with whom MANTON and KANNE, Circuit Judges, join, concurring.
The question before us today was resolved by the Supreme Court in 1972. Morrissey v. Brewer, 408 U.S. 471, 487, 92 S.Ct. 2593, 2603, 33 L.Ed.2d 484 (1972), describes the parolee’s constitutional rights in the revocation process, saying in part: “[A] determination [of probable cause at a preliminary hearing] would be sufficient to warrant the parolee’s continued detention and return to the state correctional institution pending the final decision.” It is unnecessary for us to balance interests under the Due Process Clause, and unwarranted to invite the district judge to take more evidence and strike his own balance. The Supreme Court has done it already.
Morrissey codifies a set of constitutional rules.1 As with Miranda v. Arizona, 384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694 (1966), Wolff v. McDonnell, 418 U.S. 539, 94 S.Ct. 2963, 41 L.Ed.2d 935 (1974), and other comprehensive decisions, almost all of the opinion could be labeled dicta. The details of Miranda, for example, could be disregarded on the ground that Ernesto Miranda had not been given any warning, so the Court could not pronounce on the consequences of giving three but not four of the warnings on its list. The Court has rebuffed arguments of this sort, however. See, e.g., Fare v. Michael C., 442 U.S. 707, 718, 99 S.Ct. 2560, 2568, 61 L.Ed.2d 197 (1979) (reversing a court that had altered the scope of Miranda, remarking that its mechanical nature is its chief virtue); Baxter v. Palmigiano, 425 U.S. 308, 96 S.Ct. 1551, 47 L.Ed.2d 810 (1976) (reversing an effort to alter one of the guidelines announced in Wolff). See also United States v. Jackson, 836 F.2d 324, 326-27 (7th Cir.1987). Cf. United States v. Underwood, 717 F.2d 482, 484-86 (9th Cir.1983) (en banc) (considered dictum in a codifying decision binds the lower courts).
The Supreme Court must deny inferior courts the power to change isolated parts of codifying decisions, because the rules interact. One of the four warnings in Miranda is thought sufficient only because it has company; a given step in prison discipline governed by Wolff is deemed sufficient (and other steps omitted) only because the combination is thought appropriate. So too with revocation of parole. In deciding how thorough the preliminary revocation hearing had to be, the Court necessarily considered its function: to permit the detention of the parolee until the final revocation hearing. When the Court says that the preliminary hearing entails steps A, B, and C, after which the parolee may be held, it is not offering a gratuitous view of whether he may be held; the Court has defined the steps necessary to hold the parolee. Were there to be a bail decision, the preliminary hearing might have been confined to steps A and B, or might have been omitted. What is the preliminary hearing for, except to inquire whether there is an adequate foundation to detain pending the final hearing?
Morrissey devotes three pages to the details of preliminary hearings. 408 U.S. at 485-87, 92 S.Ct. at 2602-03. The procedures the Court has prescribed — a prompt hearing before a neutral decisionmaker, at which the parolee may testify and present documentary evidence; if the parolee so requests, persons giving adverse evidence must be produced for questioning; the de-cisionmaker must summarize the evidence and give reasons for finding probable cause — are more substantial than those ac*731companying a charging decision. They are procedures appropriate to a decision to reimprison one who is already under sentence.
Since Morrissey decides the issue before us, the court’s survey of arguments pro and con is unnecessary, and its invitation to the district court to conduct still a further balancing is unwarranted. It supposes that our view matters, that judges of the inferior federal courts may decide this case the other way if they disagree with Morris-sey. Wise or not (as we mark wisdom), the Supreme Court’s discussion is authoritative.2 It precludes consideration of a “release suitability hearing” — bail by any other name, and as sweet to the parolee. The discussion of bail under the eighth amendment also is unnecessary. The Bail Clause of the eighth amendment applies only to detention before final judgment. The parolee waiting for the final revocation hearing is in custody under a sentence supported by a conviction. The state need not have “compelling” reasons to require a felon to serve additional time on a valid sentence. I therefore join only Part VI of the court’s opinion.
Only a strong hint that the Court has changed its mind since Morrissey would support a fresh look at the subject. The Court has not dropped one; any hints nudge us in the opposite direction. Several recent cases show that parole is a limited entitlement rooted in statute rather than the natural liberty of the parolee. E.g., Board of Pardons v. Allen, — U.S. —, 107 S.Ct. 2415, 96 L.Ed.2d 303 (1987) (a prisoner has a liberty interest in parole only to the extent a statute creates one); Jago v. Van Curen, 454 U.S. 14, 102 S.Ct. 31, 70 L.Ed.2d 13 (1981) (rescission of parole does not require any process); Greenholtz v. Inmates, 442 U.S. 1, 99 S.Ct. 2100, 60 L.Ed.2d 668 (1979) (no entitlements other than those provided by statute). A grant of parole or probation supplies the parolee with but a portion of the liberty enjoyed by those who have served their full sentences. A probationer’s home may be searched under circumstances that free persons need not tolerate, see Griffin v. Wisconsin, — U.S. —, 107 S.Ct. 3164, 97 L.Ed.2d 709 (1987). The parolee must accept conditions that continue control over much of his life, including where he may live and with whom he may associate. E.g., United States v. Tonry, 605 F.2d 144 (5th Cir.1979). Then there is the ultimate difference — parole may be revoked by an administrative tribunal on the basis of a preponderance of the evidence showing a violation of terms and conditions (not necessarily amounting to crimes), while only proof in court beyond a reasonable doubt of a new crime permits the state to imprison a person in the first instance. See also Gagnon v. Scarpelli, 411 U.S. 778, 93 S.Ct. 1756, 36 L.Ed.2d 656 (1973) (probationers and parolees do not automatically have a right to counsel in the revocation hearing). None of this suggests that the Court now treats parole as equivalent to the liberty *732that citizens of the United States ordinarily possess. It follows that a lesser showing— the sort required at a preliminary revocation hearing — permits the state to hold the parolee in custody pending a final revocation hearing, even though no such showing would suffice under the Bail Clause and the Due Process Clause to hold a person charged with crime. Cf. United States v. Salerno, — U.S. —, 107 S.Ct. 2095, 95 L.Ed.2d 697 (1987).
One feature of Faheem-EI’s case did not receive attention in Morrissey. The Supreme Court assumed that the final revocation hearing would follow promptly after the preliminary hearing, remarking (408 U.S. at 488, 92 S.Ct. at 2604): “A lapse of two months, as respondents suggest occurs in some cases, would not appear to be unreasonable.” Faheem-El waited a lot longer than that, and evidence in the record suggests that other parolees wait longer too. The district court, having established a per se rule that parolees are entitled to prompt consideration for release on bail, did not decide whether Illinois routinely holds parolees between stages for more than two months. I assume that this topic is open on remand as the district court considers the request for a permanent injunction. To the extent the court believes that any other topic should be open, it has authorized a district judge to second-guess the Supreme Court. The majority's thoughtful assessment of “release suitability hearings” may well persuade the State of Illinois to improve the operation of its parole consideration procedure, but whether to do so ultimately is a question of policy for the state rather than constitutional law for the court.
. A "codifying decision" as I employ that term is one that describes constitutional minima in some detail and addresses the subject comprehensively even though the case could have been disposed of by a narrowly-drawn opinion. Mor-rissey is a codifying decision because it specifies at least 15 constitutional requirements for the revocation of parole, starting with the arrest and preliminary hearing and ending with the final revocation hearing. This is the sense of codification to which Judge Friendly referred in The Bill of Rights as a Code of Criminal Procedure, in Benchmarks 235 (1967). Morrissey distinguishes a different sense of codification — that of furnishing the many other essential procedures that the Constitution does not govern— when remarking, 408 U.S. at 488, 92 S.Ct. at 2604, that it "cannot write a code of procedure; that is the responsibility of each state." Our case deals with constitutional minima, not the many other steps over which states retain control.
. My colleagues claim the right to disregard Morrissey on the ground that the Illinois statute is "significantly different” from the one presented there. At 724 n. 16. It is indeed. It is different because it offers parolees "significantly” more procedural protection than the one the Court considered in Morrissey, and so far as I can tell, significantly more than what Morrissey itself requires. Morrissey permits a "parole officer other than the one who has made the report of parole violations" (408 U.S. at 486, 92 S.Ct. at 2603) to hold the hearing; Illinois provides an independent hearing officer. 20 Ill.Adm.Code § 1610.140(a). Morrissey requires notice of "what parole violations have been alleged” (408 U.S. at 487, 92 S.Ct. at 2603); Illinois adds that the notice must tell the parolee "the manner in which [the rules] were violated.” 20 Ill.Adm. Code § 1610.140(a). Morrissey requires the state to let the parolee appear, speak, and bring documents and witnesses (408 U.S. at 487, 92 S.Ct. at 2603); Illinois gives the parolee compulsory process to produce still more evidence when “the relevance of testimony of the witness is substantial.” 20 Ill.Adm.Code § 1610.140(f). Morrissey does not hold or imply that the parolee is entitled to counsel at the preliminary hearing; Illinois allows a parolee to employ retained counsel at the preliminary hearing. 20 Ill.Adm. Code § 1610.140(e). Why should Illinois’ decision to go the extra mile be the basis of a constitutional requirement that still more is required? The only significant difference my colleagues identify is that in Illinois a parolee is automatically detained after the preliminary hearing. But since detention after the preliminary hearing is something the Supreme Court has already approved, this is not the sort of "difference” we may use as the fulcrum for additional procedures.