Full opinion text
WILLIAM E. DOYLE, Circuit Judge. PART I In this case, the plaintiffs, who are inmates of the Federal Correctional Institute at Englewood, Colorado, have challenged the interpretation that the Bureau of Prisons has given to the Federal Youth Corrections Act, 18 U.S.C. § 5005, et seq., hereinafter called the YCA. There are eleven petitioners appellees who challenge the alleged failure of the United States Bureau of Prisons to observe the requirements of the Act. A similar challenge has been filed against the United States Parole Commission in Watts v. Hadden. The issue in the present case is whether under the Act the inmates have to be segregated, such that all inmates within the institution, with perhaps minor exceptions, have to be persons who have been sentenced under the Youth Corrections Act. The Bureau of Prisons’ position is that so long as they observe the treatment and training provisions of the Act, it is perfectly permissible for the Bureau to have older prisoners within the institution. The petitions here were filed under 28 U.S.C. § 2241(c)(3), which is a federal habe-as corpus statute. Each of the petitioners was confined at the Federal Correctional Institute at Englewood under YCA sentence. They alleged that they were not being segregated from adult offenders nor receiving treatment and supervision as required by the YCA. A trial was held in the United States District Court for the District of Colorado and on April 20, 1979 a memorandum opinion was issued by United States District Judge Richard Matsch. This opinion contains extensive findings of fact together with an interpretation of the YCA. See Watts v, Hadden, 469 F.Supp. 223 (D.Colo. 1979). In general, the court concluded that the petitioners were being held in custody in violation of the YCA. But rather than ordering release from custody, the court directed the Bureau of Prisons defendants to submit a written plan which would implement the YCA as interpreted by the court. The Bureau of Prisons plan called for classification and treatment of YCA offenders at the Englewood FCI. YCA offenders were to have segregated living quarters and segregated treatment programs. At all other times, YCA inmates would not be segregated from other inmates. Thus, the plan that was submitted by the Bureau of Prisons would have allowed integration of the institution itself apart from the living quarters, housing inmates who were not sentenced under the YCA. On January 17, 1980, pursuant to Rule 54(b) of the Fed.R.Civ.P., the trial court entered a final judgment with regard to the issues involving the Bureau of Prisons. The judgment of February 5, 1980, which is referred to as the amended judgment, declares that the plan submitted by the Bureau of Prisons is in compliance with the YCA, except that it does not provide for complete segregation of YCA offenders. The judgment directs respondents to place those petitioners still in prison in a facility designed and operated so as to separate them from non-YCA offenders. It is this judgment that the Bureau of Prisons has appealed and which is in question. I. Objectives and Provisions of the Youth Corrections Act The enactment of this Act by Congress occurred in 1950. Its purpose was to provide an alternative sentencing program for federal judges for persons who were under the age of 22 years. The intent was to promote rehabilitation of convicted youths believed to show promise of becoming useful citizens, and to thus “avoid the degenerative and needless transformation of many of these young persons into habitual criminals.” Congress, in the legislative history, showed that its view was that existing methods of treatment of criminally inclined youths were not solving the problem and that mixing of youth offenders with older criminals was resulting in criminalization of the young. By herding youth with maturity, the novice with the sophisticate, the impressionable with the hardened, and by subjecting youth offenders to the evil influences of older criminals and their teaching of criminal techniques, without the inhibitions that come from normal contacts and counteracting prophylaxis, many of our penal institutions actively spread the infection of crime and foster, rather than check, it. 1950 U.S.Code Cong.Service 3983, 3985. Thus a principal purpose, as expressed by the Congress, was the segregation of these youths from older criminals who can be characterized as hardened. In the YCA, Congress sought to address the problem with a comprehensive scheme of pre-sen-tencing study, sentencing options, individualized placement and treatment, and supervised release. There have been various amendments over the years which primarily touch upon administrative and organizational schemes. However, the basic provisions, including those here relevant, have remained unchanged. Under 18 U.S.C. § 5010, a judge sentencing an individual who is between 18 and 22 years of age has a number of sentencing options. The court may suspend sentence and place the youth offender on probation. § 5010(a). In lieu of any other sentence of imprisonment provided for an offense, the court may sentence the youth offender to an indeterminate sentence of up to four years confinement and up to two years on conditional release. §§ 5010(b) and 5017(c). It is undisputed that in some cases these provisions can and do result in youth offenders spending longer periods in confinement than they would have if sentenced as adults. In the event that a term longer than six years is provided for the offense by another statute, the court may choose to sentence the youth to an indeterminate sentence for such longer period. §§ 5010(c) and 5017(d). Under 18 U.S.C. § 5010(d) the court may sentence a youth offender under any other applicable penalty provisions, but only if the court finds that the youth will not benefit from treatment under the YCA. This is a finding which is to be expressly made on the record. Dorszynski v. United States, 418 U.S. 424, 94 S.Ct. 3042, 41 L.Ed.2d 855 (1974). Offenders between 22 and 26 may be sentenced pursuant to the YCA, but only if the court finds reasonable grounds to believe the young adult offender will benefit from treatment provided by the Act. 18 U.S.C. § 4216 (formerly § 4209). Thus, in the statutes there is a strong thread that runs throughout the YCA that shows a legislative preference for sentencing under YCA for offenders under 22 years of age. While offenders older than 22 may be sentenced thereunder in special cases, it was not the preference of Congress that they be so sentenced as it was for the younger offenders. S.Rep.No.2013, 85th Cong., 2nd Sess., reprinted in [1958] U.S.Code Cong, and Ad.News 3891, 3892. Finally, § 5010(e) allows the court to commit the youth offender for study and observation in order to assist the court in determining whether treatment under the Act will prove beneficial. After the pronouncement of sentence under the YCA, every offender must first be sent to a “classification center or agency” for a complete study and recommendation regarding treatment. 18 U.S.C. § 5014. On receipt of the recommendation, the Director of the Bureau of Prisons has three options. Conditional release under supervision can be recommended to the Parole Commission. Or, the Director may transfer the youth offender to an agency or institution for treatment. Finally, the Director may “order the committed youth offender confined and afforded treatment under such conditions he believes are best designed for the protection of the public.” 18 U.S.C. § 5015(a). 18 U.S.C. § 5015(b) further permits the Director to “transfer at any time a committed youth offender from one agency or institution to any other agency or institution.” ’ 18 U.S.C. § 5006(f) defines treatment to mean “corrective and preventive guidance and training designed to protect the public by correcting the antisocial tendencies of youth offenders.” 18 U.S.C. § 5011 is the central provision of the Act regarding treatment. It provides: Committed youth offenders not conditionally released shall undergo treatment in institutions of maximum security, medium security, or minimum security types, including training schools, hospitals, farms, forestry and other camps, and other agencies that will provide the essential varieties of treatment. The Director shall from time to time designate, set aside and adapt institutions and agencies under the control of the Department of Justice for treatment. Insofar as practical, such institutions and agencies shall be used only for treatment of committed youth offenders, and such youth offenders shall be segregated from other offenders, and classes of committed youth offenders shall be segregated according to their needs for treatment. The Director is required to make periodic examinations and reports on committed youth offenders to the Parole Commission. 18 U.S.C. § 5016. The full Commission is charged with determining when and under what conditions a committed youth offender should be released. 18 U.S.C. §§ 5005, 5017. The legislative history also reveals that the YCA scheme for sentencing, treatment and release was founded on two key sources of information. An “extensive and thorough” study of punishment for crime, which spanned several years, was conducted by a distinguished committee of the Judicial Conference of the United States. A subcommittee gave particular attention to the problem of youth offenders. Reliable statistics demonstrate, beyond possible doubt, that the period in life between 16 and 23 years of age is the focal source of crime. It is during that period that habitual criminals are spawned. Sociologists and psychiatrists tell us that special causations which occur in the period between adolescence and manhood are, in a large measure, responsible for antisocial conduct trends manifested by persons in that age group. Most of the causes which contribute to anti-social conduct of youth offenders in the period between adolescence and maturity disappear when the youth reaches full maturity. The problem is to provide a successful method and means for treatment of young men [convicted of crime] —a method and means that will effect rehabilitation and restore normality, rather than develop recidivists. 1950 U.S.Code Cong.Service 3983, 3984-5. A second major source of information was the Borstal System, which had been in successful operation in England for several decades. 1950 U.S.Code Cong.Service 3983, 3987-89. The Borstal System began in 1894 as an experiment. It was carried on in a prison in which one wing was set aside for offenders age 16 to 23. By 1904 an entire institution had been devoted to intensive rehabilitation programs for youths. By 1950 thirteen institutions of varying types had become involved in the English system, with each institution providing a different type of program speciality. The system included careful study and classification to an appropriate institution, individualized treatment and assistance, and continued counseling after release. In light of the success of this system, the legislative history asserts that the YCA is “not experimental,” that the YCA plan “will work,” and that through the plan “more than 70 per cent [of youth offenders] can be rehabilitated.” 1950 U.S.Code Cong.Service 3983, 3985, 3987, 3993. Section 5012 provides that no YCA commitments are to be made until the Director has certified that proper and adequate treatment facilities are available. In 1954, the Director notified the courts in seven circuits that a facility at Ashland, Kentucky was being converted into a classification center and treatment facility. In 1956, courts in the Eighth, Ninth and Tenth Circuits were notified that the facility at En-glewood, Colorado was being similarly converted. II. Trial court findings as to changes in the segregation of youth offenders Following the receipt of evidence the trial court entered judgment, together with a memorandum opinion and order dated April 20,1979. Watts v. Hadden, 469 F.Supp. 223 (D.Colo.1979). The trial court found that Englewood was operated as a Federal Youth Center from 1956 to 1976. Some adults were housed at Englewood during this period but the facility was used primarily for' custody of YCA and Federal Juvenile Delinquency Act offenders. In 1976, the “Federal Youth Center” designation was eliminated throughout the system; En-glewood became an FCI. The court further found that the Bureau of Prisons had determined that classification of prisoners to institutions should no longer be based on an offender’s YCA status. It was determined that Englewood was to be used to house adults and YCA offenders in accordance with a new classification system under which each federal facility was rated according to the level of security provided. There are six levels of security, level six being the highest and most secure; Englewood is rated as a level three or medium security institution. Each offender is rated to determine his security requirements under a point system based i>n history of escapes, history of violence, type of detainer, severity of current offense, expected length of confinement and type of prior commitment. All offenders, whether YCA or not, are assigned to institutions based on these security ratings. John T. Hadden, Warden at Englewood, testified that judicial recommendations, location of an inmate’s home and separation of inmates for reasons of safety are other factors which are considered in placing offenders. It is not disputed that, at least in the Western Region, the Bureau of Prisons no longer makes any effort to place YCA offenders in institutions which exclusively or primarily house those who are sentenced under the Act. The trial court found that each of the petitioners was designated for confinement at Englewood under the new classification system. None of the petitioners were ever sent to a classification center or agency for study or treatment recommendation before being sent to a prison facility, nor have any of the petitioners ever been placed in an exclusively YCA facility. The record, together with the findings of the trial court, brings home the fact that several changes have occurred at Engle-wood as the result of orders of the Bureau of Prisons since this new classification system went into effect. The number of YCA inmates has declined significantly; many of the YCA inmates have been released or sent to institutions of lesser security levels. The population of the institution has increased to almost 500 although the design capacity is 350. Englewood now houses male inmates from ages 18 and up. The average age of YCA inmates is about 20, while the average age of those sentenced as adults is about 28. The emphasis in the programs provided for inmates has changed to meet the changing needs of an older population. Because fewer inmates lack the equivalent of a high school education, the programs now emphasize advanced education and vocational training. The size of the educational staff and the number of programs have been reduced due to a lowered demand. There is increased emphasis on a prison industries program, participation in which earns funds for use at the prison commissary. Many of the course offerings, particularly those at the high school level, are offered only during the daytime hours, requiring inmates to choose between those courses and the opportunity to earn money in the industries program. Although the prison staff recommends programs to inmates, no inmate is required to participate in any program. Each inmate may select his own activity without adverse consequence. Thus, no distinction is made between YCA inmates and non-YCA inmates in program offerings, and inmates are not segregated in their daily activities. III. Reasons for integrating youth offenders with older offenders The decision made by the Bureau of Prisons to integrate YCA and other offenders at Englewood and throughout the Western Region was based largely on the experience, reflected in a study conducted at the FCIs in El Reno, Oklahoma and Texarkana, Texas, that mixing of older and younger offenders resulted in decreased violence. Also, it had become necessary to remove all juvenile delinquents from facilities where they came in contact with older offenders, 18 U.S.C. § 5039. This resulted in a reduction of the population in Englewood. In recent years, the total inmate population in federal facilities has increased from 20,000 to approximately 30,000, resulting in overcrowding. At the same time, the Engle-wood facility remained relatively uncrowded. The Regional Director of the Western Region indicated that the differences between the Federal Youth Centers and other federal facilities had become less noticeable over the years. This was due largely to the fact that education and self-improvement programs had been extended to inmates throughout the prison system. The Bureau of Prisons had found that some younger and relatively unsophisticated offenders were being sentenced under adult provisions rather than the YCA. At the same time, relatively sophisticated criminals were receiving YCA sentences, particularly in the young adult offender group. This had resulted in the Bureau of Prisons placing less sophisticated young offenders in the same facilities, whether sentenced under the YCA or not, while more sophisticated YCA offenders had to be removed and placed in more secure facilities. Notwithstanding this, until the new classification system was started, YCA offenders had been, for the most part, placed in one of the six facilities which primarily housed youth offenders. At the present time, YCA offenders are scattered among twenty institutions in the federal system. The trial court found, with substantial support, that a primary reason for the integration of YCA and other offenders was a shift in penological philosophy within the Bureau of Prisons. At the time the YCA was passed the “medical model” was widely accepted by criminologists within and outside the Bureau. Under this theory, it is thought that an individual can be evaluated and treated so as to, in effect, “cure” criminal tendencies. The testimony of Warden Hadden indicated that the philosophy of the Bureau had shifted in recent years. Now it is thought that rehabilitation cannot occur unless it is desired by the inmate. The trial court found that the most important purposes of confinement are now believed to be deterence and retribution. The policy of “optional programing” provides inmates with various self-improvement opportunities from which they may select. As noted by the trial court, the commendable candor of the witnesses, as well as policy statements introduced as exhibits at the trial of this action, demonstrated the Bureau of Prisons’ view that the YCA has outlived its usefulness. Accordingly, the Bureau abandoned attempts to segregate and provide special treatment for YCA offenders. For reasons of administrative flexibility, institutional discipline, and, apparently, disillusionment with the optimistic rehabilitative approach embodied in the YCA, the Bureau began treating YCA offenders like all other offenders. IV. Analysis of the Bureau of Prison Plan In response to court decisions a separate YCA residential unit was established at En-glewood for the purpose of segregating YCA offenders from other inmates during sleeping hours. YCA and other offenders continue to share all other facilities and activities on an unsegregated basis. In the April 20, 1979 memorandum opinion and order, the trial court found that mandatory requirements of the Youth Corrections Act were being violated. Specifically, the trial court found that there was a lack of classification centers or agencies, segregated institutions, examination and reexamination of YCA offenders, and treatment according to individual needs. The court recognized that the precise manner in which the statutory requirements are to be implemented is within the discretion of the Bureau of Prisons; the trial court directed the Bureau to submit a plan for implementation of the YCA as interpreted by the court. Under the Bureau of Prisons plan Engle-wood is to be designated as a YCA classification center for the Rocky Mountain and Upper Plains region. The plan calls for a thorough study of each offender and a report within 30 days. If the offender’s needs for treatment are not met by available programs, he will be transferred to a YCA unit in a more appropriate institution. The plan further provides for a thorough review of each YCA offender’s progress every 90 days, with reclassification to a more appropriate facility a continuing option. A treatment program is to be developed and periodically reviewed for each YCA offender by the YCA unit team. Each YCA offender is to be required to participate in an assigned treatment program involving forty hours per week of activities. The plan calls for YCA offenders to be segregated from other offenders during treatment hours with one exception. If the staff determines at initial classification that “basic treatment needs” have been met, a YCA inmate will be permitted to request assignment to an industrial job. Such assignments would involve mingling of YCA and other inmates. Industrial assignments may be full or part time, provided work will not interfere with the treatment program. The plan lists a number of programs which are available to YCA inmates. It indicates that additional programs are to be developed as new treatment needs are identified. The programs listed are: adult basic education, adult secondary education, post-secondary education, special reading and math, physical education, vocational training, industrial apprenticeship, counselling, town hall, social activities, outward bound, and pre-release transfer to half-way houses for study or work release. Part of the plan calls for a YCA housing unit which is off-limits to non-YCA inmates. Thus, while they are in their unit and during treatment hours, YCA inmates are segregated from non-YCA inmates under the Bureau plan. At all other times, however, YCA inmates would under the plan be mixed with non-YCA inmates. Mixed activities would include the dining hall, movies, hobby shop, inmate organizations, chapel, visiting room, hospital, law library, gymnasium, auditorium, music and outdoor recreation. Written requests for separate use of such facilities by YCA inmates would be accommodated, if possible. The reasons given for rejection of separate YCA institutions are: First, many YCA inmates would have to be placed further from their homes and families if fewer institutions are available. Second, various types of YCA offenders would have to be mixed within each institution, a result “precisely contrary to the legislative intent.” Finally, loss of management flexibility in administering the prison system as a whole would occur. The plan also notes that greater segregation of YCA and non-YCA inmates within Englewood was considered and rejected. The reasons for this position are: The Bureau points to research suggesting that mixing of age groups reduces violence, because older inmates exert a “stabilizing” influence on YCA offenders. Secondly, the plan asserts that further segregation at En-glewood, “while physically possible ... would decrease the opportunities available to all .... ” After the court had received the plan, objections from the petitioners were also received and an evidentiary hearing was held. On January 17, 1980 the trial court entered a final judgment on the issues concerning the Bureau of Prisons pursuant to Rule 54(b), Fed.R.Civ.P. The trial court found the plan to be in substantial compliance with the YCA with regard to classification and treatment. The question now to be decided is whether complete segregation is required. The respondents contend that such complete segregation is impracticable and would be disruptive to large numbers of offenders. More particularly, they have reported that there are approximately 2,200 YCA offenders in the federal prison system and that to place all of them as close as possible to their homes but in institutions used only for such inmates would require some five institutions for males and one for females. As I indicated in my earlier Memorandum Opinion, the difficulty and expense involved in providing complete segregation of youth offenders from all other inmates are not legal justifications for the failure to perform the duty which the Congress made explicit. The qualifying phrase “insofar as practicable” [sic] is limited to the unusual and unforeseen circumstances required for particular inmates or groups of inmates as indicated in Brown v. Carlson, 431 F.Supp. 755 (W.D.Wis. 1977). Accordingly, it must be concluded that the plan is inadequate with respect to segregation. The court directed that the defendants proceed to implement the plan, and further directed that the petitioners be placed in a facility designed and operated so as to separate them from non-YCA inmates. The Bureau of Prisons defendants moved for an amendment of the judgment on the ground that total segregation and implementation of the plan at Englewood would be mutually inconsistent. If total segregation were achieved, implementation of the plan would be unnecessary. The Bureau further argued that whether petitioners should be segregated at Englewood, or elsewhere, remained within the Bureau’s discretion. On February 5, 1980 the trial court amended its judgment; by declaratory judgment, the court indicated that the plan submitted by the Bureau was in compliance with the YCA, except with regard to failure to provide for total segregation of YCA inmates. The court reaffirmed its order requiring segregation of petitioners within ninety days. Y. Issues on Appeal The Bureau has appealed only that portion of the trial court’s order which required complete segregation of the offenders. On May 5, 1980, the respondents moved for a stay of the trial court’s order pending appeal; this motion was denied May 20th. The Bureau of Prisons apparently concedes that YCA inmates must, at a minimum, receive classification and segregated treatment programs as called for in the plan submitted to the trial court. However, the plan has not been implemented, pending disposition of the present appeal. The question before this court is to what extent does the YCA require segregation of YCA offenders from non-YCA offenders? The trial court was of the view that segregation is mandated by the Act except in unusual and unforeseen circumstances. Appellants concede that 18 U.S.C. § 5011 makes segregation of YCA offenders the norm. The Bureau argues, however, that the qualifying phrase “insofar as practical” grants the Bureau discretion to place YCA offenders in unsegregated institutions and permits mixing of inmates outside residential units and treatment programs as a matter of standard operating procedure. In light of the language of the statute and the legislative history, we disagree with such an interpretation of the statute. First, the language of the statute itself, and particularly § 5011, is to the contrary. See, Lewis v. United States, 445 U.S. 55, 100 S.Ct. 915, 63 L.Ed.2d 198 (1980); Busic v. United States, 446 U.S. 398, 100 S.Ct. 1747, 64 L.Ed.2d 381 (1980). The first sentence of § 5011 provides that committed youth offenders are to undergo treatment in various types of facilities, including maximum, medium and minimum security institutions. The second sentence directs that institutions and agencies are to be designated, set aside and adapted for such treatment. The third sentence reads: “Insofar as practical, such institutions and agencies are to be used only for treatment of committed youth offenders, and such youth offenders shall be segregated from other offenders, and classes of committed youth offenders shall be segregated according to their needs for treatment.” The Bureau relies upon this third sentence for the proposition that its plan for partial segregation of youth offenders complies with the YCA. Before we undertake to construe the phrase “insofar as practical”, we will consider the question of the proper interpretation of the third sentence of § 5011. This sentence permits the use of the same facility for YCA and adult offenders, provided that segregation is not “practical.” One court has held that the third sentence of § 5011 allows the Bureau to deviate from complete segregation only with regard to use of a single facility for adults and youth offenders; segregation of youth offenders from other offenders within institutions has been held to be mandatory in all cases. United States ex rel. Dancy v. Arnold, 572 F.2d 107 (3rd Cir. 1978). There the court was of the view that the phrase “insofar as practical” modifies only the first clause of the sentence, leaving the requirement that youth offenders shall be segregated from other offenders absolute and unaffected by considerations of practicality. Two courts have reached a different result from that in Dancy. Brown v. Carlson, 431 F.Supp. 755 (W.D.Wis.1977); Johnson v. Bell, 487 F.Supp. 977 (E.D.Mich.1980). The Brown court concluded that “insofar as practical” qualifies both the first and second clauses of the third sentence of § 5011. The court in Brown clarified its position by saying that: [I]f and when it is not practical to house one or more YCA offenders in an institution or agency used only for the treatment of YCA offenders, and the YCA offender or YCA offenders are housed with non-YCA offenders, then, insofar as practical, the two categories of offenders are to be segregated from one another within the institution or agency in which they both are housed. The Brown court also makes it clear that any deviation from the norm of segregation may occur only “occasionally, in the presence of unusual and unforeseen circumstances, and only for so long as necessary.” 431 F.Supp. at 773. Brown decided that the Bureau retained some discretion with. regard to both segregation of institutions and segregation within institutions. This conclusion was based on consideration of unusual situations which might arise. As an example of possible permissible use of mixed institutions, the court noted that a special training program involving unusually expensive equipment and specialized instructors might require use of a single facility by YCA and non-YCA offenders. Moreover, while segregation of training classes within the facility might not be practical, the statute would require segregation at all other times to the extent practical. Another possibility mentioned by the Brown court was a semi-permanent exception to the use of segregated institutions. Where a consistently small number of YCA offenders require housing in a particular type institution, such as a maximum security facility, the Bureau would be allowed to place YCA offenders in such an institution with other offenders, provided that the YCA offenders are separated from other offenders within the institution “insofar as practical.” The question of whether the Dancy or the Brown/Johnson interpretation is correct is not free from doubt. The Dancy court relied on a statement in the legislative history which reads: The Director is required to designate, set aside and adapt institutions and agencies under the control of the Department of Justice for treatment. Such institutions and agencies are to be used only for the treatment of youth offenders, so far as practicable; and youth offenders are to be segregated from other offenders, and the classes of offenders are to be segregated according to their needs for treatment. 1950 U.S.Code Cong.Service, 3983, 3986. However, the phraseology used in the statute itself does not mirror the above language. Usage and ordinary understanding indicates that the phrase “insofar as practical” modifies all of the succeeding clauses of the sentence. Furthermore, such an interpretation appears to be more desirable and allows the Bureau some reasonable, limited discretion to deviate from the norm of segregation within institutions, permitting responses to the unusual types of situations as identified by the court in Brown. Concededly, it would be unfortunate if the Bureau was precluded from placing YCA offenders in a special training program offered only at a single facility because complete segregation from other inmates could not be achieved. We are not going to give the Act such an extreme interpretation. As we read § 5011, the Bureau has some discretion, although limited, to deviate slightly from the norm of segregation with respect to placement of inmates in non-segregated institutions and with respect to segregation of inmates within institutions. Acknowledging that the Bureau has some discretion, albeit slight, regarding segregation of youth offenders does not end our inquiry. The Bureau argues here that this discretion permits abandonment of any effort to maintain separate YCA facilities based on the Bureau’s determination that administrative considerations and correctional goals indicate that separate facilities are impractical. Again we have to reject this view of the statute. The language of the statute clearly contemplates a whole range of institutions which are to be designated, set aside and adapted for treatment of youth offenders. The duty to so designate, set aside and adapt facilities is not qualified by any considerations of practicality. As we have noted, the legislative history would indicate that Congress relied on the experience in England with the Borstal System, a system which in 1950 embraced some thirteen institutions of varying security levels and with differing programs of rehabilitation. It may have been the intention of Congress that the development of the system of youth facilities would take place gradually. It is now thirty years since the YCA was enacted. The record shows that not one institution is presently designated, set aside or adapted for YCA offenders. There is no alternative but to hold that this situation is completely out of harmony with the statute and the legislative history. The very first step in the treatment program for youth offenders is set forth in the statute. It directs that institutions shall be set aside for youth offenders. It is only within the context of a comprehensive system of institutions and agencies devoted to youth offenders that any deviation from the norm would be tolerated. Unquestionably segregation of youth offenders was a key element of the rehabilitation program which was envisioned by Congress when it enacted the YCA. An integral part of the treatment program was the segregation of committed persons, insofar as practical, so as to place them with those similarly committed, to avoid the influence of association with the more hardened inmates serving traditional sentences. Dorszynski v. United States, supra [418 U.S.] at 434 [94 S.Ct. at 3048]. See also, Durst v. United States, 434 U.S. 542 [98 S.Ct. 849, 55 L.Ed.2d 14] (1978). Courts which have faced challenges to YCA sentences which are potentially longer than comparable adult sentences have repeatedly emphasized the treatment which is afforded to the youth offender who is committed under the Act. Rogers v. United States, 326 F.2d 56 (10th Cir. 1963); Harvin v. United States, 445 F.2d 675 (D.C.Cir.), cert. denied, 404 U.S. 943, 92 S.Ct. 292, 30 L.Ed.2d 257 (1971); Suggs v. Daggett, 522 F.2d 396 (10th Cir. 1975). In an oft-quoted passage, the Chief Justice of the Supreme Court, who was then Judge Warren E. Burger, observed: [T]he basic theory of the [YCA] is rehabilitative and in a sense this rehabilitation may be regarded as the quid pro quo for a longer confinement but under different conditions and terms than a defendant would undergo in an ordinary prison.... [T]he YCA “provides for and affords youthful offenders, in the discretion of the judge, not heavier penalties and punishment than are imposed upon adult offenders, but the opportunity to escape from the physical and psychological shocks and traumas attendant upon serving an ordinary penal sentence while obtaining the benefits of corrective treatment, looking to rehabilitation and social redemption and restoration.” Carter v. United States, 306 F.2d 283, 285 (D.C.Cir. 1962), quoting Cunningham v. United States, 256 F.2d 467, 472 (5th Cir. 1958). Where YCA offenders are not given special treatment in accordance with the law but instead have the same terms and conditions of confinement as other offenders, the imposition of a sentence which tends to be longer on youth offenders raises a potential stumbling block of constitutional dimension. United States v. Lowery, 484 F.2d 457 (3rd Cir. 1973); United States v. Torun, 537 F.2d 661 (2nd Cir. 1976); cf., United States ex rel. Sero v. Preiser, 506 F.2d 1115 (2nd Cir. 1974), cert. denied 421 U.S. 921, 95 S.Ct. 1587, 43 L.Ed.2d 789 (1975) (New York law). We are not suggesting that total segregation of YCA inmates must be maintained at all times and under all circumstances as a constitutional matter. Nor are we suggesting that the Bureau of Prisons’ hands are tied from extending rehabilitative programs to non-YCA inmates. All such efforts by the Bureau, regardless of the nature of the inmates’ sentences, are commendable and are to be encouraged. We do say that segregation of youth offenders is directly related to and intertwined with the rehabilitative objective of the YCA. We have previously noted that the legislative history makes it clear that Congress viewed separation of youth offenders from older, more sophisticated inmates as the crucial initial step in preventing transformation of the youths into hardened criminals. We note, as well, that the record in this case suggests that provision of programs geared to youth offenders is in fact facilitated by placing youths in separate institutions. Testimony indicated that the programs at Englewood have changed in recent years to meet the needs of an older population. This situation is out of hdrmony with the Congressional intent that agencies and institutions should specialize in treatment of YCA offenders and provide rehabilitation programs which are geared to the needs of YCA offenders. The Bureau of Prisons has argued that an interpretation of a statute by the agency charged with administering it is entitled to be followed unless clearly wrong. E. g., Red Lion Broadcasting Co., Inc. v. Federal Communications Comm., 395 U.S. 367, 381, 89 S.Ct. 1794, 1801, 23 L.Ed.2d 371 (1969); Brennan v. Occupational Safety & Health Comm., 513 F.2d 553 (10th Cir. 1975). Furthermore, the Bureau points out that courts must give proper regard to the Bureau’s expertise and discretion in addressing concerns of penal administration. Procunier v. Martinez, 416 U.S. 396, 94 S.Ct. 1800, 40 L.Ed.2d 224 (1974); Bell v. Wolfish, 441 U.S. 520, 99 S.Ct. 1861, 60 L.Ed.2d 447 (1979). However, the Bureau of Prisons does not have a blank check. It is required to follow the directions that have been laid down by Congress and if there is to be a change in the law, Congress should make the change. Difference to administrative interpretations is inappropriate where the result is clearly inconsistent with Congressional intent. Espinoza v. Farah Mfg. Co., 414 U.S. 86, 94 S.Ct. 334, 38 L.Ed.2d 287 (1973); United States ex rel. Dancy v. Ar nold, supra. We must assume that the legislative purpose is set forth in the clear language of the statute. We have to give the statute such construction as will achieve the express intent of the legislature. Richards v. United States, 369 U.S. 1, 82 S.Ct. 585, 7 L.Ed.2d 492 (1962); Kokoszka v. Bel-ford, 417 U.S. 642, 94 S.Ct. 2431, 41 L.Ed.2d 374 (1974). In the light, then, of the express terms of the statutes in question, and considering the nature and character and program of the Youth Corrections Act, we are of the opinion that the Bureau’s interpretation is contrary to the express language and also to the purpose and spirit of the Act. In our view, the YCA requires the Bureau of Prisons to establish complete segregation of youtfi offenders from other offenders as the norm. The statute directs the Bureau in no uncertain terms to designate, set aside and adapt institutions for treatment of youth offenders. Only within this framework, in which complete segregation of youth offenders is the usual practice, may occasional aberrations be allowed for reasons of practicality. We are not in a position to define any such exceptions at this time, but such situations are going to arise from time to time and if they are within the context of substantial compliance with the segregation requirements of the YCA, they will pass muster. But it cannot be argued that such compliance does not exist at present. It is clear that the statute does not grant the Bureau discretion to determine that the entire legislative scheme of separate youth institutions is impractical. Discretion granted in § 5011 is limited by the specific directives of that section and by the overall scheme of classification and treatment called for in the YCA. The Director’s discretion regarding placement and transfer of YCA offenders, 18 U.S.C. § 5015, is similarly limited. The Bureau’s reasons for abandonment of segregated facilities and for partial abandonment of segregation within facilities can only be addressed to Congress. The Bureau has apparently concluded that the YCA approach is not the best way to handle offenders. But the Act and the legislative history are not to be ignored. These documents make it clear that Congress was convinced that segregation of youth offenders and earnest efforts to rehabilitate them would pay tremendous social dividends. In abandoning efforts to segregate youth offenders, the Bureau has substituted its judgment regarding desirable correctional objectives for the judgment of Congress. It may indeed be desirable to place committed offenders near to home and family; it may be possible to achieve this goal, at least in part, by designation of an adequate number of youth facilities. In any event, Congress placed primary importance on insulation of youth offenders from the corrupting influence of older offenders and on rehabilitation of youth offenders. Indeed, Congress specifically authorized and directed the Bureau to deal with the problem of mixing of various types of youth offenders by designating a variety of institutions for their treatment and by segregating them according to their needs for treatment. The Bureau of Prisons contends that the experiments which it has carried out demonstrate that the mixing of older and younger offenders may serve to reduce violence, but this is insufficient to justify disregard of a Congressional mandate. It is, of course, desirable to reduce violence but Congress is going to have to be convinced that this is a product of integration of hardened criminals with youth offenders. We do not deny the desirability of reducing violence. An institution cannot be successful unless this primary objective is achieved. The question is whether this cannot be accomplished through other means. The United States should not have to rely on inmates for maintaining discipline. Be that as it may, the Bureau is not free to reject the congressional plan. Acts of Congress cannot be avoided by administrative fiat. The majority of the courts which have dealt with this issue reached results similar to ours. United States ex rel. Dancy v. Arnold, supra; Brown v. Carlson, supra; Johnson v. Bell, supra; Watts v. Hadden, 469 F.Supp. 223 (D.Colo.1979). See also, Robinson v. Ralston, 642 F.2d 1077 (7th Cir. 1981); United States v. Howard, 449 F.2d 1086 (D.C.Cir.1971); Micklus v. Carlson, 632 F.2d 227 (3rd Cir. 1980). It is true that a few courts have reached apparently contrary results without discussion of the question of segregation. Abernathy v. United States, 418 F.2d 288 (5th Cir. 1969) (YCA offender may be placed in a maximum security institution); Foote v. United States, 306 F.Supp. 627 (D.Nev.1969) (YCA offenders may be placed in institutions with “hardened criminals”). Only one court has expressly addressed the segregation requirements of the YCA and concluded that those requirements need not be implemented by the Bureau of Prisons. Outing v. Bell, 632 F.2d 1144 (4th Cir. 1980). There, the court held that segregation of youth offenders is not mandatory under the YCA, and that a YCA sentenced prisoner must prove that segregation is practical before any relief is appropriate. A strong dissent was written in Outing by Circuit Judge K. K. Hall; needless to say we agree with the views expressed by Judge Hall. We have considered the rationale of the majority in Outing and we are not persuaded. For the reasons that are set forth above, it is our conclusion that the judgment of the trial court should be and the same is affirmed. We remand the case for further proceedings in accordance with this opinion. PART II This class action involves the second of two consolidated actions against the Bureau of Prisons and the United States Parole Commission. The two claims were filed as petitions for writs of habeas corpus pursuant to 28 U.S.C. § 2241, et seq., by inmates of the Federal Correctional Institute at Englewood, Colorado. These inmates were sentenced under the Youth Corrections Act, 18 U.S.C, § 5005, et seq. The claim in this case is that the petitioners are being held unlawfully because the Parole Commission refused to follow the requirements of the Youth Corrections Act. The trial court entered a Partial Final Judgment under Fed.R.Civ.P. Rule 54(b) against the Bureau of Prisons. The appeal of that final judgment is dealt with in our previous opinion. This part deals only with the claim against the United States Parole Commission. The trial court ordered the Parole Commission to file a written statement with the court setting forth the release date which had been determined for each of the original petitioners in this action; an explanation of the procedures by which that release date was determined; a statement of the manner in which the parole guidelines were applied when reaching that decision; a statement setting forth what, if any, information not contained in the pre-sentence report or in communications from the sentencing judge was used in making that determination; a statement of what re-examination was to be made of each of the original petitioners and when such re-examination was to be made; and what recommendations were received from the Director of Prisons with regard to each of the original petitioners. The Parole Commission’s response to the order was submitted on October 10, 1979. On January 30,1980 the Parole Commission moved to dismiss the petitions of those of the original petitioners who had been released from confinement as being moot. Therefore, on May 20, 1980 the trial court ordered that this proceeding be conducted as a class action due to the possibility that mootness might permit these issues to escape review. Following certification of the class, the trial court entered an order which directed the Parole Commission to take action within 90 days to correct its arbitrary and capricious action in failing to comply with the mandatory provisions of the YCA. Final judgment to that effect was entered by the trial court against the Parole Commission on August 15, 1980. In its decree the trial court ordered the Parole Commission to remedy the following problems: 1. The failure to obtain and use for each committed youth offender a complete classification study, including a mental and physical examination of the type explicitly required by 18 U.S.C. § 5014, and as described in the plan for implementation submitted by the Bureau of Prisons on August 31, 1979. 2. The failure to consider the inmate’s response to treatment programs and the use of guidelines which consider only the relative severity of past criminal conduct in determining the time for conditional release. 3. The failure to obtain and use periodic examinations and reports as required by 18 U.S.C. § 5016. The present practice of conducting an initial hearing with approximately three months of confinement and then deferring further consideration until the statutory time of 18 months or 24 months as required for adult offenders under 18 U.S.C. § 4208(h), is inadequate and not in compliance with the requirements of the YCA. While the YCA does not contain specific time limitations, provision for periodic reports in the Bureau of Prisons plan submitted on August 31, 1979 is sufficient and a review of the inmate’s institutional progress on a corresponding schedule would meet the statutory requirements imposed on the Parole Commission as well. 4. The failure to consider the unconditional discharge of a youth offender at the expiration of one year from the date of conditional release as authorized by 18 U.S.C. § 5017(b). The commissioner’s requirement of two years of “clean supervision” after conditional release is an improper application of the provisions of 18 U.S.C. § 4211(b). That section applies only to adult sentences. It is inconsistent with the YCA. See Bellizi v. Kaslow, 484 F.Supp. 868 (D.Colo.1980). The Parole Commission has appealed the final judgment and order on two grounds: First, the Commission questions whether it is required under the YCA to consider the youth offenders response to treatment when determining his eligibility for parole. Second, the Commission questions whether it must consider the youth offender for unconditional release at the expiration of one year of parole under 18 U.S.C. § 5017(b). The contention of the appellees, plaintiffs below, are: 1. That the Parole Commission’s application of release guidelines to Youth Act sentences has the effect of converting judicially imposed indeterminate sentences into fixed and determinate terms of confinement, which practice invades the sentencing province of the courts. 2. The Parole Commission has abused its discretion in failing to conduct periodic examinations and consider unconditional release at the expiration of one year from the date of conditional release. So, then, the issues to be determined are first, whether the Parole Commission is in violation of the Youth Corrections Act by reason of its interpretation of that Act in effect modifying and determining its meaning so as to revise its provisions and objectives. Second, has the Parole Commission acted invalidly in failing to conduct periodic examinations and in failing to consider unconditional release at the expiration of one year from the date of conditional release. I. Background The legislative history of the Youth Corrections Act states that the purpose of the Act is to promote the rehabilitation of those youths who the sentencing judge believes show promise of becoming useful citizens. H.R.Rep.No.2979, 81st Cong., 2nd Sess., reprinted in [1950] U.S.Code Cong.Service, 3983. To that end, the Act was designed to make available, for the discretionary use of federal judges, an alternate sentencing and treatment procedure for those who, in the court’s opinion, are qualified youth offenders. Id. The Act focuses on special, segregated treatment and supervision of youth offenders. The objective of the Youth Corrections Act was to substitute rehabilitative principles for retributive methods of treating antisocial behavior. Id., p. 3985. The impetus for the Act as originally adopted was a study of the English Borstal system of rehabilitating youth offenders; the legislative history of the Act includes a thorough three-page discussion of that system and an analysis of how and why the system works. Id., pp. 3987-89. The view of Congress was that the special parole procedures within the Borstal system were crucial to the success of that system. Id., pp. 3988-89. Originally, therefore, a Youth Corrections Division was created under the United States Board of Parole to carry out the purpose of the Act. 18 U.S.C. § 5005 (1950). In 1976, however, the nine member United States Parole Commission was established as an independent agency within the Department of Justice, which replaced the United States Board of Parole. 18 U.S.C. §§ 4201-4212 (1976). (The Parole Commission and Reorganization Act.) The Parole Commission, thus, replaced the Youth Corrections Division, and assumed the latter’s responsibility for carrying out the purposes of the YCA. At this time § 5017 of the Youth Corrections Act was changed as follows: Prior to the 1976 amendment § 5017(a) stated that: The Division may at any time after reasonable notice release conditionally under supervision a committed youth offender. When, in the judgment of the Director, a committed youth offender should be released conditionally under supervision, he shall so report and recommend to the Division. In 1976, § 5017(a) was amended to read as follows: The Commission may at any time after reasonable notice to the Director release conditionally under supervision a committed youth offender in accordance with the provisions of § 4206 of this Title. When in the judgment of the Director, a committed youth offender should be released conditionally under supervision, he shall so report and recommend to the Commission. Section 4206 of 18 U.S.C., referred to above, lists the criteria that are to be considered by the Parole Commission when determining when to release a prisoner on parole. That section provides as follows: (a) If an eligible prisoner has substantially observed the rules of the institution or institutions to which he has been confined, and if the Commission, upon consideration of the nature and circumstances of the offense and the history and characteristics of the prisoner, determines: (1) that release would not depreciate the seriousness of his offense or promote disrespect for the law; and (2) that release would not jeopardize the public welfare; subject to the provisions of subsections (b) and (c) of this section, and pursuant to guidelines promulgated by the Commission pursuant to section 4203(a)(1), such prisoner shall be released. (b) The Commission shall furnish the eligible prisoner with a written notice of its determination not later than twenty-one days, excluding holidays, after the date of the parole determination proceeding. If parole is denied such notice shall state with particularity the reasons for such denial. (c) The Commission may grant or deny release on parole notwithstanding the guidelines referred to in subsection (a) of this section if it determines there is good cause for so doing: PROVIDED, That the prisoner is furnished written notice stating with particularity the reasons for its determination, including a summary of the information relied upon. (d) Any prisoner, serving a sentence of five years or longer, who is not earlier released under this section or any other applicable provision of law, shall be released on parole after having served two-thirds of each consecutive term or terms, or after serving thirty years of each consecutive term or terms of more than forty-five years including any life term, whichever is earlier: PROVIDED, HOWEVER, That the Commission shall not release such prisoner if it determines that he has seriously or frequently violated institution rules and regulations or that there is a reasonable probability that he will commit any federal, state, or local crime. Added Pub.L. 94-233, § 2, Mar. 15, 1976, 90 Stat. 223. II. Under the 1976 amendments to 18 U.S.C., the Commission has the power to grant or deny an application or recommendation to parole any eligible prisoner. § 4203(b). A prisoner is eligible for release on parole after serving one-third of his term “except to the extent otherwise provided by law.” § 4205(a). The Parole Commission is required under § 4203(a)(1) to promulgate rules and regulations establishing guidelines for the exercise of its enumerated powers. It has gone forward and has promulgated such guidelines for national parole policy. See 28 CFR § 2.1, et seq. It has issued instructions for the use of those guidelines in its Guideline Application Manual. Pursuant to those guidelines and instruction manual, the Parole Commission considers the following factors when determining whether, and when, to parole a youth offender: (1) The Offense Severity Rating: Each prisoner is assigned a rating based on the severity of the offense for which he was convicted. Part III of the Guideline Application Manual states that the offense severity rating is to reflect the overall circumstances of the present offense behavior. The offense rating appropriate to the actual offense behavior that occurred is to be selected. If the actual offense behavior was more severe than the offense of conviction, the severity rating must be explained by a brief summary of the specific facts that justify the ratings. As stated in the Manual: Example: Where the offense of conviction is bank larceny, and the actual offense behavior is bank robbery, the Notice of Action would read: ‘Your offense has been rated as very High severity because you in fact used a threatening note to rob a savings and loan association.’ Any information in the file which is persuasive, in that such information comes from a reliable source and is corroborated by established facts, is to be used when determining the actual offense behavior/severity. Expressly cited as an example of such information is information contained in a count of an indictment that was dismissed as a result of a plea agreement. An offense behavior which is not listed on the guidelines is to be rated by comparison with those offenses which are listed. Further, the manual provides that if an offense behavior can be classified in more than one severity category, the more serious category is to be used. (2) The Salient Factor Score: There is a salient factor score which is given to each prisoner based on considerations of seven elements. The higher the salient factor score, the earlier the inmate will be entitled to parole. Those factors which are individually scored to comprise the total salient factor score include: (a) Previous instances of criminal conduct (including juvenile or youth adjudications); (b) Prior incarcerations (including commitment to juvenile institutions and residential treatment centers); (c) Age at first commitment (under this factor younger offenders are penalized and given a less positive parole prognosis. An inmate 26 years of age or older at the time of his first commitment receives two points; an inmate between 18 and 25 years of age at the time of his first commitment receives one point; an inmate 17 years of age or younger at the time of his first commitment receives zero points); (d) Whether the offense involved auto theft; (e) The inmate’s prior or present parole history; (f) Whether the inmate has a history of heroin or opiate dependence; and (g) Whether the inmate was employed or a full time student for a total of at least six months during his last two years in the community. The two scores, offense severity and salient factor, are the same ingredients used by the Parole Commission to'evaluate the parole prognosis of adult offenders, and are considered together when determining when a youth offender is entitled to be paroled. Guidelines are followed which state the length of time the inmate must serve in custody prior to parole in accordance with the inmate’s particular scores. In about ten percent of all cases,, however, mitigating circumstances may cause the Commission to require the inmate to spend less time in custody before he is entitled to parole. Mitigating factors which may be considered include diminished mental capacity; duress; evidence of the prisoners attempted withdrawal prior to completion of the offense; a belief by the prisoner that he had a claim of right to the property (where property offenses are involved); a d