INMATES OF SUFFOLK COUNTY JAIL, etc., et al., Plaintiffs, Appellees, v. Richard J. ROUSE, etc., et al., Defendants, Appellants; INMATES OF SUFFOLK COUNTY JAIL, etc., et al., Plaintiffs, Appellants, v. Richard J. ROUSE, etc., et al., Defendants, Appellees; INMATES OF SUFFOLK COUNTY JAIL, etc., et al., Plaintiffs, Appellees, v. Richard J. ROUSE, etc., et al., Defendants, Appellees, United States of America, Intervenor, Appellant, 129 F.3d 649

SELYA, Circuit Judge. The passage of the Prison Litigation Reform Act, 18 U.S.C.A. § 3626 (Supp.1997) (the PLRA or the Act), brought cheers to the lips of many prison administrators. In its wake, the Sheriff of Suffolk County and the Massachusetts Commissioner of Correction (collectively, the defendants) cast their gaze toward a consent decree that has governed important aspects of the county’s handling of pretrial detainees since 1979.

Spying an opportunity to sever the shackles of judicial oversight, the defendants invoked the new law and asked the supervising tribunal, the United States District Court for the District of Massachusetts, to vacate the decree or, in the alternative, to terminate all prospective relief under it. The plaintiffs questioned the Act’s constitutionality and raised a host of other objections to the defendants’ motions. The district court repulsed the constitutional attack but construed the PLRA to require only the termination of prospective relief, not the vacatur of the consent decree itself.

See Inmates of Suffolk County Jail v. Sheriff of Suffolk County, 952 F.Supp. 869 (D.Mass.1997) (D.Ct.Op.). After careful consideration of the meaning of the PLRA, we vouchsafe the Act’s constitutionality against the challenges asserted here and construe it to entitle correctional officials to the termination of existing consent decrees in civil actions involving prison conditions (except in the presence of statutorily prescribed conditions that forestall such termination). I. BACKGROUND This litigation deals almost exclusively with the effect of the PLRA on an extant consent decree.

Thus, the history of the conflict is of minimal import, and we merely sketch it. The shelves of any reasonably well-stocked law library afford readers who hunger for more exegetic detail ample opportunity to dine elsewhere. See, e.g., Inmates of Suffolk County Jail v. Eisenstadt, 360 F.Supp. 676, 679-84 (D.Mass.1973), aff'd, 494 F.2d 1196 (1st Cir.1974); Inmates of Suffolk County Jail v. Kearney, 734 F.Supp. 561, 562-63 (D.Mass.), aff'd, 915 F.2d 1557 (1st Cir.1990) (table), vacated, 502 U.S. 367, 112 S.Ct. 748, 116 L.Ed.2d 867 (1992); Inmates of Suffolk County Jail v. Kearney, 928 F.2d 33, 34 (1st Cir.1991); D.

Ct. Op., 952 F.Supp. at 871-73. In 1971 the plaintiff class, which consists of present and future pretrial detainees held or to be held in the Suffolk County jail (collectively, the plaintiffs), brought a civil action alleging that the conditions of . their confinement — particularly double bunking— violated the Eighth Amendment to the United States Constitution.

After extensive skirmishing, not relevant here, the parties reached a rapprochement, subsequently apr proved by the district court and embodied in the 1979 consent decree, in which they ratified an architectural plan for a new facility featuring single-occupancy cells. The agreement contemplated the phasing-out of the existing Charles Street jail once the new structure was in place. As the Scottish poet warned, “the best laid schemes o’ mice and men gang aft a-gley,” Robert Burns, To a Mouse (1785), and in this case time proved a formidable opponent.

Growth in prison population and delays in construction. both exceeded expectations. The new facility (the Nashua Street jail) was not completed until mid-1990 and was hard-pressed from the start to cope with the Sheriffs escalating needs. In response to these volatile conditions, the consent decree was modified by court order in 1985, 1990, and 1994. The last of these changes permitted limited double bunking at the Nashua Street facility (the Sheriff having closed the Charles Street facility prior thereto). In July 1996 the Sheriff initiated the current engagement.

He grasped the weapon that Congress had forged and moved to terminate all prospective relief pursuant to the PLRA. Not to be outdone, the Commissioner moved to vacate the consent decree outright, thus formalizing a suggestion that the Sheriff had omitted from his motion but had included in the memorandum supporting the motion: When the plaintiffs indicated that they would challenge the Act’s constitutionality as part of their opposition, the federal government intervened'.

After sorting out the components of the parties’ extensive as-severational array, Judge Keeton gave the pertinent provisions of the PLRA a narrowing construction and on that basis upheld their constitutionality. He thereupon granted the Sheriffs motion to the extent that the consent decree would “no longer be enforced by an order of specific performance,” but declined either to vacate the decree or to “terminate the obligations stated [therein]” because those obligations represented “consensual undertakings of the defendants with court approval.” Id. at 883. All parties appealed.

In an effort to cut a passable swath through this legal thicket, we start by construing the termination provision of the PLRA. We then test its constitutionality and, finding no merit in the plaintiffs’ constitutional challenges, apply the Act and evaluate the extent of the remediation to which the defendants are entitled. II. THE PLRA In parsing the PLRA, we afford de novo review. See United States v. Gifford, 17 F.3d 462, 471-72 (1st Cir.1994). Such an exercise in statutory interpretatipn always begins with the language of the statute itself. See Stowell v. Ives, 976 F.2d 65, 69 (1st Cir.1992).

At this stage, an inquisitive court should assume that the words of the statute, if not specially defined, comport with then-ordinary meaning, and that the words, so read, accurately express the legislature’s intent. See FMC Corp. v. Holliday, 498 U.S. 52, 57, 111 S.Ct. 403, 407, 112 L.Ed.2d 356 (1990). In keeping with this principle, the court should, “resort to legislative history and other aids of statutory construction only when the literal words of the statute create ambiguity or lead to an unreasonable result.” United States v.

Charles George Trucking Co., 823 F.2d 685, 688 (1st Cir.1987) (citation and internal quotation marks omitted). The PLRA is not a paragon of clarity.

In regard to existing federal court - orders, it declares that “in any civil action with respect to prison conditions, a defendant or interve-nor shall be entitled to the immediate termination of any prospective relief if the relief was approved or granted in the absence of a finding by the court that the relief is narrowly drawn, extends no further than necessary to correct the violation of the Federal right, and is the least intrusive means necessary to correct the violation of the Federal right.” 18 U.S.C.A. § 3626(b)(2).

Such prospective relief shall not terminate, however, “if the court makes written findings based on the record that prospective relief remains necessary to correct a current or ongoing violation of the Federal right, extends no further than necessary to correct the violation of the Federal right, and that the prospective relief is narrowly drawn and the least intrusive means to correct the violation.” Id. § 3626(b)(3). With regard to relief not yet obtained, the Act contains similar proscriptions.

It forbids courts from granting or approving prospective relief “unless the court finds that such relief is narrowly drawn, extends no further than necessary to correct the violation of the Federal right, and is the least intrusive means necessary to correct the violation of the Federal right.” Id. § 3626(a)(1)(A). These iterations are clear enough, but uncertainty arises when we examine the Act’s definitional instructions.

One such passage defines “prospective relief’ to include “all relief other than compensatory monetary damages,” and then defines “relief’ to “mean[ ] all relief in any form that may be granted or approved by the court ... including] consent decrees.” Id. § 3626(g)(7), (9). “Consent decree,” in turn, means “any relief entered by the court that is based in whole or in part upon the consent or acquiescence of the parties but does not include private settlements.” Id. § 3626(b)(1). In a vacuum, the interaction of the Act’s mechanics and these

definitions is manageable: terminating “prospective relief’ as prescribed by

section 3626(b)(2) would terminate “all relief,” see

section 3626(g)(7), which under

section 3626(g)(9) “includes consent decrees.” Read literally, therefore, once defendants or intervenors show their entitlement to terminate prospective relief, the Act seemingly requires termination of the consent decree itself. As the district court astutely observed, this result is counterintuitive in that it contradicts the usual understanding of both “relief’ and “consent decree.” See D. Ct. Op., 952 F.Supp. at 878.

The customary definition of “consent decree” likens such decrees to judgments, see Black’s Law Dictionary 410 (6th ed.1990) (defining “consent decree” as “[a] judgment entered by consent of the parties whereby the defendant agrees to stop alleged illegal activity without admitting guilt or wrongdoing”), and in ordinary usage a judgment is “[a] final decision of the court resolving the dispute and determining the rights and obligations of the parties,” id. at 841-42. “Relief,” on the other hand, typically is equated with “remedy,” id. at 1292, which is “the means by which a right is enforced or the violation of a right is prevented, redressed, or compensated,” id. at 1294.

Inasmuch as a remedy effectuates. the adjudication expressed in a judgment, one ordinarily would assume that “relief,” by extension, effectuates the legal decision, arrived at by consent, in a “consent decree.” Congress conflated the two terms when it described consent decrees as a form, of relief rather than as a judgment that engenders relief. The PLRA’s equation of “consent decree” and “relief’ contradicts conventional understandings and creates a situation in which a strict, language-based construction of the PLRA requires that commonplace legal terms be used in curious ways.

This circumstance fosters uncertainty, for a court cannot really tell, without further inquiry, whether the linguistic anomaly is accidental or purposeful (and, thus, whether Congress meant to uproot consent decrees themselves or merely to vitiate the relief attendant to them, when it directed federal courts to facilitate “the immediate termination of any prospective relief’ at the behest of prison litigation defendants and intervenors). This uncertainty impels us to consult extrinsic sources in search of guidance as to Congress’s intent.

In this instance, the PLRA’s legislative history persuades us to embrace the unusual. Congress passed the PLRA in an effort, in part, to oust the federal judiciary from day-to-day prison management. See 141 Cong. Rec. 14,419 (1995) (statement of Sen. Abraham) (“[N]o longer will prison administration be turned over to Federal judges for the indefinite future for the slightest reason.”); id. at 14,418 (statement of Sen. Hatch) (“I believe that the courts have gone too far in micromanaging our Nation’s prisons.”).

This evidence of ambient intent inclines us to interpret the statute literally (i.e., as directing courts to terminate consent decrees outright), for it strongly suggests that the PLRA’s sponsors wanted to truncate the federal judiciary’s involvement in prison administration. The House Conference Report provides even more powerful direction on this score.

The Report describes the “explanation of the effect of the action agreed upon by the [legislation’s] managers” and states that, by virtue of the PLRA, “[p]rior consent decrees are made terminable upon the motion of either party, and can be continued only if the court finds that the imposed relief is necessary to correct the violation of the federal right.” H.R. Conf. Rep. No. 104-378 at 166 (1995). This plain language leaves little room for doubt that Congress intended the PLRA as a last rite for those consent deerees that are incapable of surviving the rigors of

section 3626(b)(2). Of course, we recognize that the plain meaning rule, while a bedrock principle of statutory construction, may yield if giving effect to literal meaning would produce a bizarre result. See Sullivan v. CIA, 992 F.2d 1249, 1252 (1st Cir.1993); Charles George Trucking, 823 F.2d at 688. But this exception is sparingly employed, and the circumstances of this ease give it no purchase. The result that Congress’s plain language portends here involves a somewhat unusual use of terms, but it is not unreasonable. We will not paint the lily.

Given the congruence between the text of the. statute and the legislature’s easily discerned intent, we conclude that Congress meant precisely what it said — however deviant from ordinary usage that may be — when it wrote the PLRA and specially defined its operative terms. We are therefore duty bound to interpret the PLRA as mandating the termination of extant consent decrees altogether unless the district court makes the specific findings that are necessary to keep a particular decree alive. III.

THE CONSTITUTIONALITY OF THE PLRA Having construed the PLRA, we next must essay a de novo determination of whether it passes constitutional muster. The plaintiffs say that the Act’s termination provision violates the Constitution three, times over by transgressing (1) the separation of powers principle, (2) the Due Process Clause, and (3) the Equal Protection Clause. Though ably presented, none of these assertions carries the day. A. Separation of Powers. Few tenets are more central to the genius of our constitutional system than the separation of powers principle. See O’Donoghue v.

United States, 289 U.S. 516, 530, 53 S.Ct. 740, 743, 77 L.Ed. 1356 (1933) (describing separation of powers as “basic and vital” to our scheme of government). This principle has many incarnations. In one such configuration, it insulates the judiciary from unwarranted legislative intrusions. The courts’ historic independence has its roots in the Constitution, which explicitly provides that “[t]he judicial Power of the United States shall be vested in one Supreme Court, and in such inferior courts as the Congress may from time to time ordain and establish.” U.S. Const, art. Ill, § 1.

This delegation of power serves “to safeguard litigants’ right to have claims decided before judges who are free from potential domination by other branches of government.” Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 848, 106 S.Ct. 3245, 3255, 92 L.Ed.2d 675 (1986) (citation and internal quotation marks omitted). The due administration of justice demands that this separation remain inviolate. The plaintiffs lament that the PLRA infringes upon the courts’ guaranteed separateness in two distinct ways. 1. Reopening Final Judgments.

The separation of powers principle forbids Congress from reopening the final judgments of

Article III courts. See Plaut v. Spendthrift Farm, Inc., 514 U.S. 211, 240, 115 S.Ct. 1447, 1463, 131 L.Ed.2d 328 (1995). After all, if the judiciary’s power to render definitive judgments were subject to congressional control, then the judiciary would become, within its own sphere, subordinate to the legislature. Moving from the general to the particular, the plaintiffs maintain that the PLRA offends this principle by requiring a district court to rescind relief that the court already has seen fit to award.

In mounting this argument, the plaintiffs rely heavily on the Justices’ observation, made in an earlier round of this litigation, that “a consent decree is a final judgment that may be reopened only to the extent that equity requires.” Rufo, 502 U.S. at 391, 112 S.Ct. at 764. From this thread, the plaintiffs weave a syllogism: Congress cannot order the reopening of final judgments without offending the separation of powers principle, a consent decree is a final judgment, and therefore Congress cannot mandate the reopening of consent decrees.

This reasoning frays because consent decrees of the type at issue here are not “final judgments” for the purpose of a separation of powers analysis. In a recent articulation of the rule that the legislature cannot interfere with final judgments of

Article III courts, the Supreme Court carefully carved out an exception and endorsed a line of cases sanctioning legislation “that altered the prospective effect of injunctions entered by

Article III courts.” Plaut, 514 U.S. at 232, 115 S.Ct. at 1459. This exception did not spring full-blown from Justice Scalia’s brow. To the contrary, its roots burrow deep into our constitutional soil. An early exemplar is Pennsylvania v. Wheeling & Belmont Bridge Co., 59 U.S. (18 How.) 421, 15 L.Ed. 435 (1855). That hoary case established that, although a judgment at law is impervious to legislative assault, a forward-looking judgment in equity can succumb to legislative action if the legislature alters the underlying rule of law. See id. 59 U.S. (18 How.) at 431-32. More recent examples also exist.

See, e.g., Landgraf v. USI Film Prods., 511 U.S. 244, 273, 114 S.Ct. 1483, 1501, 128 L.Ed.2d 229 (1994); System Fed’n No. 91 v. Wright, 364 U.S. 642, 651-52, 81 S.Ct. 368, 373-74, 5 L.Ed.2d 349 (1961). Lower courts sometimes are required to follow precedent for precedent’s sake, no matter how much the yoke chafes. Here, however, our burden is light, for the Wheeling Bridge exception is not only mandated by precedent but also makes logical sense. The legitimacy of prospective equitable relief rests upon the presumed persistence of the conditions that originally justified the relief.

If forward-looking judgments in equity were inviolate, then one of two scenarios would develop: either the legislature would be stripped of the ability to change substantive law once an injunction had been issued pursuant to that law, or an issued injunction would continue to have force after the law that originally gave the injunction legitimacy had been found wanting (and, hence, altered). The first of these possible results would work an undue judicial interference with the legislative process, while the second would create an intolerable tangle in which some laws applied to some persons and not to others.

Since the separation of powers principle is a two-way street, courts must be careful not to embrace a legal regime that promotes such awkward scenarios. To recapitulate, consent decrees are final judgments, but they are final judgments subject to revision “to the extent that equity requires.” Rufo, 502 U.S. at 391, 112 S.Ct. at 764. Plaut and Wheeling Bridge, read together, teach that equity requires, and the separation of powers principle permits, legislatures to direct that courts respond to changes in substantive law by revisiting forward-looking injunctions. See Plyler v.

Moore, 100 F.3d 365, 371 (4th Cir.1996). The Court stated the point with great clarity earlier in this litigation: “A consent decree must of course be modified if, as it later turns out, one or more of the obligations [it imposes] has become impermissible under federal law.” Rufo, 502 U.S. at 388, 112 S.Ct. at 762. The plaintiffs try to turn these verities to their advantage by asserting that the underlying law here — the Eighth Amendment— has not changed. This is résupinate reasoning.

The relevant underlying law in this ease is not the Eighth Amendment, as there has been no finding of an ongoing constitutional violation. Rather, the relevant underlying law relates to the district court’s authority to issue and maintain prospective relief absent a violation of a federal right, and the PLRA has truncated that authority. See Benjamin v. Jacobson, 124 F.3d 162, 172 (2d Cir.1997).

The termination of a consent decree in response to the PLRA, therefore, merely effectuates Congress’s decision to divest district courts of the ability to construct or perpetuate prospective relief when no violation of a federal right exists. Given this shift in the relevant underlying law, the termination of prospective relief pursuant to the PLRA does not amount to a legislative reopening of a final judgment. 2. Rules of Decision. The plaintiffs next contend that the PLRA’s termination provision violates a different aspect of the separation of powers principle, articulated in United States v.

Klein, 80 U.S. (13 Wall.) 128, 20 L.Ed. 519 (1871). Klein had its genesis in the aftermath of the Civil War, when Congress passed a statute that permitted noncombatant Confederate landowners to recover confiscated goods upon proof of their loyalty to the Union. Klein, the administrator of the estate of Wilson, a Confederate sympathizer, attempted to recover Wilson’s goods pursuant to this statute. Mindful that the Supreme Court had previously declared that a presidential pardon was conclusive proof of loyalty, see United States v.

Padelford, 76 U.S. (9 Wall.) 531, 19 L.Ed. 788 (1869), Klein tendered evidence that Wilson had received such a pardon. While the case was pending, Congress passed a statute which declared that a presidential pardon was proof of disloyalty and directed the dismissal of any pending recovery action brought on behalf of a pardon recipient. See Klein, 80 U.S. (13 Wall.) at 131-34. The Supreme Court invalidated the new statute on separation of powers grounds.

It ruled that if the law were allowed to stand, then the trial court would have “jurisdiction of the cause to a given point; but when it ascertains that a certain state of things exists, its jurisdiction is to cease and it is required to dismiss the cause for want of jurisdiction.” Id., 80 U.S. (13 Wall.) at 146. Such a requirement “is not an exercise of the acknowledged power of Congress to make exceptions and prescribe regulations to the appellate power” and thus “passe[s] the limit which separates the legislative from the judicial power.” Id., 80 U.S. (13 Wall.) at 146-47.

The Klein Court distinguished Wheeling Bridge as a situation in which “the court was left to apply its ordinary rules to the new circumstances created by the act [whereas in Klein ] no new circumstances have been created by legislation. But the court is forbidden to give the effect to evidence which, in its own judgment, such evidence should have, and is directed to give it an effect precisely contrary.” Id., 80 U.S. (13 Wall.) at 147.

The plaintiffs argue that the ease-at hand resembles Klein more than Wheeling Bridge because the law underlying the consent decree — the Eighth Amendment — remains constant, yet the PLRA imposes a rule of decision by instructing courts to terminate prospective relief. This argument misapprehends the situation. As noted above, the relevant underlying law for present purposes is not the Eighth Amendment, but the power of federal courts to grant prospective relief absent- a violation of a federal right.

Thus, the PLRA does not run afoul of Klein because it does not tamper with courts’ de-cisional rules — that is, courts remain free to interpret and apply the law to the facts as they discern them. Because the PLRA leaves the courts’ adjudicatory processes intact, it does not transgress the Klein doctrine. See Gavin v. Branstad, 122 F.3d 1081, 1088, 1089 (8th Cir.1997). B: Due Process. The plaintiffs base their next two objections on the Due Process Clause.

The first rests on the postulate that the consent decree is a final judgment, the existence of which vests property rights in the parties that cannot be alienated by Congress. By purporting to terminate consent decrees, this thesis- runs, the PLRA not only reopens final judgments but also robs the judgments’ beneficiaries of rights secured to them thereunder. The plaintiffs’ second objection posits that the 1979 consent decree constitutes a contract and that due process limits the extent to which the federal government can enact legislation that has a deleterious effect on preexisting contracts.

Both objections lack force. 1. Vested Rights. The plaintiffs’ first objection fails because, at least in the absence of exceptional circumstances well beyond any that, are present here, frankly modifiable decrees cannot create vested rights. See Landgraf, 511 U.S. at 273, 114 S.Ct. at 1501 (noting that “relief by injunction, operates in futuro, and that [a party] ha[s] no vested right in the decree entered by the trial court”) (citation and internal quotation marks omitted).

As we have already pointed out, consent decrees are not merely final judgments, but a special species of that genre — final judgments that can be “reopened ... to the extent that equity requires.” Rufo, 502 U.S. at 391, 112 S.Ct. at 764. In the instant case, equity requires termination of the 1979 decree because Congress has withdrawn the power that animated the decree. See 18 U.S.C.A. § 3626(a)(1)(A), (b)(2). To be sure, the plaintiffs argue that this reasoning is circular. But, given the tenuous nature of consent decrees, that argument will not wash.

There is a basic difference between a money judgment and a consent decree: the former is fixted, whereas the latter is necessarily impermanent. Thus, insofar as a consent decree has prospective effect, it must on motion be adjusted to accommodate material changes of fact or law germane to its issuance. See Rufo, 502 U.S. at 393, 112 S.Ct. at 764-65. Here, the PLRA has altered the standard by which courts can continue forward-looking relief, and this profound change in the relevant underlying law entitles the defendants to termination of the decree. 2. Contract Rights.

The plaintiffs’ second due process objection is equally unavailing. Even if we make two broad assumptions that are integral to their position — namely, that the consent decree is a contract and that the PLRA impairs that contract — the objection founders. The Supreme Court delineated the standard of review for federal legislation that impairs contractual relations in National R.R. Passenger Corp. v. Atchison, Topeka & Santa Fe Ry. Co., 470 U.S. 451, 105 S.Ct. 1441, 84 L.Ed.2d 432 (1985).

If a substantial impairment of a contract right is found or assumed, “the reviewing court next determines, whether the impairment is of constitutional dimension.” Id. at 472, 105 S.Ct. at 1455. It engages in this analysis by examining the statute and identifying the parties to the contract. See id. “When the contract is a private one, and when the impairing statute is a federal one, this next inquiry is especially limited, and the judicial scrutiny quite minimal.

The party asserting a Fifth Amendment due process violation must overcome a presumption of constitutionality and establish that the legislature has acted in an arbitrary and irrational way.” Id. (citations and internal quotation marks omitted). Even though the federal government is .not a party to the “contract” in issue here (the consent decree), the plaintiffs seek to upgrade the level of scrutiny. Their gambit depends upon the Court’s opinion in Garcia v. San Antonio Metro.

Transit Auth., 469 U.S. 528, 105 S.Ct. 1005, 83 L.Ed.2d 1016 (1985), which, they say, stands for the proposition that the states have sufficient representation in the federal government to influence its actions, and that, by extension, the federal sovereign’s actions should be attributed to the states.

From this coign of vantage, they argue that, since federal laws are enacted by a government organized for the benefit of the several, states, a federal act that impairs a contract to which a state is a party should receive the same degree of scrutiny as a federal act that impairs a contract to which the United States is a party. This ratiocination is predicated on a strained reading of Garcia. The Garcia Court held that state participation in the federal government provides a sufficient safeguard to prevent federal overreaching vis-a-vis the states. See id. at 552, 105 S.Ct. at 1018.

There is, however, no basis in Garcia or elsewhere to suggest that federal legislation which benefits state governments is tantamount to self-dealing and thus subject to heightened scrutiny. We therefore summarily reject the plaintiffs’ reading of Garcia and the attendant claim that the federal government somehow became a constructive party to the 1979 consent decree. This gets the grease from the goose. Because the federal sovereign is not a party to the consent decree, either in fact or by indirection, we need only subject the PLRA to a rational basis review.

See National Passenger, 470 U.S. at 471-72, 105 S.Ct. at 1454-55; see also United States Trust Co. v. New Jersey, 431 U.S. 1, 22, 97 S.Ct. 1505, 1518, 52 L.Ed.2d 92 (1977) (holding that “[legislation adjusting the rights and responsibilities of contracting parties must be [based] upon reasonable conditions and of a character appropriate to the public purpose justifying its adoption”). Stressing that the PLRA abrogates existing responsibilities, the plaintiffs make the obligatory argument that the law is arbitrary and irrational. But these animadversions vastly overstate the case.

The PLRA only affects agreements that .have at all times remained subject to modification should circumstances change. And, moreover, by facilitating termination, the PLRA’s termination provision forges a practical, eommonsense linkage between a changed circumstance— the district courts’ newfound inability _ to grant or enforce prospective relief absent a violation of a federal right — and an existing consent decree. Consequently,

section 3626(b)(2) survives rational basis scrutiny. C. Equal Protection. The plaintiffs also advance a pair of arguments based on the Equal Protection Clause. First, they note that pretrial detainees, by definition, have not yet been convicted of the crime(

s) with which they have been charged. Thus, they enjoy both the presumption of innocence, see In re Winship, 397 U.S. 358, 90 S.Ct. 1068, 25 L.Ed.2d 368 (1970), and the right not to be punished prematurely, see Bell v. Wolfish, 441 U.S. 520, 99 S.Ct. 1861, 60 L.Ed.2d 447 (1979). Building, on this foundation, the plaintiffs assert that the PLRA is subject to strict scrutiny (which it fails) because it abridges these fundamental rights. In the alternative, they claim that the Act violates core principles of equal protection because it has no rational relationship to legitimate state interests. 1.

Fundamental Rights. Although the PLRA circumscribes a district court’s ability to provide prospective relief to pretrial detainees (as well as all other prisoners) absent a violation of a federal right, we conclude that this feature of the Act does not abridge the pretrial detainees’ right to be free from punishment. Prison conditions either violate fundamental rights (in which event they also violate federally secured rights) or they do not violate fundamental rights (in which event they do not violate federally secured rights). In the former ease, the PLRA permits relief to redeem the fundamental right.

In the latter case, the PLRA does not permit relief, but as no violation exists, the PLRA’s denial of relief does not imperil pretrial detainees’ fundamental rights. It is also possible to argue that the PLRA implicates the fundamental right of access to- the courts, see Wolff v. McDonnell, 418 U.S. 539, 578, 94 S.Ct. 2963, 2985-86, 41 L.Ed.2d 935 (1974), and that, by withdrawing the power to grant inmates prospective relief in a manner available to other blasses of people, the PLRA trammels inmates’ rights of access. This line of reasoning does not withstand close examination.

Under the PLRA, the courthouse doors remain open and the withdrawal of prospective relief — above and beyond what is necessary to correct the violation of federally protected rights — does not diminish the right of access. In a nutshell, while, there is a constitutional right to court access, there is no complementary constitutional right to receive or be eligible for a particular form of relief. See Crowder v. Sinyard, 884 F.2d 804, 814 (5th Cir.1989), abrogated on other grounds by Horton v. California, 496 U.S. 128, 110 S.Ct. 2301, 110 L.Ed.2d 112 (1990). 2. Rational Basis.

The plaintiffs’ final constitutional challenge suggests that the PLRA violates the Equal Protection Clause because it “singl[es] out a certain class of citizens for disfavored legal status or general hardship[].” Romer v. Evans, — U.S. -, -, 116 S.Ct. 1620, 1628, 134 L.Ed.2d 855 (1996). This suggestion is ill-conceived. A statute that neither abridges a fundamental right nor operates.against a suspect class receives rational basis review when it is challenged under the Equal Protection Clause. See Heller v. Doe, 509 U.S. 312, 318-19, 113 S.Ct. 2637, 2641-42, 125 L.Ed.2d 257 (1993).

The PLRA is such a statute: as we already have explained, it does not impair a fundamental right, and the plaintiffs do not assert that pretrial detainees are a suspect class. Thus, rational basis review applies. A statute survives rationality review if it “bear[s] a rational relationship to an independent and legitimate legislative end.” Romer, at -, 116 S.Ct. at 1627.

The PLRA’s legislative history indicates that the drafters intended the Act to “address the alarming explosion in the number of frivolous lawsuits filed by State and Federal prisoners,” to “mak[e] it much more difficult for Federal judges to issue orders directing the release of convicted criminals from prison custody,” 141 Cong. Rec. 14,413 (1995) (statement of Sen. Dole), and to wrest control of state, penitentiaries from federal courts so that states “will be able to run prisons as they see fit unless there is a constitutional violation,” id. at 14,419 (statement of Sen. Abraham).

These purposes are clearly legitimate. They involve the allocation of public resources, the maintenance of public safety, and the desire to institutionalize a state-centric conception of ■ our federal system. The means chosen to effect these ends are stern, but they certainly bear a reasonable relationship to the announced legislative goals. From this perspective, the PLRA easily passes rational basis review. The plaintiffs try to undermine this appraisal by asserting that an anti-inmate animus drove Congress’s approval of.the PLRA.

They claim that such an invidiously discriminatory intent violates the Court’s admonition that a legislature cannot construct legislation “for the purpose of disadvantaging the group burdened by the law.” Romer, at -, 116 S.Ct. at 1628; see also United States Dep’t of Agric. v. Moreno, 413 U.S. 528, 534, 93 S.Ct. 2821, 2826, 37 L.Ed.2d 782 (1973) (“[I]f the constitutional conception of ‘equal protection of the laws’ means anything, it must at the very least mean that a bare congressional desire to harm a.politically unpopular group cannot constitute a legitimate governmental interest.”).

We need not grapple with the intriguing question of whether the Romer Court meant to add a new “animus test” to the armamen-tarium of rationality review. The short, entirely dispositive answer to the plaintiffs’ supplication is that the evidence in the record simply does not show that the legislature inappropriately sought to disadvantage the plaintiff class. The only “proof’ that the plaintiffs offer consists of political rhetoric, such as the statement on the Senate floor that “criminals, while they must be accorded their constitutional rights, deserve to be punished.

Obviously, they should not be tortured or treated cruelly. At the same time, they also should not have all the rights and privileges the rest of us enjoy. Rather, their lives should, on the whole, be describable by the old concept known as ‘hard time.’ ” 141 Cong. Rec. 14,419 (1995) (statement of Sen. Abraham).

Passing the obvious point that such rhetoric must be taken with a grain of salt — elected officials, after all, have been known to strike poses for public consumption — the most that fairly can be said is that oratory of this sort may evince a philosophical shift; it hardly betokens an impermissible animus.

In all events, the state is well within its right to punish persons convicted of crimes, and a retributive desire to effect such punishment consequently does not offend any supposed “animus test.” Furthermore, the plaintiffs are not criminals, but pretrial detainees; they have not been found guilty of any crimes. Thus, even if the political rhetoric spotlighted by the plaintiffs qualified as animus directed at criminals, it would not constitute cognizable animus for present purposes.

In sum, an objective reading of the legislative history demonstrates that the plaintiffs’ inability to obtain prospective relief does not spring from Congress’s wish to do them harm, but from its desire to minimize the occasion for federal courts to administer state prisons. Consequently, the PLRA does not succumb to any theoretical “animus test” contained within the Equal Protection Clause. IV. APPLYING THE PLRA The plaintiffs have a fallback position.

They contend that, even if- the PLRA is constitutional, the 1979 consent decree should remain intact because (1) the district court previously made findings sufficient to save the decree by operation of the Act, see 18 U.S.C.A. § 3626(b)(2), or (2) if the findings to date are inadequate, the district court should have conducted an inquiry into whether a violation of a federal right exists currently (or probably will come into existence if the strictures of the consent decree are lifted) before implementing the PLRA’s termination provision. We reject both contentions.

Answering the question of whether prison conditions constitute an ongoing violation of a federal right under the PLRA necessitates both a definition of the right at stake and an assessment of a specific compendium of prison conditions. Accordingly, such a question comprises a mixed question of fact and law, the answer to which we review “along a degree-of-de'ference continuum, ranging from plenary review for law-dominated questions to clear-error review for fact-dominated questions.” Johnson v. Watts Regulator Co., 63 F.3d 1129, 1132 (1st Cir.1995).

Here, the question is more factual than legal: inasmuch as the double bunking of pretrial detainees does not in and of itself violate the Constitution, see Bell v. Wolfish, 441 U.S. at 541, 99 S.Ct. at 1875, the district court’s conclusion that the double bunking of which the plaintiffs continue to complain is not in violation of a federal right must be challenged, if at all, principally on the facts. Thus, the standard of review is highly deferential. See Huguley v. General Motors Corp., 999 F.2d 142, 146 (6th Cir.1993).

We have carefully reviewed the record and culled out the sparse factual findings that the court made in the relevant time frame. No useful purpose would be served by examining these findings in minute detail. Judge Keeton concluded that , they did not satisfy the requirements of

section 3626(

a) or (b). See D. Ct. Op., 952 F.Supp. at 880. A trial court generally is thought to be the best interpreter of its own prior rulings and findings, see, e.g., Martha’s Vineyard Scuba Headquarters, Inc. v. Unidentified, Wrecked & Abandoned Steam Vessel, 833 F.2d 1059, 1066-67 (1st Cir.1987); Lefkowitz v. Fair, 816 F.2d 17, 22 (1st Cir.1987), and this case is no exception. At any rate, we agree with Judge Keeton’s conclusion. It is simply implausible to suggest, on this record, that the district court’s assessment of the existing factual findings is clearly erroneous.

The plaintiffs’ follow-on argument gains them no ground. As to the conditions that presently exist, we defer to the district court’s intimate familiarity with this protracted litigation and to its informed evaluation of current prison conditions. See D. Ct. Op., 952 F.Supp. at 880 (observing that “no evidence is before the court to support findings that defendants are not in compliance with the terms of the modified Consent Decree”).

Deference is especially appropriate here since, under the terms of an order that it entered on June 14, 1994, the district court for some time had been receiving and evaluating periodic reports from the Sheriff concerning incidents of assaultive behavior, rape, disease, and the like at the Nashua Street jail. As to what the future may bring, we cannot criticize Judge Keeton’s reluctance to play the oracle. Presented with the opportunity to make further findings before deciding the defendants’ motions, the judge declined.

He noted several cogent reasons why it made sense to leave the question of whether a violation of a federal right might follow the termination of prospective relief under the consent decree to another day. See id. We discern no error. This is neither the time nor the place to press an inherently speculative claim of harm to come. The PLRA imposes no obligation on the trial court to make a predictive inquiry into future conditions before terminating an existing consent decree, and we are not aware of any other basis for burdening the court with such a requirement.

Quite often, “[p]resent fears are less than horrible imaginings.” William Shakespeare, Macbeth, act 1, sc. 3 (1605). If, in this instance, the plaintiffs’ trepidation proves justified, they remain free to initiate a new round of proceedings designed to show that post-termination prison conditions actually do violate their federally protected rights. V.

VACATING THE CONSENT DECREE Having construed the PLRA and established that its termination-of-prospective-relief provision passes constitutional muster, that the conditions for exemption have not been met, and that the Act’s mandate requires the district court to terminate the consent decree, we now mull whether that mandate means that an order must be entered not only terminating the consent decree but actually vacating it. The district court thought not. See D. Ct. Op., 952 F.Supp. at 883-84. We agree. The defendants’ opposition is easily dispatched.

Nothing in the PLRA or its legislative history speaks of vacating consent decrees. Congress chose to use the verb “terminate” and to eschew the verb “vacate.” The distinction between these two words is clear: “terminate” means “to put an end to” or “to end,” Black’s Law Dictionary at 1471, whereas “vacate” means “to annul” or “to render ... void,” id. at 1548. In the present context, this distinction may well possess practical significance. Cf. Benjamin, 124 F.3d at 177-178 (explaining that court’s view of the distinction between terminating prospective relief and vacating a consent decree).

While terminating a consent decree strips it of future potency, the decree’s past puissance is preserved and certain of its collateral effects may endure. Vacating a consent decree, however, wipes the slate clean, not only rendering the decree sterile for future purposes, but also eviscerating any collateral effects and, indeed, casting a shadow on past actions taken under the decree’s imprimatur. As nothing in the PLRA even hints that consent decrees must be vacated when prospective relief is terminated, we uphold the district court’s ruling that the PLRA does not require vacation of the 1979 decree. VI.

CONCLUSION We need go no further. To the extent that the parties advance other arguments, we reject them out of hand. None requires elaboration. For the reasons stated herein, we affirm so much of the judgment below that (

a) found the PLRA to be constitutional, (

b) terminated all prospective relief under the 1979 consent decree, and (

c) refused to vacate that decree. We direct, however, that the judgment be revised to terminate the consent decree itself and we remand for the entry of a modified judgment (together with .such further proceedings, if any, as the district court may deem necessary in light of this opinion). Affirmed as modified and remanded. All parties shall bear their own costs. . Notwithstanding the several emendations that have heen made to the original consent decree, we refer to the decree) as modified from time to time, as the "1979 consent decree.” .

Because Congress intended the PLRA to effect the termination of consent decrees, we need not elaborate upon-what consequences might follow from the termination of prospective relief alone. We note, however, that the Second Circuit has invested substantial time in exploring the potential ramifications of terminating prospective relief while leaving a consent decree otherwise intact. See Benjamin v. Jacobson, 124 F.3d 162, 177-178 (2d Cir.1997). Inasmuch as our

interpretation of the PLRA obviates the need for any such exercise, we take no view of the Benjamin court's conclusions. . This precept could not come as a surprise to the plaintiffs. In the last modification of the consent decree, under date of June 14, 1994, the district court advised the parties that it would entertain future motions to modify "upon a showing of good cause ... or upon a showing of material change in circumstances.” . The Commissioner moved below for vacation of the 1979 consent decree under Fed.R.Civ.P. 60(

b) and now appeals the denial of that motion. We need not address that aspect of the matter. At oral argument in this court, the Commissioner agreed that if the consent decree were to be terminated, the Rule 60(

b) issue could be set to one side. We take the Commissioner at his word and therefore express no opinion as to the merits of the Rule 60(

b) claim.