The Mandate Rule
By
By
Adam Crews[1]*
The “mandate rule” is a staple of federal practice. When a federal appellate court decides an appeal, a mandate issues. That mandate transfers jurisdiction back to the inferior court and constrains the inferior court on remand. As is often recited, an inferior court must comply with and carry out the mandate’s letter and spirit in further proceedings, and failure to do so is reversible on appeal or correctable by mandamus.
The mandate rule has become a source of conflict, confusion, and error in the federal circuit courts. Some courts treat the rule as jurisdictional, while others treat it as a rule of sound practice. Some courts have extended the rule from its judicial origins to administrative law, thereby claiming the power to direct agency proceedings on remand. Some courts substitute their own judgments for the district court’s, defeating the mandate’s very purpose. And some courts view enforcement of a mandate as an exception to the ordinary rules for mandamus, while others insist that the traditional requirements for the writ still apply.
This Article argues that these issues result from courts’ failures to understand and to apply the mandate rule by reference to its statutory origins. Federal circuit courts routinely recite and apply the rule as if it is inherent to the judicial hierarchy or the powers of appellate courts generally. That is mistaken. The mandate rule is, and always has been, a statutory rule, and many of its nuances can be explained by straightforward reference to the applicable but oft-overlooked statutes. By losing sight of the rule’s statutory origins, courts have created error and confusion by applying a doctrine without an anchor in their actual source of authority.
At the end of a federal appeal, there is a mandate.[2] This is more than just an abstraction (i.e., the “rule” or “holding” from the case) or a way to describe the court’s decretal language (e.g., “affirmed” or “reversed and remanded”).[3] Instead, the mandate is a legally operative set of documents. At one time, the mandate was a formal, stand-alone document that issued in the name of the President of the United States and commanded a lower court to take action.[4] Today, the typical appellate mandate “consists of a certified copy of the judgment, a copy of the court’s opinion, if any, and any direction about costs.”[5] Because these documents seem innocuous, courts sometimes refer dismissively to the mandate’s issuance as a mere ministerial act.[6] But the mandate is critical. The mandate’s issuance officially returns jurisdiction over a case to the inferior court.[7] And, under the so-called “mandate rule,” it controls subsequent proceedings in the case and empowers the superior court to coerce compliance.[8]
The mandate rule is a staple of federal practice.[9] From “its earliest days,” the Supreme Court has “consistently held that an inferior court has no power or authority to deviate from the mandate issued by an appellate court.”[10] In its most basic form, the mandate rule is the “doctrine that, after an appellate court has remanded a case to a lower court, the lower court must follow the decision that the appellate court has made in the case, unless new evidence or an intervening change in the law dictates a different result.”[11] Stated more simply, the mandate rule provides that an inferior court is “bound to honor” the superior court’s mandate.[12] This doctrine of obedience, some courts have said, is “necessary to the operation of a hierarchical judicial system.”[13]
The mandate rule really embodies several distinct rules about the effect of a superior court’s decree and the remedies available if an inferior court fails to implement that decree faithfully. The majority view is that the mandate rule is subsidiary to (but a particularly strong form of) the law of the case doctrine,[14] which holds that a decision on a rule of law should generally govern in subsequent stages of the same case.[15] But the doctrine has significant nuance. For instance, the mandate “is controlling as to matters within its compass” but not “as to other issues.”[16] Thus, “district courts are not free to decide issues on remand that were previously decided either expressly or by necessary implication on appeal,”[17] but issues not decided by the appellate court remain open on remand.[18] The rule also generally prohibits “reopening” any issue that “was ripe for review” on an initial appeal “but was nonetheless foregone.”[19] And it is well settled that a superior court can enforce its mandate “either upon a new appeal . . . or by a writ of mandamus to execute the mandate.”[20]
For such an important facet of federal practice, the mandate rule is under-explored and under-theorized. Although the Supreme Court has routinely recited the rule’s basic tenets,[21] it has never referred to the rule by name,[22] and only rarely speaks of the broader “law of the case” doctrine of which the mandate rule is generally considered a part.[23] For their parts, the federal circuit courts often recite by rote that their mandates control; that district courts must scrupulously enforce mandates; and that district courts must comply with both the letter and spirit of a mandate.[24] From these general principles, appellate courts go about the business of assessing—and, where appropriate, enforcing—compliance with their prior decrees.[25]
In applying the mandate rule, courts rarely pause to reflect on the doctrine’s origins.[26] Although this point may seem academic, it ought to matter to courts, practitioners, and litigants. Doctrines do not come from nowhere; they develop to solve particular problems.[27] When courts apply a doctrine reflexively to new areas without reflecting on its history and purpose, the doctrine’s development can become decoupled from the problem it was meant to solve.[28] The mandate rule’s reflexive and unreflective application in the federal circuit courts threatens to commit this error when several important questions are left unexamined and unanswered.
First, what is the source of the superior court’s power? A mandate might imply an obligation on the inferior court for any number of reasons. It could be that a statute or rule determines the effect, i.e., that the mandate rule is a legal directive rooted in some textual source of law. Or, it could be a necessary corollary to the appellate court’s superiority in the federal judicial hierarchy. But even then, the power can derive from any of several attributes of superiority: the very nature of the hierarchical judicial system, whether under Article III or statute; particular grants of appellate jurisdiction, with attendant inherent or statutory power to defend judgments entered under those grants; or particular grants of remedial power with respect to orders on review. These distinctions matter for practical reasons; they determine where we should look to know whether courts are applying the mandate rule correctly, and they inform the extent to which the mandate rule is subject to revision via the legislative process. For instance, if the mandate rule has a constitutional dimension, that may circumscribe Congress’s power to alter or modify the doctrine.
Second, what is the nature of the lower court’s obligation to comply? A lower court may need to comply with a mandate for one of two reasons: because the superior court’s decree is a lawful order calling for obedience, or because the lower court is without power to do anything but obey. Stated otherwise, the mandate rule could govern the inferior court’s resolution of a dispute, or it could be a limit on the inferior court’s jurisdiction to resolve certain issues. The federal circuit courts are split on the issue.[29] But this, too, matters for a practical reason: in federal court, a jurisdictional error with respect to subject matter is not forfeitable or waivable, whereas other errors are.[30] And what about state courts—can a Supreme Court mandate really withdraw jurisdiction from the courts of a separate sovereign?
Third, what can a superior federal court lawfully require through a mandate? The answer to the first two questions might inform a mandate’s lawful scope. If the mandate rule has constitutional dimensions, superior courts may have more leeway to shape their directives to inferior courts. But if the rule has a statutory source, there may be a narrower set of lawful directives that can issue. And what about when an appellate court isn’t even looking at a lower court’s judgment, but at an administrative agency’s final order? What can a federal court mandate to a body in another branch altogether, and why?
This Article aims to address these questions and, in doing so, to offer some clarity for the doctrine’s murkier points. Part II is descriptive; it explains that the mandate rule emerged from an interpretation of statute and then traces the rule’s statutory source from the First Judiciary Act to its modern successor, 28 U.S.C. § 2106, to provide context for discussing the doctrine’s current state. Part III argues that the mandate rule is statutory alone, and that any constitutional or sub-constitutional dimensions are beside the point. Part IV identifies various ways in which courts have departed from the best reading of the applicable statutes, thereby introducing conflict, confusion, or error to the doctrine. Part IV also explains how a return to the statutory text resolves these problems. Thus, this Article aims to address the three big questions outlined above: Where does the doctrine come from? What obliges inferior courts to comply and empowers superior courts to enforce compliance? And what exactly can a mandate lawfully require?
To assess the mandate rule’s modern applications, one must first know the doctrine’s source—that is, what source of law supplies the power to issue a mandate, the obligation to obey it, and the remedies to enforce it? This Part has two principal aims. First, it explains that the mandate rule is statutory and emerged as a straightforward application of the First Judiciary Act.[31] Second, it traces the development of the pertinent federal statutes over time to explain which statutes are the modern successors to the relevant provisions of the First Judiciary Act that generated the doctrine in the first place.
Federal appellate courts routinely cite the mandate rule’s precepts by reference to prior cases, which paints a picture of a doctrine inherited from judicial practice alone.[32] To be sure, a key aspect of our federal judicial system, as inherited from the Anglo legal tradition, is that prior decisions are evidence—perhaps even the best evidence—of the law.[33] But that picture is incomplete; although the modern federal appellate courts tend to rest on precedent, the precedents rest on a statute.[34]
To see this statutory grounding, a good starting point is In re Sanford Fork & Tool Co.,[35] which the Ninth Circuit has described as “the seminal case” on the mandate rule.[36] Sanford Fork states many of the mandate rule’s best known features:
When a case has been once decided by this court on appeal, and remanded to the circuit court, whatever was before this court, and disposed of by its decree, is considered as finally settled. The circuit court is bound by the decree as the law of the case, and must carry it into execution according to the mandate. That court cannot vary it, or examine it for any other purpose than execution; or give any other or further relief; or review it, even for apparent error, upon any matter decided on appeal; or intermeddle with it, further than to settle so much as has been remanded.[37]
Despite its age, the federal courts of appeals have relied on Sanford Fork as authority for the mandate rule’s contents well into recent years.[38]
Sanford Fork was before the Supreme Court on a petition for mandamus to direct entry of a judgment in accordance with an earlier mandate.[39] In reciting the mandate rule’s components, the Supreme Court—like modern federal appellate courts still tend to do—rested on its own precedents.[40] In particular, the Court relied on Sibbald v. United States for the proposition that a lower court “cannot vary” a mandate “or examine it for any other purpose than execution; or give any other or further relief; or review it, even for apparent error, upon any matter decided on appeal; or intermeddle with it, further than to settle so much as has been remanded.”[41]
For its part, Sibbald derived its rule from the First Judiciary Act.[42] Sibbald began its analysis by discussing the First Judiciary Act’s express reference, at section 24, to the Court’s “special mandate.”[43] And for the Court’s power to enforce its “special mandate,” Sibbald cited the general writ power granted to the Court under section 14 of the same statute.[44] But Sibbald was also not writing on a blank slate; in discussing its own appellate power and the conclusive effect of its own prior decisions, the Court relied on the seminal case Martin v. Hunter’s Lessee.[45] And Martin, too, rested on the First Judiciary Act, and in particular on the Court’s grant of power to review certain state court judgments and the absence of any provision that authorized the Court to revise its own judgments.[46]
The quick lesson is this: from the early days, the Supreme Court’s mandate rule evolved from the Court’s application of the First Judiciary Act.[47] That Act authorized the “special mandate”;[48] directed how the Court could review judgments and direct execution of its decrees;[49] made the Court’s judgments conclusive because no provision authorized their further review;[50] and supplied the writ power necessary to enforce the mandate.[51] This suggests that the Court never viewed the mandate rule as inherent or inevitable. The mandate rule was instead a consequence of how the First Congress designed the federal judicial system through statute. The mandate rule doctrine—from the conclusiveness of the appellate decision to the authority to direct inferior court compliance—depended on Congress.
While the mandate rule persists as handed down by the Supreme Court centuries ago,[52] the First Judiciary Act does not.[53] If the rule derives from statute, then where does it come from now? To answer that question, this Section traces the evolution of the pertinent statutes over time, with attention to how and why they changed.
Before jumping into the statutory history, it is helpful to set the stage by considering the situation when the First Congress convened. “Unlike the state constitutions, which often discussed the structure of courts in some detail, the federal Constitution was quite laconic on the subject.”[54] The Constitution merely provided for “one supreme Court”[55] exercising narrow “original Jurisdiction,”[56] with the option for Congress to “ordain and establish” other “inferior Courts . . . from time to time.”[57] Thus, “Congress could have dispensed with any lower federal courts,” leaving state courts with “original jurisdiction over all but a few federal issues.”[58] Moreover, the Constitution gave the Supreme Court “rather few constitutional tools to keep its underlings in line,” and in particular it afforded “little inherent power to punish insubordinate deputies or reward loyal ones.”[59] To the extent lower federal courts would exist, it was up to Congress to define much of the relationship among the courts.[60]
Given this constitutional backdrop, Congress faced several decisions about the roles of different courts when creating a hierarchical judicial system. Some of those decisions persist to today: although the exercise of federal judicial power is vested in all Article III courts that Congress establishes,[61] various statutes and rules divvy up particular judicial functions between the different courts.[62] There are certain things that generally only district courts do, like find facts.[63] But district courts also wield the power to enter final judgments,[64] which officially close any given case and actually fix the rights between the parties.[65] Thus, although a federal appellate court may correct an error of law on appeal, “the appellate courts’ powers necessarily operate in conjunction with the district court, where actual closure of cases occurs.”[66] In short, congressional design often makes appellate courts dependent on inferior courts to carry out their rulings to final judgment.[67]
There are sensible institutional reasons for this arrangement. Judgments are not self-executing; an award of monetary damages does not pay itself, and orders to take specific action do not carry themselves out. Sometimes people avoid paying, hide their money, or simply refuse to obey an order.[68] That’s why the federal rules provide for execution on judgments, including obtaining discovery in aid of execution and authorizing contempt for disobedient parties.[69] Purely as a matter of policy, there are good reasons to think that these functions are best carried out by district courts, which have certain institutional advantages when compared to appellate courts—for example, the ability to take testimony[70] and to act unilaterally, without the need to convene a panel of judicial officers.[71] Because district courts are responsible for entering and enforcing final judgments, the hierarchical judicial system needs something like a mandate rule; otherwise, litigants might not be able to obtain and enforce judgments in line with an appellate court’s rulings.[72]
This division of judicial authority was not inevitable,[73] but it is also not new. When Congress enacted the First Judiciary Act in 1789, it opted for a tiered federal judicial system.[74] The Act established district and circuit courts inferior to the Supreme Court[75] and specified that federal judgments would reach the Supreme Court on writs of error.[76] Once resolved, however, section 24 provided that “the Supreme Court shall not issue execution in causes that are removed before them by writs of error, but shall send a special mandate to the circuit court to award execution thereupon.”[77] In other words, the Supreme Court was not responsible for entering and enforcing its own final judgments fixing the rights of the parties to a case—that was left to the inferior federal courts, acting in accord with the Supreme Court’s ruling on appeal.[78] But the Court was not left powerless to enforce its decisions; section 14 of the Act authorized writs like mandamus to compel compliance.[79]
This division of authority made sense for various practical reasons. Under the First Judiciary Act, the Supreme Court was not a permanent fixture at the nation’s capital.[80] The Court sat “annually at the seat of government” for “two sessions, the one commencing the first Monday of February, and the other the first Monday of August.”[81] The Justices also had dual assignments on lower courts, so the Act required the Justices to ride circuit, i.e., “to climb into carriages and travel to their circuits,” initially “twice a year.”[82] A multi-member court with a four-member quorum[83] and scattered officers was hardly ideal for issuing execution on judgments.
Thus, the First Judiciary Act established the Supreme Court’s “mandate” as a means of controlling other courts in the federal system.[84] As the seminal Sibbald case described the process, “[i]f the special mandate directed by the 24th section” of the Act “is not obeyed,”then “the 14th section of the judiciary act” authorized the Supreme Court “to issue any writs which are necessary for the exercise of [its] . . . jurisdiction[], and agreeable to the principles and usages of law,” including “a mandamus or other appropriate writ” issued to the inferior court.[85] Or, as an alternative to enforcement by writ, the Supreme Court could reverse courts that did not execute the mandate “according to its true intent and meaning.”[86]
Things were a bit different for the relationship between the Supreme Court and state courts. To the extent state court judgments were reviewable on a writ of error, the appeal was generally determined “in the same manner and under the same regulations” and with “the same effect, as if the judgment or decree complained of had been rendered or passed in a circuit court.”[87] But there was a caveat: although “proceeding upon the reversal” would ordinarily be the same for state and federal courts, “the Supreme Court, instead of remanding the cause for a final decision” to a state court, had discretion to “proceed to a final decision of the same, and award execution,” but only “if the cause shall have been once remanded before.”[88] In other words, state courts had one opportunity to fall in line and enter a compliant judgment after the Supreme Court reversed them on appeal. If they did not, the Court could grant a second writ of error and, on the second reversal, enter its own judgment and order execution on it.
This different treatment of federal versus state courts may have reflected a concern that the Supreme Court did not have authority to compel state court compliance through writs like mandamus.[89] In fact, an early issue that the Supreme Court confronted was where to send a mandate when a state trial court entered a judgment; the state’s court of last resort reversed; and the Supreme Court then reversed the highest state court.[90] In the end, the Supreme Court concluded that its mandate should issue to the state trial court, which had already entered a judgment that, in the Supreme Court’s view, was properly undisturbed and ripe for execution.[91] In the Court’s words, this was because the state court of last resort’s judgment, upon “being reversed,” was “a mere nullity.”[92] Beneath the surface, there may have been concern that it was better to send the mandate to the state court that already agreed with the Court’s judgment, rather than hope that the state court of last resort would acknowledge its error and comply with the mandate.
Concern that a reversed state court might not acquiesce to the Court’s decision eventually came to fruition, as Martin v. Hunter’s Lessee shows.[93] That case began in Virginia state district court as an action for ejectment,[94] which the Supreme Court of Appeals of Virginia reversed.[95] On writ of error, the Supreme Court reversed the Virginia appellate court and affirmed the district court’s judgment.[96] The Court’s mandate issued to the Supreme Court of Appeals of Virginia, with the “cause . . . remanded to the said Court of Appeals . . . with instructions to enter judgment for” Martin.[97] On remand, however, the Virginia appellate court famously held “that the appellate power of the Supreme Court of the United States, does not extend to [that state] court”; that the Supreme Court’s writ of error “was improvidently allowed”; and “that obedience to its mandate be declined.”[98] On a second writ of error, the Supreme Court rejected that view and again reversed the state appellate court.[99] But in making that decision, the Court declined “to give any opinion upon the question, whether this [C]ourt ha[s] authority to issue a writ of mandamus to the court of appeals to enforce the former judgments.”[100]
In a separate opinion, however, Justice Johnson examined that question.[101] As Justice Johnson saw it, “the framers of [the First Judiciary Act] plainly foresaw that the state courts might refuse” to obey the Supreme Court’s mandate.[102] But those same framers were “not . . . willing to leave ground for the implication, that compulsory process”—e.g., mandamus—“must be resorted to.”[103] Indeed, Justice Johnson read section 13 of the First Judiciary Act to “restrict [the Supreme Court] in issuing the writ of mandamus, so as to confine it expressly to those courts which are constituted by the United States.”[104] This specific rule about mandamus, in Justice Johnson’s view, trumped section 14’s broader grant of authority to issue “all . . . writs . . . which may be necessary for the exercise of their respective jurisdictions, and agreeable to the principles and usages of law.”[105] Because the Supreme Court lacked mandamus power over the state courts, Justice Johnson reasoned, Congress allowed the Supreme Court “to execute its own judgment,” but this power was discretionary because “it could only be necessary in case of the refusal of the state courts.”[106] And, importantly, this discretion was allowed only in the event of reversal, because in the event of affirmance there would be no tension between the state courts and the Supreme Court requiring some means to vindicate the federal ruling.[107]
So far, this review has covered the mandate rule’s vertical constraint on inferior courts. But the mandate also had a horizontal constraint on the Supreme Court itself.[108] Discussing the mandate’s conclusive character, Justice Story remarked: “Whatever had been formerly before the Court, and was disposed of by its decree, was considered as finally disposed of[.]”[109] This, too, followed from statute. In Martin v. Hunter’s Lessee, for example, the Court rejected a challenge to its jurisdiction to enter an earlier decree in the same case because “in ordinary cases a second writ of error has never been supposed to draw in question the propriety of the first judgment,” and “[a] final judgment of this court is supposed to be conclusive upon the rights it decides, and no statute has provided any process by which this court can revise its own judgments.”[110] Just a few years later, the Court reaffirmed the basic principle.[111] When a party moved for rehearing after the mandate issued, the Court denied the motion because “it was too late to grant a rehearing in a cause after it had been remitted to the Court below, to carry into effect the decree of [the] Court, according to its mandate.”[112] And on a second writ of error, the Court had no authority to conduct “an inquiry into the merits of the original decree.”[113] This rule of conclusiveness—even as to the Supreme Court itself—came to be called “the law of the case.”[114] And embedded in that doctrine was the now well-settled “mandate rule.”[115]
The prior Subsection established that the mandate rule’s initial contours came from the Supreme Court’s interpretations and applications of the First Judiciary Act. So how did we get to where we are today, with the mandate rule alive and well principally in the federal circuit courts? After all, the First Judiciary Act is no longer in effect, and the mandate rule initially derived from provisions of that Act dealing specifically with the Supreme Court’s power.[116] In short, why do today’s federal circuit courts enforce their own mandates by reciting language from the Supreme Court that was rooted in statutory text that is no longer in force and that applied only to the Supreme Court?
To explain the journey from 1789 to today, the first stop is in 1872. That year, Congress codified the powers of appellate courts (whether the Supreme Court or the initial circuit courts) when reviewing judgments on writs of error.[117] In relevant part, Congress provided that the “appellate court may affirm, modify, or reverse the judgment, decree, or order brought before it for review, or may direct such judgment, decree, or order to be rendered, or such further proceedings to be had by the inferior court as the justice of the case may require.”[118] This formula of words would come to define the federal appellate power, even today.[119]
Shortly after the 1872 statute, Congress undertook a comprehensive recodification of federal law, a project that had been explored since 1866.[120] The project came to fruition in 1874, when Congress authorized publication of the Revised Statutes.[121] As codified, the Revised Statutes repealed any statute pre-dating December 1, 1873, to the extent the provisions were contained in the new Revised Statutes.[122] The new codification was “in force” instead.[123] As relevant here, the Revised Statutes provided, at section 701, that the “Supreme Court may affirm, modify, or reverse” federal judgments, “or may direct such judgment, decree, or order to be rendered, or such further proceedings to be had by the inferior court.”[124] The provision further restated the rule from the First Judiciary Act that the Supreme Court not issue execution, but instead issue “a special mandate” to the inferior court.[125]
Several years later, Congress passed the Judiciary Act of 1891,[126] often called the Evarts Act.[127] This Act created the federal circuit courts of appeals that are familiar today, i.e., intermediate appellate courts with jurisdiction to review district court judgments.[128] As part of this reform, Congress provided that the Supreme Court and the federal circuit courts could “remand[] . . . for further proceedings to be there taken in pursuance of” the appellate court’s “determination” of the case.[129] The Act further provided that “all provisions of law now in force regulating the methods and system of review, through appeals and writs of error,” would regulate appeals in the new federal circuit courts.[130]
The Evarts Act came to be understood as extending to the federal circuit courts the Supreme Court’s powers to affirm, modify, or reverse judgments and to direct further proceedings in the inferior court. This seems to have begun with a privately published compilation of federal statutes from 1913.[131] A statutory note in that compilation stated that the Evarts Act “superseded to a great extent” section 701 of the Revised Statutes, which governed the Supreme Court’s determination of appeals.[132] To the extent section 701 “remained in force, it was made applicable to appeals and writs of error provided for . . . in respect of the circuit courts of appeals,” such that the provision relating to the “special mandate” should be “regarded as applicable” to the circuit courts.[133] Courts agreed. Citing these very provisions, the Seventh Circuit held in 1915 that it was “vested with power to modify, as well as to affirm or reverse, any judgment of the District Court.”[134] Other courts soon followed suit.[135]
With the transition from the Revised Statutes to the U.S. Code, the various provisions supporting the mandate rule found themselves scattered throughout title 28.[136] Beginning in the 1934 edition, § 344 dealt with the Supreme Court’s power to “reverse, modify, or affirm” certain state court judgments and to “award execution or remand”;[137] § 876 dealt with the Supreme Court’s powers to “affirm, modify, or reverse” and to “direct such judgment, decree, or order to be rendered, or such further proceedings to be had” with respect to district court judgments in prize cases;[138] and § 877 dealt with remands to district courts.[139] In 1940, the codified § 876 was broadened to state that the “Supreme Court may affirm, modify, or reverse any judgment, decree, or order of a district court lawfully brought before it for review, or may direct such judgment, decree, or order to be rendered, or such further proceedings to be had by the inferior court,” but shall “send a special mandate” rather than “issue execution.”[140]
These scattered provisions converged in 1948, when Congress revised and codified title 28 as positive law.[141] In a new section, 28 U.S.C. § 2106, captioned “Determination,” Congress provided:
The Supreme Court or any other court of appellate jurisdiction may affirm, modify, vacate, set aside or reverse any judgment, decree, or order of a court lawfully brought before it for review, and may remand the cause and direct the entry of such appropriate judgment, decree, or order, or require such further proceedings to be had as may be just under the circumstances.[142]
This new § 2106 made several changes to the pre-existing law:[143] (1) it consolidated pertinent parts of §§ 344, 876, and 877 of the 1940 version of the U.S. Code; (2) on the authority of circuit court precedent,[144] it specified that “any . . . court of appellate jurisdiction” shared in the power to “affirm, modify, vacate, or set aside” a judgment on review; (3) it broadened the power to “remand” to “permit a remand by the Supreme Court to a court of appeals”;[145] and (4) it dropped entirely the key formula of words, dating back to the First Judiciary Act, providing that the Supreme Court should not issue execution but instead send a special mandate to the inferior court to award execution.[146]
What happened to the mandate reference? In explaining that omission from the 1948 codification, the legislative history cites Rule 34 of the Supreme Court’s revised rules, as well as the newly codified 28 U.S.C. § 1651.[147] Rule 34 dealt with mandates; it provided that mandates would generally issue “as of course after the expiration of twenty-five days from the day judgment is entered,” but not “upon the denial of a petition for writ of certiorari.”[148] Section 1651 re-codified the All Writs Act,[149] which provides that all federal courts “may issue all writs necessary or appropriate in aid of their respective jurisdictions and agreeable to the usages and principles of law.”[150] It therefore appears that the mandate reference dropped from statute because it was provided for by rule, and the general writ power conferred on the federal appellate courts all the power they needed to enforce their judgments.[151]
The new § 2106 sweeps broadly. Several courts have cited this most recent statement of federal appellate determination power as the source of authority for, among other things, staying a lower court’s mandate following an initial remand;[152] reassigning cases on remand;[153] changing a district court’s grant of summary judgment into a dismissal for lack of jurisdiction;[154] directing proceedings to occur “without undue delay”;[155] and granting, vacating, and remanding a judgment in light of a confession of error.[156] Simply put, courts view § 2106 as an almost catch-all source of power for directing further proceedings in inferior federal courts.[157]
But § 2106 is also the successor to the First Judiciary Act’s grant of power to issue a “mandate” to inferior courts to enforce the Supreme Court’s decisions on appeal.[158] Now that power is even more explicit, with each federal appellate court empowered to “require such further proceedings to be had”—including no further proceedings at all, if the record is sufficient to “direct the entry of [an] appropriate judgment, decree, or order.”[159] So, even though § 2106 no longer mentions the mandate by name, the provision still supplies the same basic authority to appellate courts that the mandate rule has long embodied.
Section 2106 authorizes appellate courts to direct and control many aspects of litigation on remand.[160] But that power, by its terms, is limited to review of an “order of a court lawfully brought before it for review.”[161] In the early twentieth century, as what would become § 2106 was taking shape, Congress was also enacting new statutes to deal specifically with federal appellate review of orders from other bodies—the emerging agencies that would make up the modern administrative state.[162] Although a comprehensive review of these statutes is unnecessary, it is helpful to have a sense for how Congress regulated judicial determination in cases of agency review as compared to cases originating in the courts.
Shortly after the turn of the twentieth century, what has come to be called the “appellate review model of the relationship between reviewing courts and agencies” emerged from a back-and-forth between Congress and the Supreme Court.[163] The basic issue was how to balance judicial review of agency action against maintaining appropriate separation of the judicial and administration roles.[164] The end result was a system in which judicial review of agency actions looked a lot like—but not entirely identical to—the relationship between appellate and trial courts in civil litigation.[165]
Consider the Radio Act of 1927 as an example.[166] That Act provided for judicial review of radio station licensing decisions in an Article I federal appellate court, which would “hear, review, and determine the appeal” on the agency record and any additional evidence admitted, with the power to “alter or revise the decision appealed from and enter such judgment as to it may seem just.”[167] This was viewed as imbuing the court with “administrative rather than judicial” power; the appellate court was “a superior and revising agency.”[168] A few years later, after the Supreme Court held that it could not hear appeals from an Article I court exercising these non-judicial functions, Congress amended this review provision.[169] The amendment restrained the scope of judicial review by limiting the appellate court’s review to questions of law and sufficiency of evidence.[170] The revised statute also made the court-agency relationship look a bit more like the relationship between appellate and district courts; the statute expressly authorized the appellate court to “remand the case” to the agency “to carry out the judgment of the court.”[171]
This trend of review in the circuit courts of appeals, limited to questions of law, continued.[172] For several agencies, including the Federal Communications Commission as successor to the Federal Radio Commission, the trend culminated in the Administrative Orders Review Act,[173] commonly referred to as the “Hobbs Act.”[174] That statute, which governs judicial review of several agencies’ final orders, provides that the reviewing court may only “enjoin, set aside, suspend (in whole or in part), or . . . determine the validity of” the agency action.[175] Thus, the Hobbs Act specifies the determinations a circuit court of appeals may make regarding an agency order on review, much like § 2106 specifies the determinations a circuit court of appeals may make regarding a judicial order on review.[176] But the two are not coextensive; by contrast to review of judicial orders, the Hobbs Act (for example) does not specify a power to direct procedures on remand to the agency.[177]
This Part has aimed to demonstrate two key points. First, the mandate rule evolved in the Supreme Court by reference to specific statutory provisions governing the Court’s review of judicial orders.[178] In the First Judiciary Act, the pertinent provisions dealt with how the Supreme Court would carry out its judgments—generally, via a special mandate to the inferior court.[179] Over time, Congress specified the types of “determination[s]”[180] that the Supreme Court (and later, federal circuit courts) could make when reviewing judgments.[181] Ultimately, Congress removed the long-standing direction about sending a “special mandate” because (1) it was unnecessary in light of the Court’s own rule for sending mandates and (2) the Court could enforce its “determination” in the judgment via mandamus regardless.[182]
Second, even though the statutory rule for appellate determination looks much different today when compared to the First Judiciary Act,[183] the mandate rule has not changed. In a way, this makes sense. Although the modern statute is more specific about what an appellate court can determine and direct on remand, those specific provisions are broad and afford substantial power to the appellate court.[184] That power readily encompasses the mandate rule’s core historical attributes. But the statute by its terms does not cover the wide world of orders that the circuit courts of appeals might review today, most notably orders by various federal agencies.[185]
Part II showed, as a descriptive matter, that the mandate rule is historically a statutory doctrine. But to assess whether the doctrine accords with its source of authority, it is worth considering whether other sources of power complement or supplement that statutory grant. If so, it could be those other sources that independently support all or part of the modern mandate rule.
This Part addresses the possibility of two other sources for the mandate rule: (1) a constitutional source premised in Article III’s judicial hierarchy and (2) a sub-constitutional source premised on the inherent power of courts. Ultimately, this Part argues that neither source can independently justify the mandate rule, thereby rendering the doctrine entirely statutory.
A theme in some modern appellate decisions applying the mandate rule is that the doctrine is “necessary to the operation of a hierarchical judicial system.”[186] As Judge Sutton has written: “The rule springs from the hierarchical structure of our judicial system and leaves no room for discretion,” such that inferior courts are “duty bound to follow the mandate of the superior court” on remand.[187]
This hierarchy rationale might imply a constitutional source for the mandate rule.[188] This is textually plausible on its face. The Constitution tells us that, in the federal system, some courts are necessarily “inferior” to others;[189] that even among the courts “inferior” to the Supreme Court, Congress can “ordain and establish” a system of its own devising, including one with multiple tiers;[190] and that even state courts can find themselves subject to the Supreme Court’s “appellate Jurisdiction.”[191] From this constitutional design, one might infer a “duty” to follow the superior court’s mandate.[192]
Despite the intuitive appeal of inferences from constitutional design, a constitutional basis for the mandate rule does not hold up well to scrutiny. As an initial matter, there are sound reasons not to overread the Constitution’s distinction between a “supreme” and “inferior” court as necessarily mandating a judicial hierarchy. As Justice (but then-Professor) Barrett explained in a similar context, the Supreme Court’s more general supervisory power, “the terms ‘supreme’ and ‘inferior’ do little, standing alone,” to address the Supreme Court’s power to supervise other courts.[193] This is because relative rank does not necessarily imply subordination.[194] In fact, some evidence suggests that Founding-era usage may have understood “supreme” and “inferior” to refer to different scopes of territorial jurisdiction under which a “supreme” court is nationwide while an “inferior” court is more confined, but not necessarily subordinate.[195] Article III’s structure does little to resolve this potential textual ambiguity,[196] in particular because the Supreme Court’s appellate jurisdiction is subject to qualification and revision via the Exceptions and Regulations Clause.[197] And history, too, reveals no well-settled English, colonial, or early American practice treating supervision of inferior court procedures as a necessary feature of a hierarchical court system.[198] Indeed, at common law, the customary practice was that an appellate court had only the power to affirm or reverse a judgment on appeal—not to modify the judgment or direct a proper judgment on remand.[199] It would be surprising if Article III, of its own force, upset these settled practices.
To be sure, the mandate rule is a bit of a different beast from a general supervisory power. Whereas the Supreme Court has invoked its general supervisory power to impose several rules of general practice or procedure,[200] the mandate rule is really about effectuating the appellate court’s judgments in specific cases. In other words, the mandate rule is not so much about the process to decide all cases as it is a rule to enforce what has been decided in particular cases. But the mandate rule is, ultimately, a rule about what a lower court must do (or refrain from doing) on remand. Thus, the lessons about implying a procedural supervisory power from the Constitution are still apt: Article III’s plain text and structure do not necessarily imply a federal judicial system built around the subordination of some courts to others,[201] and the Constitution left Congress free to design a non-hierarchical federal system.[202] So, to the extent that the mandate rule is an inference from judicial hierarchy, it would rest more soundly on the statutes implementing our particular hierarchical system than on the Constitution.[203]
The constitutional explanation is also incomplete. A “mandate” is entirely the creature of legal texts other than the Constitution.[204] Even if Article III’s implied hierarchical structure had some significance, that structure would not necessarily require the mandate rule as it exists today. Although we are accustomed, for example, to the Supreme Court’s deciding discrete questions embedded within a case,[205] the Court’s constitutional judicial power extends to the entirety of the case or controversy.[206] There is nothing necessary to the federal judicial system about a superior court’s reliance on an inferior court to enter and execute a final judgment in line with a mandate that resolves one or two relevant legal questions. In other words, there is no obvious reason why the superior court could not (as a constitutional matter) enter and enforce its own judgments without the aid of a inferior court. Even at the time of the First Judiciary Act, that was the prescribed course in some circumstances.[207] And of course, the practical concerns existing at the time of the First Judiciary Act, such as limited terms of court and extensive circuit riding, do not exist today,[208] meaning that the Supreme Court and federal circuit courts are better positioned to carry out their own judgments today than they were in years past. The mandate is, on this account, a relic of problems that do not necessarily persist. But just because a practice is “familiar” does not mean that it is “constitutionally required.”[209]
Even if the Constitution’s text and structure are not independent sources for the mandate rule, then perhaps there is a sub-constitutional source. By this, I refer to what Professor Sachs calls a “constitutional ‘backdrop[]’: rules of law that aren’t derivable from the Constitution’s text, but instead are left unaltered by the text, and in fact are protected by the text from various kinds of legal change.”[210] The concept of a constitutional backdrop is a possible explanation for many legal doctrines that are supposedly “inherent,” such as the congressional contempt power[211] or state sovereign immunity from suit.[212]
Federal courts likewise possess “inherent” power.[213] A corollary to the Constitution’s grant of “the judicial power of the United States” is the “inherent authority to protect their proceedings and judgments in the course of discharging their traditional responsibilities.”[214] This inherent power supports many commonplace judicial functions: imposing fines for contempt,[215] vacating judgments obtained by fraud on the court,[216] and general docket and courtroom management,[217] to name only a few. One might argue that inherent in the judicial power is an unalterable grant of authority allowing courts to see that their judgments have effect.[218] From some of its earliest days, the Supreme Court has included on the list of “universally acknowledged” inherent judicial powers the ability “to impose … submission to [a court’s] lawful mandates.”[219] Perhaps that inherent power supports the mandate rule?
This argument, too, is unsatisfying. Consider first the object of the inherent power: whom, exactly, can a court coerce in furtherance of its judgment? It is one thing to say that a court can coerce the parties to the dispute into compliance with its own orders, but it is quite another to say that a superior court can coerce an inferior court to use the inferior court’s power to coerce the parties into compliance with the superior court’s judgment. Although the power to coerce parties has been recognized as inherent,[220] the power to coerce other courts has not. Indeed, that unbridled coercion of inferior courts seems to have been foreign to the common law of appellate remedies at the Founding.[221]
This makes sense. An inherent power is generally necessary to carry into execution the power vested in the institution.[222] The “judicial Power” vested in the federal courts is generally party-focused;[223] it is all about fixing the vested rights of specific parties by applying law to facts.[224] Thus, for example, the Constitution freezes in place—even against legislative alteration—the rule that a valid final judgment is conclusive of the rights between the parties regarding a specific occurrence at the heart of a case or controversy.[225] The relationship between courts is unnecessary to any single court’s decision about the relationship between parties. If an inferior court balks at obeying a lawful mandate, the superior court could (in theory) simply vacate the defiant judgment, issue its own judgment, and award execution on that judgment. There is no obvious inherent need for final judgments to be entered only at a trial court, nor by extension for inter-court conflict and forced submission of the inferior court to the superior one.[226] We could have another system and the judicial power would still work just fine.
Nevertheless, some courts have described “the authority to order compliance” with a mandate as “part of [the] inherent powers” of federal courts.[227] To be sure, it is settled law that in “appropriate circumstances” a court can use mandamus against non-parties to an action that “are in a position to frustrate the implementation of a court order or the proper administration of justice.”[228] A superior court’s mandamus to an inferior court to implement a mandate might be viewed as of that character.[229] But that power is also not inherent; it rests in Congress’s authorization that “all courts established by Act of Congress” can “issue all writs necessary or appropriate in aid of their respective jurisdictions and agreeable to the usages and principles of law.”[230] The federal writ power has always been viewed in the United States as a recognition by the political branches that courts will encounter “interstices” in their federal judicial power that require resort to an extraordinary writ.[231] The All Writs Act, no less than the mandate itself, traces back to the First Judiciary Act.[232] It seems that the First Congress—which included eight representatives and eleven senators who had served as delegates to the 1787 Constitutional Convention[233]—did not view enforcement of mandates as inherent in the judicial power, but as requiring a legislative act.[234]
Even if some enforcement powers were inherent, it does not follow that unbridled use of that inherent power survived the First Judiciary Act and its successor statutes, given the view that Congress had plenary power to set federal court procedures.[235] As modern doctrine makes plain, “the exercise of an inherent power cannot be contrary to any express grant of or limitation on the . . . power contained in a rule or statute.”[236] For good reason: Congress’s power “[t]o make all Laws which shall be necessary and proper” extends to “carrying into Execution . . . all other Powers vested by this Constitution in the Government of the United States, or any Department or Officer thereof.”[237] Because any inherent judicial authority flows from the Constitution’s vesting of judicial power,[238] Congress can regulate that authority by statute. It follows that, to the extent the obligations under a mandate and enforcement of the same are inherent to the judicial power, Congress can supplant the background rule, define a lawful mandate’s scope, and regulate the means of enforcement.
The purpose of this Part is not to answer definitively the scope or content of any mandate-rule-like doctrine rooted in constitutional or sub-constitutional sources. Instead, this Part’s principal goal is to make a narrower claim: before falling back on constitutional inference or inherent power as a basis for the mandate rule, one should ask whether a statute or valid rule speaks to the issue, as those source can likely inform or supplant whatever background rule might exist (to the extent there is a background rule at all). As Part II showed, Congress has from the earliest days spoken comprehensively on the relationships among courts at various levels within the nation’s judicial systems,[239] and those statutes became the mandate rule’s source, as some courts already recognize.[240] Absent some sound argument that these statutes are unconstitutional—and, indeed, the better view is that they carry out Congress’s power to regulate the federal courts—I suggest that there is nothing to be gained from looking anywhere else but the statutes as supporting the mandate rule.
Having established that the mandate rule is fundamentally a statutory doctrine, this Part explores how certain aspects of the doctrine have departed from the statutory basis. In other words, this Part looks at instances where courts have applied the doctrine, without reference to its statutory source, in a way that has created conflict, confusion, or error. Along the way, this Part suggests how straightforward application of statutory text can inform existing disputes in the doctrine and correct certain of the errors.
One of the mandate rule’s key precepts is that the appellate mandate is binding on the lower court on remand, so that the lower court must carry out the mandate.[241] Some courts have taken this further: because the lower court must comply with the mandate, it cannot do anything foreclosed by the mandate; and to say that a court cannot do something is simply to say it is without jurisdiction to take that action.[242]This has led to a seemingly deep circuit split on whether the mandate rule is jurisdictional or, like the remainder of the law of the case doctrine, merely a rule of practice guiding the court’s discretion.[243] This Section argues that, in different contexts, both sides are correct.
The current split in authority is relatively stale. The cases holding that the mandate rule is jurisdictional tend to be older than those treating it as a rule of practice.[244] One possible explanation is the Supreme Court’s relatively recent project to clean up the meaning of the term “jurisdiction” to refer generally to “prescriptions delineating the classes of cases a court may entertain . . . and the persons over whom the court may exercise adjudicatory authority,”[245] but not to other rules governing how cases are processed through the federal judicial system.[246] The mandate rule may well be thought of as falling into that latter category; it is a rule for how issues are to be resolved as a case works its way through various stages of litigation, from judgment to appeal to proceedings on remand.[247] But the split at issue predates the Supreme Court’s jurisdictional clean-up efforts, and older decisions treating the mandate rule as jurisdictional may have been using the term “jurisdiction” with the looseness of which the Supreme Court would disapprove today.[248]
An alternative driving force behind the split, however, may be the extent to which the mandate rule is part of the law of the case doctrine. One view is that the mandate rule is just a specific application of law of the case,[249] which is not jurisdictional.[250] But the Ninth Circuit, for example, treats the mandate rule as jurisdictional and as a doctrine “broader” than law of the case; although “both doctrines serve an interest in consistency, finality and efficiency,” the mandate rule serves a separate interest “in preserving the hierarchical structure of the court system.”[251] If the mandate rule is just a subset of a non-jurisdictional doctrine, then it should also not be jurisdictional; if the mandate rule is materially different, then maybe it is jurisdictional.
The jurisdiction-versus-procedural-rule debate is somewhat of the Supreme Court’s own making. The idea of the mandate rule as jurisdictional comes from language in the Court’s early cases. From the earliest days, the Supreme Court applied the mandate rule to foreclose attacks on the validity of its own prior judgments, even if there was cause to believe that the Court was without jurisdiction to issue its initial decision.[252] Thus, for example, Martin v. Hunter’s Lessee held that the Court’s judgments are “supposed to be conclusive” because “no statute has provided any process by which [the Court] can revise its own judgments.”[253] As a result, a recurring view was that the Supreme Court was without “[a]ppellate power”—i.e., jurisdiction—“to review its decisions” after the mandate issued.[254] This view persevered into the twentieth century.[255] As the Eighth Circuit once put it, “the [S]upreme [C]ourt has no appellate jurisdiction over its own judgments” and therefore “cannot review or modify them after the case has once passed, by the issuance of the remittitur, from its control.”[256]
Over time, the Supreme Court began to intermix these rulings, which were rooted in an interpretation of the statutes governing appellate power, with the more general principle of adherence to issues of law previously decided in a particular case.[257] For example, the Supreme Court began citing its mandate rule cases in support of the general proposition that the Court will decline to reconsider any question of law decided in an initial appeal.[258] This principle came to be called the “law of the case,” which “as applied to the effect of previous orders on the later action of the court rendering them in the same case, merely expresses the practice of courts generally to refuse to reopen what has been decided, not a limit to their power.”[259] This distinction between practice and power rested on an intuitive ground: where a second appeal simply re-raises an issue decided in the first appeal, the practice was not to “dismiss that appeal for want of jurisdiction,” but to “entertain[] jurisdiction” and affirm the inferior court’s judgment applying the law handed down on the first appeal.[260]
The Supreme Court’s non-jurisdictional view also makes sense when state courts are considered. If the mandate rule is always jurisdictional, that raises an important problem about the relationship between the Supreme Court and state courts. As a statutory doctrine, the mandate rule cannot withdraw from a state court the right to exercise its own judicial power unless some grant of federal authority authorizes such a withdrawal of jurisdiction.[261] Although state courts wield judicial power, they do not exercise the “judicial Power of the United States” that is vested under Article III and, by extension, unambiguously subject to congressional regulation under Article I’s Necessary and Proper Clause.[262] It is beyond this Article’s scope to discuss in full when Congress can abrogate state court jurisdiction, a topic that has been discussed elsewhere.[263] Rather, it suffices here to say that there are good reasons to doubt that 28 U.S.C. § 2106 withdraws jurisdiction from state courts when the Supreme Court issues a mandate in a case on appeal from that state’s court system.[264] The statute contains no expressly jurisdictional language,[265] and various federalism presumptions may fairly counsel against inferring such a rule as to state courts.[266] Yet, the language in the statute is the same for any “order of a court lawfully brought” up for review.[267] If the statute is not jurisdiction-restricting for state courts, it follows that it should not be read as jurisdiction-restricting for federal courts.
The current division in authority over whether the mandate rule is jurisdictional reflects an all-or-nothing approach. But a careful reading of § 2106 reveals that neither side of the jurisdiction-versus-procedural-rule debate is exactly right. In many situations, the mandate rule is no more jurisdictional than any other rule directing how to process a case. After all, the mandate rule emanates from a statutory command unrelated to the types of cases or persons over whom the court has power.[268] Instead, to say that the mandate requires an inferior court to take some action is just to identify 28 U.S.C. § 2106 as supplying a rule of decision for certain disputes. In other words, some disputes on remand from an appellate court will implicate instructions that an appellate court already gave, and many of those instructions will be lawful under § 2106’s broad grant of power to “remand,” “direct” particular judgments and orders, or “require . . . further proceedings.”[269] In those situations, § 2106 provides a rule of decision: follow and apply the mandate. At the same time, there are some circumstances in which the mandate rule is jurisdictional. Specifically, it is fully consistent with the contemporary law of federal jurisdiction, and with the history informing the statutory mandate rule, to characterize an inferior court as having no jurisdiction to question a mandate’s validity.
The logic of the Supreme Court’s mandate rule cases—and, in particular, their holdings that the rule is not jurisdictional—makes sense as applied to the highest court in the system, such as the Supreme Court itself. Federal appellate power is all about which courts get to review which judgments.[270] When a judgment is “lawfully brought before it for review,” the appellate court is imbued with express statutory authority under § 2106 to examine the judgment and then enter a judgment of its own regarding the effect of the judgment on review.[271] But appellate jurisdiction runs in only one direction: up the chain from inferior court to superior court.[272] Thus, on remand from the Supreme Court, a district court might enter a final judgment in line with the Supreme Court’s ruling; the losing party might notice an appeal of that final judgment;[273] and if the Supreme Court (for whatever reasons) grants review again,[274] nothing in § 2106 requires it to adhere to its own prior decision when assessing whether the final judgment now on review should be “affirmed, modified, vacated, set aside, or reverse[d].”[275] Instead, the statute allows for sweeping review of the judgment on review’s lawfulness.[276] The Supreme Court has all the power it needs from the certiorari statute and § 2106 to correct errors of law, including those in its own prior decision. What holds the Court to respect its own prior ruling on a point of law is the settled rule of practice known as the law of the case doctrine,[277] not any deficiency in statutory authority.
But the world is different for inferior courts. Suppose that, on remand from the federal circuit court, a party opposes entry of final judgment in the district court on the ground that the appellate judgment is invalid as beyond the circuit court’s jurisdiction. Perhaps the appellant noticed its original appeal too late, thereby depriving the circuit court of appellate jurisdiction,[278] but no one caught the error. Of course, a judgment in excess of subject matter jurisdiction is not merely voidable, but outright void.[279] Can the district court examine the circuit court’s judgment for validity? The mandate rule says no: among the things an inferior court cannot do is “examine” the mandate “for any other purpose than execution.”[280] Thus, for example, the Supreme Court held in Skillern’s Executors v. May’s Executors that a circuit court was bound to carry out a mandate directing resolution of the merits even though the circuit court on remand from the Supreme Court identified a defect in its own jurisdiction.[281] Even in the absence of federal jurisdiction, the circuit court could not disregard the Supreme Court’s mandate as invalid.[282]
For the hypothetical district court described above, this rule is jurisdictional. Section 2106 confines the examination of “any judgment, decree, or order” to a “court of appellate jurisdiction” reviewing a judgment “lawfully brought before it.”[283] The district court cannot avail itself of this power.[284] The way to lawfully bring the circuit court’s judgment for review in a different forum was by certiorari to the Supreme Court.[285] Once the mandate is sent down, however, its validity is conclusive in the same proceeding.[286]
If this rationale seems familiar, that is because it follows the same basic logic as the Rooker-Feldman doctrine.[287] That doctrine “prevents the lower federal courts from exercising jurisdiction over cases brought by state-court losers” who seek to challenge in federal court the final state court judgment against them.[288] The rule follows from a negative inference: by statute, Congress generally empowered the Supreme Court alone among the federal courts to “reverse or modify” state court judgments, so district courts may not entertain original actions by state-court losers asking the court to declare a final state court judgment null and void, as that “would be an exercise of appellate jurisdiction” reserved to the Supreme Court.[289] So, too, in the mandate rule context: by assigning the power to review judgments only to courts whose appellate jurisdiction has been properly invoked, § 2106 implies that other courts lack that power.[290] And “it is commonplace” for the word “jurisdiction” to be used as “specifying the remedial power[] of the court.”[291] A restriction on an inferior court’s power to disregard a higher court’s judgment in the same case—or, to borrow § 2106’s term, to “set aside” that judgment as void[292]—is no less a jurisdictional limitation than any other withdrawal of power.
What I have just described is a very narrow application in which the mandate rule is jurisdictional; it is particular to those situations in which an inferior court is asked not to give effect to a superior court’s judgment because of some legal defect in the judgment itself. In that situation, to say that the inferior court cannot “examine” the judgment, even “for error apparent,”[293] is simply to recognize that the examination power does not apply except when a judgment is lawfully brought up to a superior court for appellate review.[294] But the mandate rule covers a wide array of other obligations, including the inferior court’s adherence to questions of law actually resolved and obedience to directions about “the entry of such appropriate judgment, decree, or order” or “further proceedings” to be had.[295] Where the validity of the directive is not at issue, the lower court’s obligation to carry out the mandate is not a restraint on its jurisdiction, but a requirement that flows from § 2106’s rule of decision identifying the directive as lawful. Put differently, when the only issue is the inferior court’s obedience to a mandate, disobedience is no more an act in excess of jurisdiction than any other failure to apply a statute or valid procedural rule correctly.
The lesson of this Section is that both sides of the jurisdiction-versus-procedure debate are right in some situations. There are times when the mandate rule is properly understood as jurisdictional, and there are times when it is merely a rule of practice. This matters for a practical reason: if courts treat every obligation under the mandate rule as a limit on jurisdiction, then an aggrieved party who fails to object to—or affirmatively invites—a district court’s deviation from the mandate can prevail on appeal by raising the mandate rule for the first time.[296] After all, jurisdictional objections of this sort are generally not subject to waiver or forfeiture.[297] The way to distinguish the two situations is by careful reference to § 2106 and its allocations of power to different courts.[298] If the inferior court purports to use power under § 2106 that it does not have, that is a jurisdictional error. But if the inferior court merely disobeys a superior court’s use of § 2106’s power, the error is merely procedural.
A cornerstone of modern federal administrative law litigation is the petition for review, a standard means for challenging agency action.[299] On a petition for review, a party aggrieved by agency action challenges that action’s lawfulness directly in a circuit court of appeals.[300] An inferior court is not involved; the agency is both the party bound by the resulting judgment and the initial decision-making body to which a remand occurs.[301]
To what extent does the mandate rule bind an agency? The prevailing view is that the doctrine applies with equal force.[302] As a leading treatise puts it: “An administrative agency is bound by the mandate of a reviewing court much as a lower court is bound by the mandate of a higher court,” which the treatise calls a “general principle” that “is easily recognized.”[303] Intuitive though that may be, there is significant nuance that some courts appear to miss—namely, the scope of the authorized statutory review and determination power.
As this Article has discussed, the “mandate rule” evolved from statutes that gave the Supreme Court broad power to correct errors in an inferior court’s judgment.[304] Over time, that statute evolved to state with particularity the things that an appellate court can require of a lower court, ranging from straightforward affirmance or reversal of its judgment to remanding with instructions “direct[ing] the entry of such appropriate judgment, decree, or order” or “requir[ing] such further proceedings to be had as may be just under the circumstances.”[305] But that statute is plain: it applies to review of a “judgment, decree, or order of a court.”[306] If appellate courts are looking to supervise agency action on remand, they better look elsewhere for authority.
And, of course, such authority exists, just not always as broad as that enjoyed with respect to inferior courts. As discussed in Part II, the early twentieth century saw Congress experiment with approaches to judicial review of agency action, culminating in statutes that form a backbone of modern administrative law and that generally provide a judicial scheme “borrowed from the understandings that govern the relationship between appeals courts and trial courts in civil litigation.”[307] For example, the Hobbs Act governs judicial review of several agencies’ final rules, regulations, and orders, but provides that the reviewing court may only “enjoin, set aside, suspend (in whole or in part), or . . . determine the validity of” the agency action.[308] Similarly, the Federal Energy Regulatory Commission’s judicial review statutes provide federal appellate courts with the power “to affirm, modify, or set aside” challenged orders.[309] These statutes do not expressly provide for directing the entry of particular orders or requiring particular proceedings on remand—i.e., instructions that a federal appellate court may hand down to an inferior federal court under § 2106.[310] Nevertheless, appellate courts have imported the judicial mandate rule into the realm of administrative law, sometimes in a manner incompatible with the narrower scope of authority provided by the applicable judicial review statutes.[311]
A key early offender was the D.C. Circuit. In Pottsville Broadcasting Co. v. FCC, that court reversed the Federal Communications Commission’s denial of an application to construct a broadcast station.[312] On remand to the FCC, the agency started a new proceeding in which it set for hearing three applications—Pottsville Broadcasting’s initial application plus two new applications—with the license to go to the applicant that would best serve the public interest.[313] Pottsville Broadcasting returned to the D.C. Circuit, seeking “a writ of mandamus to require the Commission to grant [its] application . . . on the record as submitted to and considered by” the court in the initial appeal.[314] The court agreed with Pottsville Broadcasting, reasoning that “as far as is practicable the order of the court entered on an appeal from the Commission ought to have the same effect and be governed by the same rules as apply in appeals from a lower federal court to an appellate federal court in an equity proceeding.”[315] The court then invoked the mandate rule against the agency and issued the writ.[316]
The Supreme Court reversed.[317] With respect to the mandate rule, the Court observed that this “was not a mandate from court to court but from a court to an administrative agency.”[318] The Court reasoned that the relationship between the federal courts is not necessarily the same as the relationship between federal courts and administrative agencies, and so
technical rules derived from the interrelationship of judicial tribunals forming a hierarchical system are taken out of their environment when mechanically applied to determine the extent to which Congressional power, exercised through a delegated agency, can be controlled within the limited scope of ‘judicial power’ conferred by Congress under the Constitution.[319]
But the Court did not rest on these abstract differences; it ultimately rooted its analysis in the applicable judicial review statutes.[320] As the Court observed, the original Radio Act of 1927 authorized a reviewing court to “alter or revise the decision appealed from and enter such judgment as to it may seem just.”[321] This made the reviewing court “a superior and revising agency in the same field” of delegated authority as the agency itself.[322] But Congress soon amended that initial act to limit the court’s role to “affirming or reversing the decision of the commission” based on the resolution only of questions of law.[323] In light of that revised scope of authority, the Court held that the D.C. Circuit’s initial ruling “did not create rights of priority” in the license applicant.[324] In other words, the D.C. Circuit could enter a judgment that conclusively determined the lawfulness of the initial denial of the application, but it had no power to direct a certain outcome or manner of proceeding on remand.[325]
Notably, the Supreme Court did not disavow the mandate rule’s application to administrative agencies.[326] Instead, after reasserting its own power to interpret the D.C. Circuit’s mandate de novo, the Court turned to a “much deeper issue” about the legal effect that the mandate could have, not just whether the D.C. Circuit’s gloss on its own mandate was correct.[327] In this respect, the Court’s analysis focused on the appellate court’s power and, by extension, the scope of the obligation that the mandate created for the agency.[328] And as it has always done, the Court situated the mandate’s lawful effect in the applicable statutory framework for review.[329]
Pottsville was not necessarily well received. In 1952, Congress responded to the Pottsville decision by enacting 47 U.S.C. § 402(h).[330] Under that provision, if a court enters an order reversing the FCC, the court “shall remand the case to the Commission to carry out the judgment of the court,” which “unless otherwise ordered by the court” will carry out the judgment “upon the basis of the proceedings already had and the record upon which said appeal was heard and determined.”[331] The basic idea was to give the appellate court “a measure of control commensurate with the dignity and responsibility of that tribunal,”[332] with some commentators observing that Section 402(h) “closely resembles the mandate rule enunciated by the lower court” in Pottsville.[333] But that Congress may have abrogated Pottsville’s immediate holding does not undermine that case’s rationale.[334] The rule is simple enough: judicial supervision of inferior courts is not necessarily coextensive with judicial supervision of administrative agencies, and the right place to look for mandate-rule-like supervisory power is the governing statutory review scheme.[335]
But Pottsville’s lessons did not seem to stick, as City of Cleveland v. Federal Power Commission shows.[336] In that case, the D.C. Circuit held on petition for review that the Federal Power Commission committed legal error when it refused to investigate whether a filed rate was unauthorized and without effect.[337] The court reversed the agency’s disposition of a single issue and remanded to the agency “for further proceedings consistent with [its] opinion.”[338] On remand, the FPC focused its investigation on the one issue specifically reversed but not all “other energy charges” in dispute.[339] That prompted a motion to enforce the initial mandate, which the court granted.[340] The court acknowledged Pottsville, which it cited for the proposition that the mandate rule makes “no exception for reviews of administrative agencies.”[341] But the D.C. Circuit distinguished that case in substantial part, reasoning that its initial ruling addressed “only purely legal questions,”[342] as permitted by statute. And the resolution of those questions required the agency to follow the court’s initial, broad determination “outlaw[ing] all unilateral alterations of agreed-upon rates” and “to determine whether the energy charges”—any of them at issue—“are of that ilk.”[343] This, the court reasoned, followed from “the duty of a lower court to follow what has been decided by a higher court at an earlier stage of the case.”[344]
The City of Cleveland approach has become well-accepted. In Office of Consumers’ Counsel v. FERC,[345] for example, the D.C. Circuit addressed FERC’s conclusion that an interstate gas pipeline’s practices were not “abuse” under the Natural Gas Policy Act of 1978.[346] The court held that the second part of FERC’s two-part test for “abuse” was “either flatly at odds with the plain meaning of the statute or incomprehensible,” and therefore reversed and remanded for FERC “to reconsider its interpretation of abuse and construct a new test that adheres to the statutory requirement” before then “be[ing] required to reevaluate [the] challenged conduct in light of the revised test of abuse.”[347] A year later, the D.C. Circuit granted a motion directing compliance with the mandate.[348] The court broadly stated, with citation to City of Cleveland alone, that it “has the authority, through the process of mandamus, to correct any misconception of its mandate by a lower court or administrative agency subject to its authority,” a “recourse” that a party “always has.”[349] But the court never paused to consider the key question: What did the mandate lawfully require of the agency, given Congress’s applicable grant of statutory judicial review authority?
These cases reflect a strong view of the mandate rule as applied to administrative agencies. City of Cleveland and Office of Consumers’ Counsel each equate an appellate court’s authority over an administrative agency with its authority over a lower court, essentially treating the mandate rule as identical regardless of the body supposedly bound by the mandate.[350] This strong view of the mandate rule for agency action has become an article of faith not just in the D.C. Circuit,[351] but elsewhere.[352] Thus, a prevailing view in the federal circuit courts is that the mandate rule operates identically as between courts and agencies.[353]
The prevailing view is mistaken. This is not necessarily because there are no situations in which the mandate rule operates on an agency the way it operates on a court. Rather, the view is mistaken because it takes that parity for granted, often without undertaking the careful parsing of statutory authority that the Supreme Court directed in Pottsville.
Part of the problem in applying the mandate rule to administrative agencies may be the tendency to conflate the mandate’s dual aspects. Because an agency is both a party to the judicial proceeding and the initial deciding body, the judicial proceeding results in both a judgment that settles the agency’s rights and obligations vis-à-vis the petitioner and a mandate governing how to implement that judgment. More confusing still, the mandate ordinarily consists of little more than the court’s opinion and “a certified copy of the judgment.”[354] The same documents serve two distinct roles. The judgment controls what the agency owes the petitioner; the mandate controls how the judgment is given effect and implemented. Put differently, a judgment fixes vested rights between parties, but a mandate directs the inferior body to act.[355] Importantly, however, different legal rules govern those documents’ effects in their different roles. A final federal judgment’s effect is governed by federal constitutional law (in particular, the separation of powers and vesting of judicial power in the federal courts) and federal common law.[356] But, as this Article has argued, the lawful effect of a judicial mandate is governed by statute.[357]
Underlying the prevailing treatment of the mandate rule as applied to agencies is the assumption that the court can dictate the how question. That intuition comes honestly; with respect to inferior courts, § 2106 confers broad power on the appellate court to direct action on remand.[358] But there are good reasons to doubt that the same control exists in the court-agency relationship. As Pottsville recognized decades ago, agencies have power that exceeds, and is different from, “the conventional judicial modes for adjusting conflicting claims”;[359] and it remains true today that “agencies, unlike lower courts, are not faithful agents of the remanding tribunal,” because agencies have their own agendas and their own obligations to the President and Congress.[360] The relevant statutes reflect this by insulating agency orders from over-intrusive judicial review.[361] As previously discussed, Congress experimented with more sweeping grants of judicial revision and supervisory power before settling on narrower judicial review statutes that generally limit courts to affirming, reversing, or modifying agency actions to correct legal error and only on an existing record—an approach made necessary by the Supreme Court’s consistent holdings that any more intrusive review amounted to impermissible administrative action by federal courts.[362] It is not always the case that agency judicial review statutes confer the same breadth of appellate supervision for agencies as for inferior courts, especially given the separation of powers concerns that led to the modern model of appellate review of administrative action.[363]
At bottom, federal appellate courts should be more attentive to the scope of their power of review. Consider again City of Cleveland.[364] In that case, the court recited its view that the mandate rule applies to agencies and then faulted the FPC for not, on remand, carrying the logic of the D.C Circuit’s opinion to its furthest reach.[365] Although the D.C. Circuit had necessarily decided only a single point on the initial appeal,[366] it concluded that the agency on remand should have conducted proceedings that encompass everything that “inexorably” followed from the opinion’s logic because the mandate rule “applies to everything decided, either expressly or by necessary implication.”[367] Notably, the court in the initial proceeding remanded “for further proceedings consistent with the opinion,”[368] a directive that would ordinarily entail an obligation to carry out what an opinion implicitly required. But the missing piece of the court’s analysis is whether that directive was lawful in the first place, given the statutory review scheme.
So, how should City of Cleveland have proceeded? For starters, the baseline presumption is that administrative agencies are different from inferior courts in constitutionally material ways.[369] The question is always what level of judicial supervision the Constitution allows Congress to authorize, and what Congress then actually authorized within that sphere.[370] At the time of City of Cleveland, the applicable judicial review statute provided only for a “judgment and decree . . . affirming, modifying, or setting aside, in whole or in part,” the order on review.[371] Notably, that statute did not provide for remand or direction on remand, even though the same statute authorized the court to direct proceedings in another way.[372] In particular, if a party to a petition for review adduced new and material evidence that it reasonably failed to present to the agency in the first instance, the court could “order such additional evidence to be taken before the Commission and to be adduced upon the hearing in such manner and upon such terms and conditions as to the court may seem proper.”[373] In other words, Congress authorized judicial direction of agency proceedings in one situation (i.e., supplementation of the record), but not with respect to remands following reversal or vacatur.[374] Thus, the Pottsville baseline rule against judicial interference with agency proceedings should have held, and the court should have recognized the absence of power to mandate how a proceeding occurs on remand.
To be sure, the court might have concluded on a second petition for review that the logic of its prior opinion compels the conclusion that the FPC erred by taking a certain approach on remand (e.g., because the agency’s approach was contrary to law).[375] But that is different from situating the error in defiance of the mandate. The former is a violation of statute correctable on petition for review, whereas the latter is a violation of judicial order correctable by mandamus, a process that can disrupt and derail agency proceedings by injecting the courts mid-stream in the administrative process.[376] Perhaps some supervision is desirable, but that is ultimately a decision for Congress. A serious objection to the current state of the mandate rule as applied to agencies is that Congress’s designed relationship between courts and agencies, and the formers’ scope of review of the latter, reflects an informed choice about when courts should become involved.[377] Generally, the answer is only on final action.[378] An administrative mandate rule doctrine that disregards the significance of the statutory review scheme’s design—including the general emphasis on finality and the absence of an express, broad power to direct proceedings on remand—could be subverting Congress’s will by aggrandizing judicial power under the guise of enforcing mandates, when in fact the mandates exceed their statutory authority.
That said, there are situations in which resolution of a purely legal question will necessarily require a particular outcome on remand. Consider this hypothetical: A party files a petition for rulemaking for an agency to repeal a rule on the ground that the rule is inconsistent with the authorizing statute.[379] The agency denies the petition on the ground that the rule is lawful. The party then files a petition for review, arguing that the denial was arbitrary, capricious, and contrary to law because the rule is in fact legally invalid.[380] The court agrees and grants the petition. That judgment fixes the parties’ rights with respect to the petition for repeal: it conclusively determines that the agency’s denial of the petition for repeal was unlawful, and the prevailing petitioner now has a vested right, under the judgment, to have the petition granted.[381] If the agency still refuses to grant the petition, the party can seek further relief to “compel agency action unlawfully withheld or unreasonably delayed”[382]—relief that courts frequently recognize is identical to mandamus.[383] In that situation, there is functional equivalence between the judgment and the mandate. On remand, there is only one action the agency can take because statutory rules imbue the judgment with the same effect (subject to the same remedies) as § 2106 imbues a judicial mandate to an inferior court.[384] But it does not follow that the resolution of every legal question leaves the agency with only one option on remand.
As noted, City of Cleveland does not stand alone. Its approach has spread to several opinions across circuits, which now uncritically assume sweeping mandamus power without careful reference to the statutes authorizing the scope of judicial determination.[385] The purpose of this Article is not to canvas comprehensively all such statutes. Instead, the central point is that the scope of a lawful mandate to an agency on remand—and, by extension, what an appellate court can enforce via mandamus—depends on grants of power embodied in statutes. Before invoking the mandate rule, courts should confirm that the supposed mandate to the agency (as opposed to the judgment fixing rights) has a statutory basis.
Sometimes there might be an agency-specific mandate-rule-like statute. Recall 47 U.S.C. § 402(h),[386] which some have described as codifying the mandate rule with respect to judicial review of the Federal Communication Commission’s orders.[387] Congress enacted § 402(h) to abrogate Pottsville’s holding that, on remand from appeal of an adverse licensing decision, the FCC could consider the prevailing appellant’s application alongside newly filed applications.[388] Section 402(h) provides that an appellate court that reverses an FCC order “shall remand the case to the Commission to carry out the judgment of the court,” and the FCC’s duty is “to do so upon the basis of the proceedings already had and the record upon which said appeal was heard and determined,” unless the court orders otherwise.[389] This provision imbues appellate courts with mandate-rule-like power only insofar as they can relieve the FCC of compliance with the default statutory rule for re-adjudicating matters on remand.[390] The obligation for how to carry out the judgment does not come from the court; it comes from Congress, which directed a method of adjudication.[391]
Even if § 402(h) were understood as a type of agency-specific mandate rule, the appropriate statutory parsing that this Article advocates suggests that it applies only to appeals, not to petitions for review. To see the distinction, look elsewhere in § 402. That provision creates a two-avenue system for review of FCC orders: actions on applications and licenses are subject to appeal under § 402(b), whereas all other final agency actions are subject to a petition for review under the Hobbs Act and § 402(a).[392] Every other subsection in § 402 refers to an “appeal” or “the appellant.”[393] From context, the implication is that § 402(h) is likewise particular to appeals under § 402(b),[394] a result that accords with its genesis as an abrogation of Pottsville,[395] which was an appeal rather than a petition for review.[396] As such, even if § 402(h) is some form of statutory mandate rule, courts should be careful to cabin it to its appropriate role—decisions on appeal under § 402(b), not decisions on petitions for review under § 402(a).[397]
The purpose of this § 402(h) exercise was to emphasize that looks can be deceiving. Before reflexively applying the mandate rule to administrative agencies, federal appellate courts should look carefully at the statutes governing the petition for review to ascertain the scope of their lawful mandates on remand. And even where a statute reflects a mandate-rule-like decision rule, only a careful parsing of the statute—with reference to text, structure, context, history and purpose, and common sense[398]—can confirm whether the appellate court has the power to direct particular proceedings on remand.
In sum, federal circuit courts assume almost by habit that the mandate rule applies equally to agencies as to courts.[399] But if, as this Article maintains, the mandate rule derives from statutory authority governing the determination of cases, then any application of the mandate rule to an agency must have its own statutory hook. Often, courts asserting the mandate rule against an agency fail to engage in a meaningful search for, or analysis of, this authority. As this Section has shown, that may be responsible for significant error, as courts lose sight of their narrower scope of control of agency proceedings as compared to judicial proceedings in inferior courts.
So far, this Part has tackled some of the bigger issues that have resulted from imperfect attention to the mandate rule’s statutory bases. This Section surveys a few other odds and ends in the doctrine that, although perhaps less pressing than the earlier issues, are still examples of the doctrine’s divorce from statute.
The Fifth Circuit has a unique practice among the federal circuit courts. Sometimes, rather than reverse a judgment and remand, that court will instead “reverse and render.”[400] When that happens, the appellate mandate becomes the case-terminating final judgment.[401]
The court appears to have adopted this practice from Texas appellate procedure.[402] Texas’s counterpart to § 2106 lists among the types of appellate judgments: “reverse the trial court’s judgment in whole or in part and render the judgment that the trial court should have rendered.”[403] Texas’s own version of the mandate rule then comes into play. When an appellate court renders the judgment that a trial court should have rendered, “that judgment becomes the judgment of both courts.”[404] The trial court’s duty is to enforce the appellate court’s judgment as its own.[405]
In a small and largely insignificant way, the Fifth Circuit’s adoption of Texas practice violates § 2106.[406] As Part II explained, § 2106’s predecessors reflected a consistent policy choice: the inferior federal courts were responsible for rendering and enforcing execution on judgments.[407] As those statutes evolved, they repeatedly stated that the appellate court could direct what judgment to render.[408] Under the existing statute, the Fifth Circuit still surely has the power to “direct the entry” of a particular judgment.[409] But no provision in § 2106 allows for the appellate court’s rendering its own judgment, even though it can “set aside” the judgment on review.[410] Although one might respond that the Fifth Circuit could “require” the district court to adopt the appellate judgment as its own final judgment,[411] that would be an awkward reading of the statute. Section 2106 delineates case or issue terminating directives (“direct the entry of such appropriate judgment, decree, or order”) and case continuing directives (“require such further proceedings to be had”).[412] It would be odd to read the latter as encompassing the former.
The distinction between reversing and rendering and reversing and remanding with instruction to render a particular judgment is largely academic. But the Fifth Circuit’s blind imitation of a Texas Rule of Appellate Procedure and Texas’s state-law mandate rule, rather than strict adherence to its own statute governing determination of appeals, is emblematic of a larger problem of inadequate attention to the mandate rule’s statutory source.
The Ninth Circuit has its own quirk. Traditionally, the mandate is enforceable via two avenues: a second appeal or a writ of mandamus.[413] Over the centuries, the Supreme Court has articulated various requirements for mandamus: (1) clear and indisputable entitlement to relief; (2) no other adequate means of redress, such as an appeal; and (3) appropriateness under the circumstances.[414] The Ninth Circuit has further refined these requirements into a well-established, five-guideline test first announced in Bauman v. U.S. District Court.[415] But in a series of cases seeking mandamus to enforce the mandate, the Ninth Circuit began to disregard the Bauman factors[416] before holding outright that “Bauman does not apply when mandamus is sought on the ground that the district court failed to follow the appellate court’s mandate.”[417] Whether the ordinary rules for mandamus apply in the mandate rule context is now the topic of a split with the D.C. Circuit.[418]
The authority to issue mandamus in defense of a mandate has always rested in statute, namely the All Writs Act.[419] Today, that statute confines the writ power in federal courts to what is “necessary or appropriate in aid of their respective jurisdictions and agreeable to the usages and principles of law.”[420] When discussing mandamus, the Supreme Court generally nods to that basis of authority.[421] It follows that what is “necessary or appropriate” to aid a federal court’s jurisdiction and “agreeable” to the law is what the Supreme Court elaborates when exercising mandamus.[422] And tellingly, the Supreme Court has declined to issue mandamus where, for example, the petitioning party could simply appeal again.[423] That is consistent with long-standing general principles of mandamus, including that the writ should not issue where another remedy lies.[424]
The Ninth’s Circuit’s view that the mandate rule is an exception to the ordinary rules of mandamus is just another example of disregard for the governing statutes. The Ninth Circuit cases that ignored the ordinary mandamus standard never cited the statute or analyzed the precedents applying it.[425] Instead, the Ninth Circuit adopted an essentially per se rule that mandamus is appropriate to correct deviation from its mandates, a rule it seems to have derived from General Atomic Co. v. Felter. [426] But in General Atomic, the Supreme Court declined to issue the writ—a result that explains that decision’s failure to discuss at length the traditional factors governing mandamus.[427]
In short, the Ninth Circuit’s exceptional view of mandamus in the mandate rule context rests on inapt precedents rather than a faithful application of the All Writs Act, which carefully circumscribes use of the extraordinary writ.[428] By contrast, the Ninth Circuit’s Bauman factors derive from the accumulated guidance of appellate decisions analyzing mandamus against the backdrop of the All Writs Act.[429] From the First Judiciary Act to the present day, the All Writs Act has been a statutory cornerstone for the mandate rule’s enforcement.[430] In the Ninth Circuit, a return to applying the mandate rule in line with its statutory source requires a return to applying the Bauman factors.
This Article has argued that the mandate rule ultimately derives from statutory grants of authority governing determinations by federal appellate courts. For the whole of this country’s history, the mandate rule has reflected statutory choices about (1) the allocation of responsibility for entering and enforcing final judgments; (2) the scope of appellate review power; (3) the power to direct specific proceedings in inferior courts; and (4) the remedies available for noncompliance. The mandate rule is not rooted in the Constitution’s implied judicial hierarchy or in any unbridled inherent powers that courts possess. A statute controls.
Unfortunately, many courts have lost sight of these statutory origins. As a result, many modern appellate decisions recite and apply the doctrine without reference to the precise powers Congress conferred in different situations. This has created several errors and conflicts in the doctrine, including deep disagreement over when (if ever) the doctrine limits jurisdiction; the extent of judicial control of agency action on remand; and the ordinary rules for what decrees and remedies an appellate court can hand down. Fortunately, the solution to the problems that plague the doctrine and divide the courts is simple: just read the statute.
It divided the country into districts, each district generally coextensive with a state, each with a Federal District Court, and a District Judge. The districts, in turn, were grouped into three circuits. In each circuit, a circuit court, made up of two Supreme Court justices and one district judge, sat twice a year. In general, the circuit courts handled cases of diversity of citizenship . . . . In certain limited situations, the circuit courts heard appeals.
Friedman, supra note 53, at 142–43; see First Judiciary Act §§ 9, 11. ↑
inherent authority to manage their dockets and courtrooms with a view toward the efficient and expedient resolution of cases.”). ↑
u. e of the so-called prerogative writs—principally writs of mandamus under the All Writs Act . . . .”); Hyatt v. U.S. Pat. & Trademark Off., 146 F. Supp. 3d 771, 781 & n.25 (E.D. Va. 2015) (noting that the “relief under § 706(1) is functionally identical” to the “writ of mandamus”). ↑