Legislative power and judicial power.
Saint Augustine remarked that he understood time until someone asked him to explain it. (1) That legislative and judicial power are conceptually distinct may seem obvious, but explaining the difference between them is not so easy.
This Essay will argue that there are two possible theories of the difference between the legislative and judicial powers referred to by the Constitution, and that both of them have surprising implications. According to one account, the difference between the two powers is wholly structural, in that it depends entirely on the functions of government institutions. This account regards the difference between legislative and judicial power as the difference between the power to make legal rules and the power to apply them. The surprising implication is that although the powers differ in the form through which they are exercised, the legislative power can bring about any legal effect the judicial power can bring about, including the alteration of legal positions established by prior litigation. Some legal effects, however, can be achieved only by legislation. If the two powers are understood wholly structurally, the judicial power amounts to a sub-part of the legislative power.
The alternative to that conclusion is that the difference between the two powers is substantive in that the legislative power is limited in its operation with respect to some legal interests but not others. On this account, the limits of legislative power relative to judicial power are marked by legal interests that legislation may not change but that may be operated on by judicial power pursuant to preexisting rules. Some rights are vested. Those rights are identified, not by distinguishing between making and applying rules, but on other grounds. The classic nineteenth-century doctrine of vested rights was often described in terms of the distinction between legislative and judicial power, so the substantive roots of one leading account of that distinction are reasonably well known. As I will explain, the Supreme Court's current doctrine limiting Congress' power to undo final judgments is also substantive and not structural.
Of those two conclusions, perhaps the more surprising is the first, and it may seem so implausible that it cannot be correct. If legislative power can accomplish anything judicial power can, what is the point of assigning them to distinct institutions that are designed to be politically independent of one another? Even if the legislature has that much power, which legislatures subject to constitutional limitations do not, judicial independence can force legislators to exercise their authority through written law, and hence publicly and with as much clarity as is needed to constrain the courts.
This Essay begins by explaining how the power to make rules is in effect a perfect substitute for the power conclusively to apply them. It also points out that two possible constraints on legislative power that might be thought to be structural--requirements of generality and prospectivity--cannot plausibly be attributed to the legislative power granted by Article I of the Constitution. If the difference between legislative and judicial power involves only the functions of government institutions, a statute can do anything a judgment can do. The Essay then argues that understandings of the difference between the two powers that do constrain the legislature rest, not on different functions of government, but on the differences among legal interests. That was quite clear about the nineteenth-century doctrine of vested rights. Under that doctrine, courts held that some legal interests were immune from change by legislation enacted after the interest was created. (2) That doctrine was often justified as reflecting the separation of legislative and judicial power, but protected only some legal interests, interests that were identified on grounds of justice and the public good. Perhaps more surprising is that the Supreme Court's current doctrine concerning legislative interference with judgments has the same feature: It protects some interests and not others, and identifies the protected interests on grounds that do not derive from different government functions. Indeed, the distinction the Court draws closely resembles old-style vested rights doctrine. The Essay concludes by explaining how separation of powers furthers the rule of law even if the legislature has complete control over the law's content.
I. MAKING AND APPLYING LEGAL RULES
Perhaps the most natural account of the difference between legislative and judicial power is that it is the difference between two functions of government institutions: making legal rules and authoritatively applying legal rules.
If that is the difference, then a legislature can in effect do anything a court can do, but not vice versa. When they decide lawsuits, courts bring about two kinds of legal results. First, they conclusively resolve disputed questions of law and fact. (3) If A sues B for breach and the court concludes that B is in breach, that conclusion will bind the parties in the future. Declaratory judgments do explicitly what all merits judgments do implicitly, conclusively establishing the legal relations of the parties under the law as it stands when the judgment is issued. (4)
Adjudication can also involve the creation of new legal rules that bind the parties. If A prevails in a suit for damages, the judgment creates a new legal obligation, into which A's preexisting claim is said to be merged. (5) Now B must pay A, and A has no other claim against B based on those facts. Injunctions similarly create new, highly specific, legal rules by imposing obligations on the party enjoined. A manifestation of this point is the crime of contempt for violating an injunction, which is distinct from the substantive law on which the injunction rests. (6)
A government institution that can make the legal rules whatever it wants them to be can in effect perform both of those functions. If two parties have a dispute about the consequences of pre-existing legal rules for their current relationship, a lawmaking institution can provide that whatever those consequences may have been, they shall henceforth be as the institution prescribes. If A and B are in an automobile accident, an institution with that power can say that henceforth B shall have no obligation to pay A damages arising out of that accident, whatever B's obligation previously may have been. A new legal rule could also say that future relations between A and B shall be governed on the assumption that certain events had happened in the past, and thus effectively establish facts the way a court can. An institution that can make rules thus can produce the same result as a court does in a case for damages. It can also provide that B shall have an obligation to pay, again without regard to what the parties' relationship may have been. It can state rules for B's conduct that replace or go beyond the rules that existed before, and so generate the same result as an injunction.
Anything a law-applying institution can do a law-making institution can do, but the reverse does not hold. A law-applying institution cannot announce a wholly new rule, not derived from existing law, and use it to create new obligations like a damages judgment. Application is application of the pre-existing rules. (7) If legislative and judicial power are understood as law-making and law-applying power, judicial power is a sub-part of legislative power from the standpoint of legal consequences though not of legal formalities.
Despite the long-standing association of legislative power with generality and prospectivity, neither of those features necessarily accompanies the power to make legal rules, because legal rules can be specific and retrospective. An example of specificity in the Constitution itself is in the Twenty-Second Amendment, which provides that it "shall not apply to any person holding the office of President when this article was proposed by the Congress." (8) Congress proposes constitutional amendments at particular times, and only one person can be President at one time. Despite its extreme specificity, the proviso is a legal rule. It has legal consequences when applied to facts.
In similar fashion, a rule can be retrospective and still be a rule; if that were not possible, the problem of retrospectivity would not arise. Although legal rules cannot change the past, they can direct that future actions depend on events that have already occurred, and so be both prospective and retrospective at the same time. An ex post facto law, the classic example of an objectionably retrospective rule, can operate precisely because it is a rule for future conduct that depends on past events: under an ex post facto law, officials are directed to impose criminal punishment on the basis of past events. (9) Unrestricted law-making power can be retrospective as easily as it can be specific. (10)
Perhaps the legislative power granted by Article I of the Constitution is not the unrestricted law-making power I have described, because it is limited to general and prospective rules. Such a conception of legislative power would be structural and not substantive in that it would apply alike to all legal interests, depending on a description of the functions of government institutions and not the different interests involved in different situations. That understanding of legislative power would complement a description of judicial power according to which it alone acts specifically and retrospectively.
Accepted legislative functions, however, include highly specific rules, as American constitutional practice reflects. Congress has long conferred benefits on named individuals, for example by indemnifying executive officials who have incurred personal liability in the course of their responsibilities. (11) Highly specific statutes that impose disadvantages are constitutionally problematic, but that is because they impose disadvantages, not because they are highly specific. (12) Legislative power can legitimately operate with complete specificity. (13)
Whether the legislative power in Article I may operate retrospectively is a more involved question, because the distinction between prospective and retrospective legislation requires some careful thought. The broadest conception of retrospectivity includes all rules that change the legal consequences of events that have already occurred. A law that does so need not require any knowledge about the past for its application. If a legislature were to eliminate the remedy of specific performance, the consequences of earlier contracts would change, but a court would not need to know whether the parties had actually made a contract in order to apply the rule. Nineteenth-century judges who were zealous in their protection of vested rights recognized substantial authority to change the law of remedies as it applied to existing contracts, provided the substance of the right involved was not impaired. (14) Even less controversial examples of acceptable exercises of legislative power that change legal consequences are easy to find. Changes in the formalities of title transfer, for example, alter the legal powers that property owners acquire when they take title. Laws like that are in the broadest sense retrospective, but are also within the legislative power as it is generally understood. (15)
A more plausible place to draw the line between law-making and legislative power may seem to be at rules that require some information about past events for their application. Ex post facto laws have that feature. But so do other rules that are clearly exercises of legislative power. As with specificity, laws conferring benefits provide historical examples. In the later nineteenth century, Congress provided pensions for Union veterans of the Civil War, not by name, but by a rule that depended on their earlier military service. (16) Even some burdens imposed retrospectively are unproblematic exercises of legislative power. When a state raises its drinking age, the new and more restrictive rule depends for its application on earlier events, but is no less legislation for that reason. (17) Quarantines for individuals who have been exposed to a contagious disease fall into the same category, requiring information about the past for their application, but also qualifying as legislation. (18)
Ex post facto laws, by contrast, are defined not only by their reference to the past, but by criteria having nothing to do either with past or future or the functions of government. Only criminal punishments qualify as ex post facto laws, and criminal punishment is identified by the interests on which the law operates and the reasons on which it rests. (19) As the quarantine example demonstrates, not all burdens based on prior events are punitive because some are imposed for other purposes. Much the same is true with respect to specificity and bills of attainder: a law can be extremely specific, can even impose a burden as opposed to providing a benefit, and still not be a bill of attainder. If a class of one is legitimate, the rule using it is not a bill of attainder, and legitimacy goes to the reasons on which the legislature acts.
Judicial decisions are based on past events, but they actually affect future events, including the future consequences of prior occurrences. The power to make rules for the future therefore can accomplish anything courts can accomplish. If legislatures hold that power by virtue of being legislatures, they can do anything a court can do.
II. VESTED RIGHTS, OLD AND NEW
Another understanding of the difference between the two kinds of government authority does not turn wholly on the distinction between making and applying rules. That understanding takes some but not all legal interests to be immune to some forms of alteration by the legislature, but subject to adverse action by courts pursuant to pre-existing law, as when property is lost through a criminal fine. The nineteenth- and early twentieth-century doctrine of vested rights combined substance and separation of powers in that fashion: although sometimes explained and justified on the basis of separation of powers, it protected some but only some legal interests from alteration by the legislature. Seen as a separation of powers principle, it assumed that differences among legal interests were built into the definitions of legislative and judicial power. Although the Supreme Court has largely abandoned that doctrine, as I will explain, its current case law concerning legislative invasion of the judicial sphere resembles the old approach in that it protects some but only some legal interests from legislative alteration. Old and new doctrines are alike in that they do not rest simply on the difference between making and applying rules, but on differences among legal interests.
A. NINETEENTH-CENTURY VESTED RIGHTS AND SEPARATION OF POWERS
For many decades, the basic doctrine of American constitutional law was that the government could not deprive people of vested rights. It could not take the property of A and give it to B. (20) That principle was often attributed to the difference between legislative and judicial power. (21) Only the latter could work deprivations of vested rights, and it could do so only pursuant to pre-existing legal rules, because judicial decisions must rest on the law as it stands. Legislatures may make new legal rules, but may not destroy vested rights when they do so. The vested rights understanding of the separation of powers built differences among legal interests into the definitions of legislative and judicial power. Legislatures could not interfere with vested rights, not because they had to act generally and prospectively, but because they could not interfere with vested rights. As long as they avoided doing so, legislatures could change the legal consequences of earlier events, and could operate with great specificity. Perhaps the most striking illustration of this point came in those few states that actually banned retrospective laws. Their courts understood those provisions to bar only laws that interfered with vested rights. (22) Avoiding all retrospectivity was unreasonable and probably impossible. Avoiding legislative forfeitures was mandatory.
The criteria by which nineteenth-century courts distinguished between legal interests that legislatures could affect and those they could not reflected considerations of justice and the public interest. Judge Thomas Cooley, the pre-eminent scholar of vested rights doctrine in its mature phase later in the century, explained that legislation could not invade vested rights, even when the legislature acted in the public interest, but that vested rights were not defined in any technical sense. Rather, a vested interest was one "which it is equitable the government should recognize, and of which the individual cannot be deprived without injustice." (23) Insofar as the doctrine he expounded rested on separation of powers, it also rested on conceptions of legislative and judicial power that turned on questions of justice, not simply the formal properties of different institutional actions.
More recently, Professor Caleb Nelson has shown how nineteenth-century concepts of judicial, legislative, and executive power incorporated principles of political philosophy that distinguished among the interests an individual might hold under the law. (24) Only the judiciary, not the legislature or the executive, could "declare that a competent private individual no longer retained core private rights previously vested in him." (25) Core private rights were defined in terms of the state of nature of Lockean political theory, and contrasted with privileges of individuals that were created by the legislature to further the public interest. (26) The difference between natural rights and state-created privileges, like the difference between just and unjust deprivations, reflects normatively relevant differences among legally protected interests. It does not simply reflect the difference between making and applying law. Like Cooley, Nelson elaborates on the substantive content of nineteenth century conceptions of legislative and judicial power.
B. THE COURT'S CURRENT DOCTRINE, SEPARATION, AND VESTED RIGHTS
Little is left of the once basic doctrine. Congress has substantial authority to change the future legal consequences of past events. That authority extends to the consequences of many past events that created property rights. (27) Most commentators today probably would explain this shift as a change in substantive due process doctrine, perhaps neglecting the separation of powers rationale of the earlier version.
According to the Supreme Court today, separation of powers (at least at the federal level) does impose one quite specific limit on Congress' ability to change legal relations after the fact. Linder Plaut v. Spendthrift Farm, (28) Congress may not reopen final judgments for damages in federal court. As I will explain, Plaut is like nineteenth-century jurisprudence in that it attributes to separation of powers a result that can be explained only by distinguishing some legal relations from others, and so recognizes a form of vested rights.
Plaut grew out of Congress' response to an earlier decision by the Court, Lampf Pleva, Lipkind, Prupis & Petigrow v. Gilbertson. (29) In Lampf, the Court resolved a question that had divided the courts of appeals, and concluded that certain private federal securities fraud suits had to be brought within one year of discovery of the violation and three years of the violation itself. (30) After Lampf, a number of securities-fraud cases that were in progress when the case was decided were time barred, even though they had been timely filed according to court of appeals precedent applicable when and where they were brought. (31) While some of the plaintiffs in those cases pursued appeals, others let their cases go to final judgment for the defendant. Plaut was one of those. (32) In response to Lampf, which defeated expectations based on court of appeals precedent, Congress added Section 27A to the Securities Exchange Act of 1934. Section 27A provides that the limitations period for securities fraud cases filed before the day on which Lampf was decided shall be the law applicable in the jurisdiction as of that date. It also provides that cases like Plant, which were dismissed as time-barred under Lampf but would have been timely under the limitations period set out in the statute "shall be reinstated on motion by the plaintiff" if the motion is made within a period set out in the statute. (33)
When the plaintiffs in Plaut moved to reinstate their action pursuant to the statute, the defendants challenged that provision on constitutional grounds, claiming that the instruction to re-open final judgments impermissibly intruded on the judicial power. The Supreme Court agreed, finding that Congress had exceeded the permissible bounds of legislation and sought to disrupt the judicial power by making a previously final judgment non-final. (34)
On one reading Plaut was a purely formal decision, keyed to the formality that Congress employed to achieve its goal. The statute provided that cases already decided should be reopened. It did not give the plaintiffs in those cases a new cause of action, based on the same facts as underlay the earlier cases but subject to a different limitations period. The latter kind of statute would not, strictly speaking, have required that any case be reopened, though it would have produced the same result. It would have undermined the effect of a final damages judgment by limiting the judgment's preclusive reach, keeping it from applying to the newly-enacted rule governing prior events. Especially because the Court in Plaut spoke through Justice Scalia, it is possible that its holding turned on the formal point that Section 27 A told courts to reopen prior judgments. If that is so, Plaut is consistent with the principle that the legislative power can accomplish anything the judicial power can, but means that legislatures must make law the way legislatures make it, providing the rules that will govern cases rather than purporting to direct courts in their decision.
More likely Plant concerns the substance of congressional power and not the form through which it operates, and the Court would have come to the same conclusion had Section 27A said that it created a new right for parties whose claims had been dismissed. The Court traced its understanding of separation of powers to the Framers' rejection of "a system of intermingled legislative and judicial powers, which had been prevalent in the colonies long before the Revolution, and which after the Revolution had produced factional strife and partisan oppression." (35) That reasoning treats the principle being applied as an important feature of the structure of government, not just a matter of proper drafting. The case does not, however, stand for the proposition that Congress may never require that judgments be revised in light of changes in the law. Justice Scalia explicitly did not disturb a line of cases in which the Court has found that injunctions should be adjusted to reflect changes in the law on which they rest, be it the law governing primary legal obligations or the law of remedies. (36)
The foundational decisions in that line of cases were made in the 1850s, in a case in the Court's original jurisdiction. In Pennsylvania v. Wheeling & Belmont Bridge Company (Wheeling Bridge I), (37) the State of Pennsylvania complained that the Wheeling Bridge, which spanned the Ohio River at Wheeling, Virginia, was an obstruction to interstate commerce and a nuisance. The Court agreed, and ordered that the bridge be raised or removed. (38) Congress then passed a statute providing that the bridge was a lawful structure. (39) When Pennsylvania returned to the Court seeking enforcement of the injunction, the Court concluded in Wheeling Bridge II that the injunction should be dissolved because the applicable law on which it rested had been changed by the legislature. (40) Congress, the Court found, had not sought illicitly to exercise the judicial power by deciding what was and was not an obstruction to commerce under the applicable law. It had exercised its power to regulate commerce by changing the law and making the bridge lawful, not finding that it was. Once the law had changed, the Court was obliged to adjust its decree accordingly. (41) The Court continues to follow Wheeling Bridge II, adjusting ongoing injunctions in federal court so that they match legislative changes in both the substantive and remedial law. (42)
Plaut thus means that any statute with the effect of overturning a final damages judgment is unconstitutional, even if formulated purely in terms of substantive entitlements like a right to damages. (43) That doctrine does not rest on the difference between making and applying rules. The provision at issue in Plaut, considered as a change in legal relations, did not contradict the Court's decision in Lampf, because it was not and did not need to be an application of the law as it stood when Lampf was decided. Section 27A was consistent with the assumption that Lampf correctly applied the law as it then stood. Supporters of legislative relief for parties disfavored by the decision may well have believed that the Court was right but that its decision nevertheless worked an injustice because it defeated reasonable expectations. One way to act on that belief is to create a new rule of law applying to earlier events. The parties disfavored by Lampf were like Civil War veterans who received a benefit based on prior events that was not available to them under the law as it stood when those events took place. When Congress provided additional veterans' benefits after the fact, it did not contradict the conclusion that soldiers did not accrue those benefits when they served; it changed the law.
Justice Scalia in Plant seems not fully to have appreciated the Promethean character of the power to create any rule the lawmaker wishes and attach new consequences to prior events. Explaining why Section 27A invaded the judicial power, he explained that "[h]aving achieved finality, however, a judicial decision becomes the last word of the judicial department with regard to a particular case or controversy, and Congress may not declare by retroactive legislation that the law applicable to that very case was something other than what the courts said it was." (44) Congress in Section 27A did no such thing. That provision was consistent with the correctness of Lampf when Lampf was decided. Congress did not declare what the law was when Lampf came down, but rather what it was to be henceforth. (45) To say that the law was one thing at time 1 and another at time 2 is simply to say that the law has changed. To say that the law cannot change, as the Court in effect did in Plant, is simply to say that some rights are vested.
Although Plaut does not rest on the difference between making and applying rules, it might be thought to derive from the distinction between legislative and judicial power as they appear in Articles I and III. The Court's reasoning suggests that it found in the latter a form of claim preclusion according to which a final judgment relative to certain facts bars later litigation based on those facts but on new law. If Article III itself mandates any form of claim preclusion for federal court judgments, it is not the version that Plaut assumes. The scope of claim preclusion depends on the scope of claims. Justice Scalia assumed what the Restatement (Second) of Judgments calls the modern or transactional view. According to that understanding of claims, a claim includes all the grounds for a remedy growing out of some transaction, which it to say, some set of facts. (46) As the Restatement explains, earlier approaches took a narrower view. Under common law pleading, the claim associated with one writ often was distinct from that associated with another. Different claims were also thought to be associated with different primary rights. If any two primary rights differ, a right that existed when events took place and a right that did not exist then are different. (47) Article III was drafted in the days of common law pleading and the separation of law and equity, not the Federal Rules of Civil Procedure, which replace common-law writs and bills in equity with one form of action. (48)
Conceived as the authority conclusively to apply law to particular facts, the judicial power much more plausibly gives rise to issue and not claim preclusion, to use contemporary terminology. (49) In particular, a court can apply only the law that exists when it decides a case, and a legislature that creates new law does not call into question the court's earlier decision, nor does it reverse that decision the way an appellate court does. Claim preclusion extends beyond issue preclusion precisely because it reaches issues that the court did not decide.
Plaut does not rest on the difference between legislation and adjudication, nor on a plausible view of the effect of judgments that is required by Article III. It is a doctrine of vested rights: the effects of past events with respect to monetary liability produced by a damages judgment may not be changed subsequently.
Whether the Court's current doctrine should be characterized as substantive or structural is a more difficult question, mainly because those categories have an uncertain boundary. (50) However those concepts should be applied, the doctrine is based on differences in legal interests as opposed to the characteristics of government institutions. As the Restatement explains, a damages judgment gives rise to a new right, a debt owed the plaintiff by the defendant in the amount of the judgment. (51) Debts are payable in money, and so represent wealth or property in a quite abstract form, dissociated from any particular asset. An injunction gives rise to a quite different interest, the interest in the defendant's compliance with the court's order. (52) Those legal interests are distinct, without regard to the functions of government institutions.
Perhaps more striking is that undoing the judgment in a damages case, which means eliminating or creating a debt, is a pure transfer of wealth. For the beneficiaries of Section 27A who had meritorious claims (the statute of limitations disregarded), that provision simply created a right to receive payment from the defendants that otherwise did not exist. Undoing a judgment for the plaintiff in a damages action would create a new obligation for the successful plaintiff to pay an amount equal to the judgment. Pure transfers were the canonical violation of the principle of vested rights: the legislature was not allowed to take the property of A and give it to B. (53) Eliminating an injunction, by contrast, benefits the defendant and harms the plaintiff, but may have other consequences and lacks the exact correspondence of benefit and burden. Undoing an injunction is thus not necessarily pure redistribution.
The protection for judicially-created rights found in Plant is thus substantive in two important senses. First, it reflects differences in the interests protected by judgments that may and may not be undone by the legislature. Second, it reflects a difference that is of great importance from a policy standpoint, if the policy is one of limiting the legislature's power to bring about pure wealth transfers. In both respects today's doctrine closely resembles that of the nineteenth century, albeit it favors only rights created by judicial decree. (54)
III. SEPARATION OF POWERS AND THE RULE OF LAW
The main contention of this Essay is that the separation of legislative and judicial power has a surprising feature, however it is understood. If the distinction between the two powers tracks the difference between making and applying legal rules, then legislative power functions as a perfect substitute for judicial power. If not, that is because the difference between the two powers has built into it some more substantive component, the kind of substance that underlay classic vested rights doctrine. I have not sought to argue that one or the other conception is a better account of the powers referred to by Articles I and III of the Constitution. (55) In this section, I will argue that the wholly structural understanding cannot be dismissed on the grounds that separating the powers is pointless if the legislative power is so sweeping.
Dismissal of that conception for that reason may seem unavoidable. Justice Scalia in Plaut said that the Constitution's system of judicial independence was founded on the "ruins" of early state constitutions in which the two powers were not separated, as they were not in Britain at the time. (56) Article III creates a form of judicial independence. The judges are selected by joint action of two separately-elected institutions, President and Senate, and may be removed only with much difficulty. (57) Because they are independent, the courts are not simply the agents of the legislature the way lower-level executive officials may be thought to be agents of the President. (58) But if legislatures can in effect decide and re-decide particular cases, courts are nothing but agents of the legislature and making the two institutions independent is pointless. Hence, there must be something that courts can do that legislatures cannot, not just in formal but in more constraining terms.
Judges' independence of the legislature, however, is justified even if the legislative power can accomplish any effect the judicial power can. First, Congress, like every state legislature, is subject to constitutional limits that go beyond the separation of legislative and judicial power. For that reason, no American legislature can make the law whatever it wants, and so none wields the entire lawmaking power. Courts are obliged to apply the law, including those parts of the Constitution that limit or displace legislative acts. If the judges were dependent on the legislature for their salaries, they might be unwilling to perform the important function of looking to the superior law when it conflicts with a statute.
More interesting is that judicial independence is useful even where the legislature does have the discretion to make the law what it wants. When the legislature has that ability, and the judiciary is independent, the legislature can control the judiciary but must do so through the written law (provided the courts are motivated to follow the law). While the written law will be effective, intimations from influential members of the legislature, implicitly backed by the threat of sanction or the promise of reward, will not be. If the legislature can neither harm nor hurt the judges, it cannot influence them through channels other than statutes. In a system in which the legislature has complete control over the substance of the law, judicial independence nevertheless furthers the goals of publicity and clarity. Independent courts, even if they have no power to fix the law in place, can facilitate the rule of law.
* James Madison Distinguished Professor, University of Virginia School of Law. Thanks to participants in a workshop on an earlier version at the University of Virginia.
(1.) Saint Augustine The Confessions 217 (Philip Burton, trans., Robin Lane Fox, ed., 2001) ("What, then, is time? As long as no one asks me, I know; but if someone asks me and I try to explain, I do not know.").
(2.) An example well known to contemporary readers is Trustees of Dartmouth College v. Woodward, 17 U.S. 518 (1819), in which the Supreme Court held that the Contracts Clause protected Dartmouth College's corporate charter from legislative change. In Fletcher v. Peck, 10 U.S. 87 (1810), the Court through Chief Justice Marshall held that Georgia legislation purporting to rescind a grant of land previously made by the legislature similarly violated the Contracts Clause. Marshall's opinion in Fletcher shows that he viewed the constitutional protection of contracts as only one manifestation of a broader principle that legislatures could not interfere with vested rights. "But, if an act be done under a law, a succeeding legislature cannot undo it. The past cannot be recalled by the most absolute power. Conveyances have been made, those conveyances have vested legal estates, and, if those estates may be seized by the sovereign authority, still, that they originally vested is a fact, and cannot cease to be a fact." Id. at 135.
(3.) Restatement (Second) of Judgments [section] 27 (1982) (judicial decision of litigated issues is conclusive on the parties in subsequent cases).
(4.) Under the Declaratory Judgment Act, federal courts may "declare the rights and other legal relations of any interested party seeking such [a] declaration." 28 U.S.C.A. [section] 2201 (West 2010).
(5.) Restatement (Second) of Judgments [section] 18 cmt. a (1982) (judgment for plaintiff extinguishes plaintiff's claim and substitutes the judgment for it); Id. at cmt. c (judgment for plaintiff creates a debt in that amount from defendant to plaintiff).
(6.) Violation of a federal court's injunction is the crime of contempt. 18 U.S.C.A. [section] 402 (West 2016). As that statute shows, federal court orders create obligations under federal law. Those orders are often based on non-federal law, like the law of contract. See, e.g., Green Stripe, Inc. v. Berny's Internacionale, 159 F. Supp. 2d 51 (E.D. Pa. 2001) (granting preliminary injunction to remedy breach of contract and unjust enrichment).
(7.) Courts arc law-applying and not law-making institutions in the sense in which I am using the terms, although some courts make law in a manner of speaking by setting precedents. When a court's interpretation of a legal norm is taken as conclusive by later courts, as can happen under standard American principles of stare decisis, the earlier court's interpretation functions as if it were the norm itself, and so announcing the interpretation can function as the creation of a norm. Especially when the norm is itself unwritten, as are common law principles, it is quite common and natural to characterize the courts' conclusive expositions of those principles as judge-made law. Despite that functional similarity, courts do not make law in that they are supposed to act only on the basis of some law or body of law that they did not exist before they interpreted it. The distinction can become attenuated in practice, but in principle it is central to the concepts of adjudication and judicial power.
(8.) U.S. CONST, amend. XII, [section] 1.
(9.) After describing Caligula's supposed practice of writing laws in small print and posting them on a high pillar, Blackstone turned to ex post facto laws: "There is a still more unreasonable method than this, which is called making of laws ex post facto', when after an action ... is committed, the legislator then for the first time declares it to have been a crime, and inflicts a punishment who has committed it." See Sir William Blackstone, Commentaries on the Laws of England, 46 (1893). Although the law comes after the action, the punishment comes after the law.
(10.) Just as the Constitution provides an example of extreme specificity, so it provides an example of a retrospective rule. Section 3 of the Fourteenth Amendment imposes political disabilities on certain individuals who participated in insurrection or rebellion before it was adopted. U.S. CONST, amend. XIV, [section] 3.
(11.) Congress indemnified Captain George Little of the Navy for the judgment against him personally in Little v. Barreme, 6 U.S. 170 (1804), in which the Supreme Court found that Little had acted in excess of his statutory authority in seizing the plaintiff's vessel. Acl for the Relief of George Little, ch. 4, 6 Stat. 63 (1807). See James E. Pfander & Jonathan L. Hunt, Public Wrongs and Private Bills: Indemnification and Government Accountability in the Early Republic, 85 N.Y.U. L. Rev. 1862 (2010) (describing early congressional indemnification practice). Congress began passing private bills for the relief of private individuals who had suffered tort-like damage from government activities in the 1790s. See Alexander Holtzoff, The Handling of Tort Claims Against the Federal Government, 9 L. & CONTEMP. PROBS. 311,311 (1942).
(12.) The Supreme Court found that a statute that in effect discharged three specifically identified individuals from federal employment was a bill of attainder because it imposed punishment on them. See United States v. Lovett, 328 U.S. 303,316 (1946).
(13.) See Nixon v. Adm'r of Gen. Servs., 433 U.S. 425 (1977). When former President Nixon entered into an agreement concerning his presidential records pursuant to which some of them would be destroyed if he so directed, id. at 431-432, Congress passed a statute that referred to Nixon by name and forbade destruction of the records, id. at 429. In response to the argument that the legislation was a bill of attainder, the Court found that Nixon was a "legitimate class of one" because Congress acted on the basis of permissible reasons--the preservation of presidential records--that at the time applied only to him. Id. at 472.
(14.) An important example was the abolition of imprisonment for debt, which the Court regarded as permissible with respect to existing and future contracts. "The distinction between the obligation of a contract, and the remedy given by the legislature to enforce that obligation, has been taken at the bar, and exists in the nature of things. Without impairing the obligation of the contract, the remedy may certainly be modified as the wisdom of the nation shall direct. Confinement of the debtor may be a punishment for not performing his contract, or may be allowed as a means of inducing him to perform it. But the state may refuse to inflict this punishment, or may withhold this means, and leave the contract in full force. Imprisonment is no part of the contract, and simply to release the prisoner, does not impair its obligation." Sturges v. Crowninshield, 17 U.S. 122, 200-01 (1819) (Marshall, C.J.).
(15.) Although it may seem natural to say that a change in the formalities of transfer operates only prospectively, that is not correct. Because all laws are in one sense prospective, it is easy to emphasize the prospective aspect of a rule as a way of expressing the conclusion that it is permissible. The fact that a rule with some retrospective feature is acceptable does not make it wholly prospective, however. Acquisition of ownership brings with it legal powers like transfer, and changing those powers changes the legal consequences of acquisition.
(16.) For example, in 1890 Congress substantially increased the disability benefits of Union Civil War veterans, giving them pensions for disabilities that were not related to war service. The new benefit was not given by name, but was based on past events, being available to honorably discharged veterans who served more than ninety days in the War of the Rebellion. Act of June 27,1890, ch. 634,26 Slat. L. 182, [section] 2.
(17.) States that raise their drinking age may "grandfather" individuals who have acquired the right to drink but are still below the new drinking age, like someone who recently turned eighteen when that was the minimum age and now faces a minimum of twenty-one. E.g., 64 Del. Laws 546 (increase in drinking age from twenty to twenty-one not applicable to individuals who are twenty on effective date). But someone who turned eighteen shortly after the change was subject to the new more restrictive and retrospective law. As to that person, the law is retrospective: its application depends on past events (the person's date of birth) and it changes the future consequences of those events, so that someone who would have been allowed to drink upon reaching the old drinking age now is not. One might say that the person who had turned eighteen had acquired the right to drink while the person who had not had never done so. That is true, but also demonstrates the attraction of a distinction based, not simply on retrospectivity, but on vested rights. An increase in the drinking age that is not grandfathered operates retrospectively upon both those who lose the permission to drink and those whose permission is only delayed, but only the former will have had that permission and lost it. The tendency to say that only the former have been subject to a retrospective law shows how deeply vested rights and retrospectivity are associated. They are nevertheless conceptually distinct.
(18.) Under federal quarantine law, the Public Health Service may detain individuals in order to prevent the spread of communicable disease, without regard to whether the individual involved became infected before or after the quarantine order was imposed. See 42 U.S.C.A. [section] 264 (West 2016). Quarantines also make the point that legislation may be keyed, not only to prior events, but to prior decisions by individuals, for example the decision to travel to a place that later proves to be a source of infectious disease.
(19.) A classic statement of the standard interpretation of the Ex Post Facto Clauses appears in Justice Chase's seriatim opinion in Calder v. Bull, 3 U.S. 386 (1798). The clause in Article I, Section 10, which applies to the states, requires "that the Legislatures of the several stales, shall not pass laws, after a fact done by a subject, or citizen, which shall have relation to such fact, and shall punish him for having done it." Id. at 390. Justice Chase explained that the clause protects "personal security" from criminal punishment, but not "private rights of either property, or contracts." Id. He went on to draw a distinction between ex post facto laws and the broader category of retrospective laws to which the former belong, id. at 391, and to argue that a law that "takes away, or impairs, rights vested, agreeably to existing laws, is retrospective, and is generally unjust, and may be oppressive," id., but that some retrospective laws, like laws of oblivion and pardon, may be just and for the benefit of the community, id. Chase evidently did not believe that all laws that altered the legal effects of prior events were beyond the legislative power; if he had, he would have thought that acts of oblivion were impossible.
(20.) A classic discussion is Edwin S. Corwin, The Basic Doctrine of American Constitutional Law, 12 Mich. L. Rev. 247 (1914).
(21.) Professors Chapman and McConnell have recently explored in depth the separation of powers thinking that underlay the doctrine of vested rights. See Nathan S. Chapman & Michael W. McConnell, Due Process as Separation of Powers, 121 YALE L.J. 1672 (2012). They do not dwell on the point that the separation of powers principles they discuss were themselves substantive in that they distinguished among legal interests, limiting the legislature only with respect to vested rights and not other legal advantages.
(22.) From its adoption in 1784 to today, the Constitution of New Hampshire has provided that "Retrospective laws are highly injurious, oppressive, and unjust. No such laws, therefore, should be made, either for the decision of civil causes, or the punishment of offenses." N.H. CONST., pt. I, art. XXIII. In the nineteenth century, the Supreme Court of New Hampshire concluded that the provision banned only those retrospective laws that interfered with vested rights. A leading case was Willard v. Harvey, 24 N.H. 344 (1852). In Willard, the defendant relied on a limitations period for the enforcement of judgments that had been adopted after entry of the judgment on which the plaintiff sought to collect. Id. at 351-352. Although the statute changed the legal consequences of the earlier judgment, the court upheld it, explaining that not all retrospective laws fell under the constitutional ban on retrospective laws. "The broadest construction of the constitutional rules which forbid retrospective legislation, would require that all statutes affecting in any way a civil cause, must be so entirely prospective, that no new rule could be applied in the decision of a cause, which did not exist when the right of action accrued. But a construction so broad as this could not be reasonably held, since the effect would be that no change could be made in the courts or course of justice which would affect the actions or causes of action then existing." Id. at 352. The court's solution was the standard distinction between rights, which vest, and remedies, which do not. "The courts, therefore, have everywhere recognized a distinction between statutes affecting rights, and those affecting remedies only. The rights of parties cannot be changed by legislation; but no party has a vested right to any particular remedy." Id. (citation omitted).
Tennessee also has a ban on "retrospective" laws. TENN. CONST., art. I, [section] 20. In the 1820s, the Supreme Court of Tennessee said that "This clause, taken in its common and unrestrained sense, extends to all prior times, persons and transactions, whether civil or criminal, yet, certainly, there are some cases coming within its general scope, to which it does not extend." Townsend v. Townsend, 7 Tenn. 1, 15 (1821).
(23.) Thomas M. Cooley, The Constitutional Limitations Which Rest Upon the Legislative Power of the States of the American Union 358 (1868). For Cooley, the protection against legislative alteration that came with a right being vested was a conclusion, derived from considerations of policy and justice, and not part of the concept of a right. His conception avoided circularity, but only by looking to considerations that distinguished some legal interests from others. Cooley also explained that specificity by itself did not make a legislative act invalid, id. at 355, and that "there are numerous cases which hold that retrospective laws arc not obnoxious to constitutional objection, while in others they have been held to be void," id. at 370. Cooley discussed several categories of permissible retrospective laws, such as changes in remedies, id. at 361, and in limitations periods. Id. at 364.
(24.) Caleb Nelson, Adjudication in the Political Branches, 107 Colum. L. Rev. 559 (2007).
(25.) Id. at 565.
(26.) "Inspired by Lockean political theory, [nineteenth century Americans] distinguished what I will call 'core' private rights (which Lockean tradition associated with the natural rights that individuals would enjoy even in the absence of political society) from mere 'privileges' or 'franchises' (which public authorities had created purely for reasons of public policy and which had no counterpart in the Lockean state of nature)." Id. at 567 (footnote omitted).
(27.) For example, in Usery v. Turner Elkhorn Mining Co., 428 U.S. 1 (1976), the Court upheld a federal statute that required mine operators to compensate miners for injuries that took place before the statute was adopted. The Court explained that "[i]t is by now well established that legislative Acts adjusting the burdens and benefits of economic life come to the Court with a presumption of constitutionality, and that the burden is on one complaining of a due process violation to establish that the legislature has acted in an arbitrary and irrational way." Id. at 15. The Court recognized that retroactivity could pose problems that a prospective rule would not, id. at 17, but found that "the imposition of liability for the effects of disabilities bred in the past is justified as a rational measure to spread the costs of the employees' disabilities to those who have profited from the fruits of their labor the operators and the coal consumers," id. at 18. Demanding such a rational connection is far from absolutely protecting vested rights, or treating retrospective imposition of liability as per se beyond legislative power.
(28.) 514 U.S. 211 (1995).
(29.) 501 U.S. 350 (1991).
(30.) Id. at 364.
(31.) Justice Stevens described the situation in his dissent in Plaut. See Plaut, 501 U.S. at 246-47.
(32.) Id. at 213-14.
(33.) Id. at 213-15.
(34.) The Court reasoned that the judicial power granted by Article III is the authority to render dispositive judgments, and that "by retroactively commanding the federal courts to reopen final judgments, Congress has violated this fundamental principle." Id. at 219.
(36.) Id. at 232-233.
(37.) 54 U.S. 518 (1851).
(38.) Id. at 578. Wheeling is in the part of Virginia that became West Virginia in the 1860s.
(39.) The Post Office Appropriation Act of 1852 declared the Wheeling Bridge to be a lawful structure. Act of August 31, 1852, ch. CXL, [section] 6-7,10 Stal. Ill, 112.
(40.) 59 U.S. 421,436-437 (1855).
(41.) Id. at 429-431.
(42.) In Miller v. French, 530 U.S. 327 (2000), the Court upheld a provision of the Prison Litigation Reform Act that stayed existing injunctions concerning prison conditions while those decrees were reconsidered in light of the new standards for injunctive relief provided by the act. The Court explained that "as Plaut and Wheeling Bridge II instruct, when Congress changes the law underlying a judgment awarding such relief, that relief is no longer enforceable to the extent it is inconsistent with the new law." Id. at 329. The Court recently reaffirmed and applied that principle in Bank Markazi v. Peterson, 136 S. Ct. 1310 (2016). "Congress, our decisions make clear, may amend the law and make the change applicable to pending cases, even when the amendment is outcome determinative." Id. at 1317.
(43.) Justice Scalia emphasized the retroactive character of Section 27A. Plaut v. Spendthrifl Farm, 514 U.S. 211, 227 (1995). That focus on retrospectivity raises the possibility that he meant to distinguish, not between damages judgments and injunctions, but between wholly prospective injunctions and all other remedies. No injunction is wholly prospective, however, in that every injunction rests on determinations about past events, just as damages judgments do. In order to prevail, the plaintiff seeking an injunction must establish facts identifying the plaintiff as the holder of an interest that is eligible for protection through a judicial order. A plaintiff seeking specific performance, for example, must show that the parties made a contract.
(45.) Perhaps Justice Scalia and the Court were led astray by their famous pronouncement that the judicial power is the power to say what the law is. See Marbury v. Madison, 5 U.S. 137,177 (1803). Although Chief Justice Marshall may have meant to speak in an eternal present, with the fall of vested rights doctrine that statement is true only in the momentary, changeable present. Judicial power says what the law is at the time of decision, and sometimes what it was when the relevant events took place. It does not lock legal relations in place, unless it creates vested rights.
(46.) Restatement (Second) of Judgments [section] 24 (1982).
(47.) "But in the days when civil procedure still bore the imprint of the forms of action and the division between law and equity, the courts were prone to associate claims with a single theory of recovery, so that, with respect to one transaction, a plaintiff might have as many claims as there were theories of the substantive law on which he could seek relief against the defendant.... In those earlier days there was also some adherence to a view that associated claims with the assertion of a single primary right as accorded by the substantive law, so that, if it appeared that the defendant had invaded a number of primary rights conceived to be held by the plaintiff, the plaintiff had the same number of claims, even though they all sprang from a unitary occurrence." Id., [section] 24 cmt. a.
(48.) Fed. R. Civ. P. 2 (one form of action).
(49.) Restatement (Second) of Judgments [section] 27 (1982) (resolution of issues litigated and decided by court with jurisdiction has preclusive effect between the parties in later litigation).
(50.) The doctrine of Wheeling Bridge It itself, as opposed to current doctrine, was definitely substantive and not structural in that it distinguished, not between damages and injunctions, but between injunctions based on public rights, and decrees, whether granting damages or injunctions, based on private rights. The Court agreed that in general "the act of [C]ongress cannot have the effect and operation to annul the judgment of the court already rendered, or the rights determined thereby in favor of the plaintiff. This, as a general proposition, is certainly not to be denied, especially as it respects adjudication upon the private rights of parties. When they have passed into judgment the right becomes absolute, and it is the duty of the court to enforce it." Pennsylvania v. Wheeling & Belmont Bridge Company (Wheeling Bridge II) 59 U.S. 421,431 (1855). Public rights, like the right of navigation, were different. "The case before us, however, is distinguishable from this class of cases, so far as it respects that portion of the decree directing the abatement of the bridge. Its interference with the free navigation of the river constituted an obstruction of a public right secured by acts of [C]ongress." Id.
(51.) Restatement (Second) of Judgments [section] 27 (1982).
(52.) The Court has not had occasion to discuss the status of possessory decrees under Plaut, and is not likely to have occasion to do so. A possessory remedy like replevin rests on the conclusive resolution of questions of ownership (and a judgment for the defendant in a possessory proceeding may also rest on the court's resolution of that question, though it need not). Ownership of a particular asset is so much like ownership of a specified sum, and so different from a pure rule of conduct that possessory decrees very likely would be treated like damages judgments under Plaut.
(53.) Corwin, supra note 20, at 247 (the two fundamental antebellum constitutional doctrines were the complementary principles of vested rights and the police power).
(54.) Rule 60(b) of the Federal Rules of Civil Procedure contemplates changes in injunctions but does not mention damages or other judgments. That suggests another possible ground of distinction: injunctions may be changed because every injunction has built into it the possibility of change, so alteration of an injunction does not eliminate any right in the injunction created. Justice Scalia did not indicate that a similar provision with respect to all federal court judgments would be constitutional, and he very likely would regard one as an impermissible intrusion into the judicial power; he did not rely on Rule 60(b) in distinguishing cases like Wheeling Bridge II. A rule about alteration of judgment that turned on the law of judgments as it stood when a case was decided would be another form of vested rights doctrine, one that would enable legislatures to create interests that they could not alter in the future by deciding whether to provide for alteration in the law of judgments.
(55.) Answering that question requires resolving an interpretive difficulty that is itself important. Many, perhaps most, legally sophisticated people at the time of the Framing believed that some legal interests were immune from legislative alteration. Two lines of reasoning were available to support that conclusion. One was that the concept of legislative power included a limitation in favor of vested rights. The other was that vested rights were immune from legislative alteration because of a separate principle that operated as a limit on legislative power, as the First Amendment limits Congress' taxing power. Both lines of reasoning produce the same result, and so the fact that the result was widely accepted cannot itself support a choice between them. The problem of separating the meaning of constitutional concepts from non-constitutional legal principles that were generally accepted at one time but are not accepted now is a recurring one. Any resolution I could offer to the particular example of that recurring problem that involves separation of powers and vested rights would be at best provisional. It is not necessary to choose between the two possible understandings of the distinction between legislative and judicial power in order to see that there are two.
(56.) Plaut v. Spendthrift Farm, 514 U.S. 211, 219 (1995).
(57.) Federal j udges are appointed by the President with the advice and consent of the Senate, U.S. CONST., art. II, [section] 2, and serve on good behavior, id.', id. at art. Ill, [section] 1, subject to impeachment by the House of Representatives and removal by a two-thirds vote of the Senate, id. at art. I, [section] 2 (House shall have sole power of impeachment); id. at art. I, [section] 3 (Senate shall try all impeachments, with two-thirds needed to convict).
(58.) See Myers v. United States, 272 U.S. 52, 117 (1926) (President must have power to remove subordinate executive officers because he holds the power to execute the laws and has the duty to do so).
|Printer friendly Cite/link Email Feedback|
|Author:||Harrison, John (American law professor)|
|Date:||Jun 22, 2016|
|Previous Article:||The meming of substantive due process.|
|Next Article:||Madison's Hand: Revisiting the Constitutional Convention.|