The suggestion that “‘the judicial Power of the United States . . . can no more be shared’ with another branch than ‘the Chief Executive, for example, can share with the Judiciary the veto power’” has an axiomatic ring to it.1 Yet in practice the federal courts are substantially creatures of Congress. Congress has broad authority to steer some disputes into non–Article III tribunals; may tailor federal jurisdiction with few express constraints;2 may withdraw without obvious limit the federal government’s consent to suit; and may, of course, make the law federal courts apply. Against this backdrop, the Supreme Court has continued to recite a nominal limit: if nothing else, Congress “usurp[s] a court’s power to interpret and apply the law”3 when it writes a statute that reads “in ‘Smith v. Jones,’ ‘Smith wins.’”4 Last Term, in Patchak v. Zinke,5 a fractured Court made clear that that principle limits very little. Though no line of reasoning won a majority, six Justices voted to uphold a statute difficult to distinguish from “Patchak loses.” The decision is a defeat for the proposition, typically traced to the Court’s 1871 decision in United States v. Klein,6 that Article III meaningfully constrains congressional power to direct results in pending cases.7 And if the defeat turns out to be final, little harm little foul. Inquiries into the inviolate scope of the judicial power are almost hopelessly indeterminate,8 and Klein’s limit did little to no work in broader constitutional context.
David Patchak’s suit centered on a dispute over a parcel of land, the Bradley Property, in southwest Michigan.9 In 2005, the Secretary of the Interior announced her intent to take the Property into trust for the Match-E-Be-Nash-She-Wish Band of Pottawatomi Indians, which hoped to build a casino there.10 In 2008, Patchak, an unhappy neighbor, sued under the Administrative Procedure Act (APA) to enjoin the action, arguing that the authority invoked — the Indian Reorganization Act11 (IRA) — permitted Interior to take land into trust only on behalf of tribes that were under federal jurisdiction when the statute passed in 1934.12 It soon became clear Patchak’s argument was a winner. By 2009, the Court had embraced this interpretation of the IRA in another party’s suit.13
Notwithstanding that good turn, Patchak still faced jurisdictional obstacles that would take years to litigate. In his first trip to the Court, he overcame objections that he lacked prudential standing and that federal sovereign immunity barred his suit.14 It now seemed obvious Patchak would win on remand — but in the interim, the Band had built its casino.15 While there was no appetite in Congress to overturn wholesale the Justices’ interpretation of the IRA,16 lawmakers took note of the reliance interests now at stake and passed by overwhelming margins the Gun Lake Trust Land Reaffirmation Act,17 a short statute providing:
SEC. 2. REAFFIRMATION OF INDIAN TRUST LAND
(a) In General. — [The Bradley Property] is reaffirmed as trust land, and the actions of the Secretary of the Interior in taking that land into trust are ratified and confirmed.
(b) No Claims. — Notwithstanding any other provision of law, an action (including an action pending in a Federal court as of the date of enactment of this Act) relating to the land described in subsection (a) shall not be filed or maintained in a Federal court and shall be promptly dismissed.18
As Patchak would object on remand, the Act seemed designed to resolve his case and his case alone, without changing otherwise applicable law.19 No other challenges to the land’s trust status were pending; no others could be brought because the APA’s statute of limitations had run.20 All the same, pointing to § 2(b), the district court with custody of Patchak’s suit dismissed it.21 This over his objections that the statute violated the separation of powers; that it violated the Petition Clause; that it extinguished his property interest in his suit without due process of law; and that it amounted to a Bill of Attainder.22 The D.C. Circuit affirmed on each question.23 The Supreme Court granted certiorari on the first and, splintered across five opinions, affirmed.24
Writing for a plurality, Justice Thomas25 found no threat to Article III. Patchak had invoked the ghost of Klein, the confused26 opinion in which the Court announced that Congress may not “prescribe rules of decision to the Judicial Department of the government in cases pending before it.”27 On its face, Klein’s rule-of-decision principle can’t be quite right. The Court has long made clear that, where a judgment isn’t yet final, “a law [that] intervenes and positively changes the rule which governs . . . must be obeyed.”28 Later cases have read Klein more narrowly, Justice Thomas noted, for the proposition that “Congress violates Article III when it ‘compel[s] . . . findings or results under old law’”29 but may freely “change[] the law,”30 even if the change’s application to pending suits “effectively ensures that one side wins.”31 Here, he reasoned, the Act was best read as an exercise of Congress’s jurisdictional authority, and “[s]tatutes that strip jurisdiction ‘chang[e] the law’ for the purpose of Article III, just as much as other exercises of Congress’ legislative authority.”32 To the retort that this change applied to just one case — the dread “Smith wins” hypothetical — Justice Thomas answered that “[n]othing on the face of § 2(b) is limited to Patchak’s case.”33
That isn’t to say the plurality ignored the context in which the statute arose. “[W]e recognize,” the opinion acknowledged, “that the Gun Lake Act was a response to this Court’s decision in Patchak I.”34 Nor did Justice Thomas seriously dispute that Congress’s intervention was “unfair,”35 emphasizing that “the question in this case is ‘[n]ot favoritism, nor even corruption, but power.’”36 After all, the Court largely disposed of Klein challenges predicated on the complaint that “there is something wrong with particularized legislative action”37 two years ago in Bank Markazi v. Peterson.38 In that case, the Court upheld a statute designed to ease the enforcement of a judgment in a single proceeding, which Congress had helpfully identified by its docket number.39 No degree of targeting here, then, was likely to undermine the formal conviction that “when Congress strips federal courts of jurisdiction, it exercises a valid legislative power no less than when it lays taxes, coins money, declares war, or invokes any other power that the Constitution grants it.”40
Of course, Klein itself involved a jurisdiction-stripping statute, applicable in cases where penitent Confederates — seeking to recover property — submitted presidential pardons as proof of loyalty.41 Why a different result for Klein than for Patchak, then? Per the plurality, the core Article III objection in Klein was that Congress had bent jurisdiction in an attempt to “alter[] the legal standards governing the effect of a pardon” — that is, to give pardons an effect other than the one the Constitution assigns — “standards Congress was powerless to prescribe.”42 This was the understanding centered in Bank Markazi, even as that Court defended the “Smith wins” principle. This is also a very thin reading of Klein; the notion that Congress can’t by ordinary legislation amend the Constitution has been secure since Marbury.
Justice Breyer, who joined the plurality in full, wrote separately to assert the relevance of § 2(a).43 No one, he noted, had attacked that provision before the Court,44 and its import was that no suit challenging the Secretary’s action could succeed. That sufficed to distinguish Klein, “a congressional effort to use its jurisdictional authority to reach a result . . . that it could not constitutionally reach directly.”45
Justice Ginsburg, joined by Justice Sotomayor, concurred only in the judgment.46 Without entering Klein’s thicket, Justice Ginsburg read the statute as a withdrawal of consent to suit, a one-to-one response to Patchak I’s holding.47 To this Justice Sotomayor added a standalone concurrence disputing the plurality’s approach to Klein. Concisely, she “agree[d] with the dissent that Congress may not achieve through jurisdiction stripping what it cannot permissibly achieve outright, namely, directing entry of judgment for a particular party.”48
Finally, in an opinion joined by Justices Kennedy and Gorsuch, Chief Justice Roberts dissented.49 The plurality’s defense of formal power over jurisdiction, he objected, was both “undoubtedly correct” and “undoubtedly irrelevant.”50 This because Klein’s functional heart was its insight that “not every congressional attempt to influence the outcome of cases, even if phrased in jurisdictional language, can be justified as a valid exercise of a power over jurisdiction.”51 To hold otherwise — to hold that stripping jurisdiction in a single case works enough of a change to satisfy Klein — would “‘provide[] no limiting principle’ on Congress’s ability to assume the role of judge and decide the outcome of pending cases.”52 Continuing to defend ground largely lost in Bank Markazi, the Chief Justice maintained that “the concept of ‘changing the law’ must imply some measure of generality or preservation of an adjudicative role for the courts.”53 And here, while the “relating to” standard “could theoretically suggest a broader application,”54 in practice, he warned, the plurality “disavow[ed] any limitations on Congress’s power to determine judicial results, conferring on the Legislature a colonial-era authority to pick winners and losers in pending litigation as it pleases.”55
Though only briefly, the dissent also took issue with the suggestion that the Act broke no new jurisdiction-stripping ground. Justice Thomas had invoked Ex parte McCardle56 for the proposition that “Congress generally does not violate Article III when it strips federal jurisdiction over a class of cases.”57 Chief Justice Roberts replied that McCardle addressed only a withdrawal of the Court’s appellate jurisdiction, not a denial of any forum for the litigant’s federal claim.58 Patchak “ha[d] no alternative means of review anywhere else.”59
It’s difficult to say anything wholly new about Klein,60 and in a sense Patchak says nothing new about Klein. Four Justices saw no Article III trouble, four Justices sounded some alarm, and one expressed no opinion.61 But it would be difficult to encourage a client to bring a Klein challenge after Patchak, unless the law read on its face “Smith wins.” The Court hasn’t identified a rule-of-decision violation since Klein;62 it seems increasingly likely that it never will. The question, in Patchak’s wake, is whether the principle is better praised or buried. And the better but the boring argument, probably, is that “Smith wins” is unconstitutional when unconstitutional for an articulable reason — but not always, no thanks to Klein, and not on inchoate separation of powers grounds. In that sense, the outcome in Patchak is a victory for the suggestion that “the purposes of the separation of powers are too general and diverse to offer much concrete guidance” where no particular constitutional guarantee is at stake.63 The dissent may be right that a sharp division of powers “secure[s] individual freedom,”64 but the energy expended over more than a century clarifying Klein is a cautionary tale about “treat[ing] a broad purpose of the separation of powers — safeguarding liberty — as if it were a judicially manageable constitutional standard.”65
The waste is best measured relative to the stakes, and if the fall of the rule-of-decision principle sounds troubling, it bears emphasizing how little work is left for Klein to do once other prohibitions on “Patchak loses” are accounted for. Consider a few ways of articulating what was wrong with the Act. Was it that it singled Patchak out for differential treatment? The Equal Protection Clause, that great guarantee of legislative generality,66 only requires that lawmakers draw lines rationally, and choosing the Band’s economic certainty over Patchak’s quiet enjoyment was at least minimally rational. Beyond this, as aired more fully in Bank Markazi, there is no absolute or freestanding generality requirement, “or else we would not have the extensive jurisprudence that we do concerning the Bill of Attainder Clause . . . [or] a case [holding] that Congress may legislate ‘a legitimate class of one.’”67 Rationality is a deeper rule-of-law value than generality in any event,68 and a judgment that an otherwise applicable rule would be a bad fit for certain facts makes for a valid subrule.69 “Smith wins on our say-so” is likely unconstitutional; “Smith wins for good reason” is lawmaking.70
Alternatively, was the problem with “Patchak loses” that it made the legal and factual judgments of a non–Article III body — Congress — conclusive on Article III courts, leaving them “[no] role . . . beyond that of stenographer”?71 As to questions of law, the objection evokes anxieties about deference to administrative agencies72 and, in the habeas context, state courts,73 settings in which federal courts are asked to give effect to something other than their own view of the law. The concern recalls, too, Henry Hart’s remark that “if Congress directs an Article III court to decide a case, I can easily read into Article III a limitation on the power of Congress to tell the court how to decide it.”74 An extended discussion of those considerations is (wildly) beyond the scope of this comment.75 But it bears interrogating whether the Article III question is often, or for that matter ever, squarely presented. Chevron deference can just as well be articulated as a rule about the scope of substantive agency authority,76 while limits on post-conviction relief can be defended as constraints on the remedial — rather than interpretive — power of the federal courts.77 And Patchak, it’s worth remembering, did not press before the Court the argument that § 2(a) “unlawfully impose[d] ‘Congress’s own interpretation of the IRA’ on the federal courts”78 — probably because, per Justice Breyer, Congress didn’t so impose. Given how difficult it is to find a statute flatly requiring that courts apply something other than law, perhaps Klein forbids such laws in the sense in which Saint Patrick drove the snakes out of Ireland.
With respect to questions of fact, it bears emphasizing that the Court has never suggested federal courts must be absolutely free to do all their own factfinding (the history of agency adjudication is very much to the contrary79), and where it has struck down arrangements that abridge the courts’ factual judgment, the ground has usually been due process rather than Klein80 — the objection that an individual has been deprived of a constitutionally protected interest without an opportunity to raise relevant constitutional arguments.81 This concern tends to flow into the last intuitive critique of the Act: Namely, is the problem with “Patchak loses” that Patchak’s merits argument was a clear winner, and that in kicking him out of court, Congress denied him a remedy to which he was constitutionally entitled? If anything is surprising about Patchak, it’s the dearth of discussion — especially measured relative to the concern’s prominence at argument82 — about this longstanding jurisdiction-stripping anxiety.83 And certainly the plurality account includes few express safeguards.84 But Patchak needn’t cast doubt on the notion that, “when substantive constitutional rights exist, the Constitution requires that some court have jurisdiction to provide sufficient remedies to prevent those rights from becoming practical nullities.”85
Recall that underneath the procedural wrangling, Patchak was seeking an injunction to redress a violation of statutory law. Whatever the list of constitutionally necessary remedies,86 it probably doesn’t include that one: Congress’s power to preclude review of bare statutory violations is fairly well established.87 And Congress didn’t succeed, contra Chief Justice Roberts’s concern,88 in insulating any separation of powers violation from review. A majority of the Court saw no substantive separation of powers problem with § 2(b), and any failure to review § 2(a) on the merits is better attributed to Patchak’s litigation decisions than to Congress or the Court. In that respect Patchak’s bottom line is readily defended against the charge that it represents a new “highwater mark of legislative encroachment on Article III.”89 Neither in Patchak nor in broader constitutional context does Congress’s latitude to authorize constitutional violations rise and fall with Klein.90
In each of these respects, Patchak was typical of cases that appeal to Klein: something feels wrong in general, but nothing, on close examination, is convincingly wrong in particular. The same objection could be levied in a range of contexts that, under precedents old and new, require freeform separation of powers inquiries, largely unmoored from text or history.91 Here, the Court refused to be drawn into the morass, all but announcing it would never again be tempted by this facet of Klein. Perhaps the rule-of-decision principle is worth keeping on the nominal books in case some future suit brings the Article III objection into sharper relief. But if Patchak marks the last impassioned invocation of the rule, little of constitutional significance will have been lost.