be detained. Thus, overprediction is a substantial problem. The lowest false positive to true positive ratio would be 1.5 to 1.0. To detain all true positives, 225 non-recidivists would be included along with the 41 recidivists. A point at which 63% of the recidivists would be retained would still represent a false positive to true positive ratio of 1.5 to 1.

Much more startling are the results of the unsophisticated DS-1 scale. At its best, this scale would select correctly only about 30% of the time — that is, 7 false positives for every 3 true positives detained. To ensure that all recidivists were detained, in effect 8 nonrecidivists would have to be detained for every recidivist. Only by detaining fewer recidivists can this high degree of overprediction be avoided: When 40% of the recidivists were detained, the ratio was considerably reduced but was still high, a ration of 2.88 to 1.

Neither of these sets of results actually measure what would happen in pretrial detention. First, the accurate and detailed information used in the study is not normally available to the judicial officer in a § 739 proceeding. Second, the Harvard study's definition of recidivism is not limited in time but relates to arrests and convictions occurring perhaps months following initial arrest and release. An accurate measure of the effectiveness of pretrial detention must consider the number of persons who would have committed crimes during the detention period only, and under this measure, the 60% accuracy figure for DS-2 falls to about 15%, and the 30% figure for DS-1 falls to about 5%.

In addition to the substantial body of literature in agreement that no reliable statistical prediction has been devised, the Harvard study indicates why the family court judge's determinations would be unreliable. Even to the extent that a set of variables could be attributed to the judge's evaluation, they would not have been weighted in a manner to enhance statistical reliability, and thus, the expectable range of accuracy would be that of the DS-1 scale. Further, since the judge's assessments bear only on pretrial detention, the expectable range of accuracy would likely be very low. Finally, the difficulty with overprediction would be substantial.