Concept
The Michigan v. Long Presumption
The “adequate and independent state ground”
The Michigan v. Long Presumption
The independence requirement turns on a hypothetical: would the state ground still hold if the Supreme Court corrected the state court's view of federal law? That question is easy to answer when the state court says which law it relied on, and hard when it does not. State judges write opinions to decide the case before them, and many opinions cite the state constitution, a federal clause, and a mix of state and federal precedent without saying whether the state provision stands on its own footing or was read to mirror the federal one. In Michigan v. Long the Michigan Supreme Court held a vehicle search invalid under both the Fourth Amendment and the state constitution. The opinion gave the Court reason to suspect that the state ground depended on federal law, yet it also expressly invoked both constitutions as the basis for the decision.1 A state ground and a federal ground sat side by side, and nothing on the page showed whether the state ground could survive alone.
Before 1983 the Court had no single way to resolve that uncertainty. It had sometimes sent the case back to the state court for clarification, sometimes examined state law itself for signs of independence, and sometimes declined jurisdiction on the theory that doubt should be resolved against review.2 Long rejected each of those approaches and put a presumption in their place. Where it is unclear whether a state decision rests on adequate and independent state grounds, the Court presumes that it has jurisdiction and that the decision rests on federal law, unless the state court clearly and expressly states that its judgment rests on state-law grounds independent of federal law.3 The allocation of the burden follows from who holds the information. The state court knows which law drove its decision and can say so in a sentence, whereas the Supreme Court can only infer that intent from an opinion that was not written with the inference in mind. The Court also reasoned that the presumption avoids the cumbersome alternatives of remanding or surveying state law, and that it leaves state courts free to build independent bodies of their own law.4
The presumption operates only in cases of uncertainty, where the state ground fairly appears to depend on, or to be interwoven with, federal law, and a state court can overcome it by a clear statement that its ground is independent. The subsections below follow the doctrine in four steps. The first covers the methods the Court used before Long and the reasons it rejected each one. The second states the plain statement rule itself. The third examines what the rule asks of state courts and how they have answered, and the fourth presents the criticisms of Long and the remand practice the Court has not entirely abandoned. The presumption has its sharpest effect when a state court has given more protection than federal law requires, because a state ground that adds rights is adequate and only its independence is in doubt.5 The first subsection, Approaches Before Long, begins with the methods the presumption replaced.
Approaches Before Long
When a state court's opinion leaves open whether its state ground stands on its own footing, the independence inquiry has nothing clear to work from. Before 1983 the Supreme Court met that uncertainty with a shifting assortment of responses, and in Michigan v. Long the Court admitted that it had never developed a satisfying and consistent approach to the problem.6 Three methods stand out. Each is worth learning together with the objection Long raised against it, because the presumption Long adopted makes sense only as an answer to those objections.
The methods, in the order the Long opinion surveyed them, were these:
- Dismissal. Under the strict view of cases like Lynch v. New York ex rel. Pierson, the Court dismissed whenever the ground of decision was at all unclear, on the theory that the party invoking jurisdiction had failed to show that it existed.7 The flaw is that an opinion resting primarily on federal grounds then goes uncorrected whenever the independence of a possible state ground cannot be seen from the face of the opinion, which leaves the need for uniform federal law unmet.8
- Vacatur or continuance. In cases such as Herb v. Pitcairn, the Court vacated the state judgment or continued the case so that the state court could clarify the basis of its decision.9 The flaw is that this practice caused delay and reduced judicial efficiency, and it placed on state courts the burden of demonstrating whether the Supreme Court had jurisdiction.10
- Examination of state law. In more recent cases such as Texas v. Brown, the Court looked at state law itself to decide whether the state courts had used federal law merely to guide their application of state law or as the actual basis for the result.11 The flaw is that the Court had to interpret state laws with which it was generally unfamiliar and which the parties often had not discussed at length.12
The second method shows both how the practice worked and where it broke down. In Minnesota v. National Tea Co. the Court vacated a state judgment whose basis was unclear and sent the case back for clarification, and if the state court then said that its ruling rested on an independent state ground, the matter ended there.13 The method depended on an answer that might never come. In Dixon v. Duffy the Court continued the case twice to obtain clarification, and none was forthcoming, because the California court informally advised counsel that it doubted its jurisdiction to make such a determination; the Court then vacated the judgment and remanded.14 A state court that does not answer leaves the Supreme Court where it began, with added delay.
The three methods also failed as a group. The Court called the resulting ad hoc practice "antithetical to the doctrinal consistency that is required when sensitive issues of federal-state relations are involved."15 No method had been chosen as the preferred one to the exclusion of the others, so the disposition could vary from case to case. In Oregon v. Kennedy the Court declined to remand for clarification even though the state decision rested in part on a state case, because it concluded that the state case itself rested on federal grounds.16 A litigant facing an unclear opinion could not predict whether it would be dismissed, sent back, or read by the Justices themselves, since the outcome turned on which tool the Court happened to pick. A single rule replaced all three, and the next subsection states it.17
Long did not abolish requests for clarification altogether. The opinion acknowledged that clarification may sometimes be necessary or desirable and said that the Court would not be foreclosed from taking appropriate action.18 Before Long the Court remanded in three situations: where changes in state law after the state court's judgment might affect the disposition, where the record did not clearly reflect the facts underlying the federal issue, and where the actual ground of decision was unclear. The first two situations survive, while the third has been dispensed with, though only sometimes.19 The Court took the remand course in the litigation over the 2000 Florida presidential recount, finding considerable uncertainty about the precise grounds of the Florida court's decision, even though Long would have allowed it to review.20 Students lose this point when they assume Long forbids every remand, because the presumption governs the ordinary unclear-ground case while clarification stays available in exceptional circumstances.
The Plain Statement Rule
Michigan v. Long replaced the Court's patchwork of responses with a rule that has three working parts: a trigger that says when the rule applies, a presumption the Court then adopts, and a plain statement by which a state court can defeat that presumption. The trigger comes first because the presumption does nothing until it is met. Under the holding, when a state court decision "fairly appears to rest primarily on federal law, or to be interwoven with the federal law," and when the adequacy and independence of any possible state-law ground "is not clear from the face of the opinion," the Court accepts as the most reasonable explanation that the state court decided the case as it did because it believed federal law required it to.22 Both conditions must be present. The first looks at what the opinion is built from, asking whether federal precedent is doing the work or whether state and federal law are so mixed that the state holding cannot be separated from the federal one. The second asks whether the page itself answers the independence question.
Confining the inquiry to the face of the opinion is deliberate. Examining state law directly would require the Court to interpret statutes and decisions it knows little about, often without full briefing from the parties, and the approach adopted in Long avoids that task in most cases.23 An opinion built on a body of state decisions and statutes therefore fails the first condition, and an opinion that says plainly that it stands on state law leaves nothing unclear for the second. The face of the opinion is also the one thing the state court fully controls when it writes, which is why the rebuttal described below operates there.
When the trigger is met, the Court presumes that it has jurisdiction and that the decision rests on federal law, unless the state court clearly and expressly states that its judgment rests on state-law grounds independent of federal law.24 The presumption is a prediction about what the state court would do if the Court corrected its federal reasoning. A court that believed federal law required its result would lose the premise of its decision once told that the premise was wrong, so a reversal on the federal question could change the judgment, and a ruling on that question would be more than an advisory opinion. The Court said as much: the plain statement requirement does not authorize advisory opinions, because the Court merely assumes there are no independent state grounds when the opinion does not make them clear and it fairly appears that the state court rested primarily on federal law.25
Placing the burden on the state court follows from who holds the information. The state court knows which law drove its decision and can say so in a sentence, whereas the Supreme Court can only infer that intent, and each earlier alternative (dismissing, vacating for clarification, or surveying state law) imposed costs on the Court, on the state courts, or on the uniformity of federal law.26 The presumption also removes a temptation. A state judge who wants to protect a defendant might be inclined to describe the result as compelled by federal law, which shifts responsibility for an unpopular decision to the federal government, and a rule that rewards an open statement of state-law reliance asks the state court to own the protection it recognizes and exposes that choice to the state political process.27 The design also leaves state courts free to develop independent bodies of state law, since they need only say that they are doing so.28
The plain statement is the state court's answer to the presumption, and it takes one of two forms depending on how the court used federal law. A court that relies on federal precedents just as it would on the precedents of any other jurisdiction need only make clear, in its judgment or opinion, that it cites the federal cases for guidance and that they do not themselves compel its result.29 A court that gives alternative holdings can instead indicate clearly and expressly that its decision also rests on bona fide separate, adequate, and independent grounds, and the Court will not undertake to review the decision.30 Either way, the statement supplies the missing information on the face of the opinion, which is the one place the Court looks.
The statement must be genuine. The rule excuses review only for "bona fide" separate, adequate, and independent grounds, and the Court has long held that it is incumbent on the Court to decide for itself whether an asserted non-federal ground independently and adequately supports the judgment.32 A recital that a state ground exists therefore carries no weight if the opinion's actual reasoning shows that the state provision was read only to mirror the federal one, or if the asserted ground could not support the judgment, and the Court may disregard such a statement.33
Long itself shows the trigger and the presumption working together. Michigan officers searched the passenger compartment of Long's car after a roadside stop and found marijuana, and the Michigan Supreme Court reversed the Court of Appeals, which had upheld the search, holding the interior search invalid and suppressing the marijuana from both the interior and the trunk.34 The Michigan opinion referred to the state constitution twice, once in a footnote citing both constitutions and once in its concluding sentence, which said the search was barred by the Fourth Amendment and by the state constitution's search provision.35 Long argued that Michigan courts had given greater search-and-seizure protection under the state constitution than the Fourth Amendment provides, so those references supplied an independent state ground.36
The Court's answer ran through the face of the opinion. Apart from its two citations to the state constitution, the Michigan court relied exclusively on its understanding of Terry v. Ohio and other federal cases, cited not a single state case for its holding, and declared the search unconstitutional because the lower court had erroneously applied the principles of Terry to the interior of the vehicle.37 The constitutional references in no way indicated that the decision was independent of the court's reading of federal law, and even if Michigan's constitution had been read to give independent protection for rights also secured by the Fourth Amendment, it fairly appeared that the court rested primarily on federal law.38 With no plain statement that the decision rested on an adequate and independent state ground, the Court found that it had jurisdiction, and it went on to hold the search justified under the principles of Terry.39 Because Michigan had added to motorists' rights, adequacy was never in doubt, and independence alone decided whether the Court could hear the case.40
Foster v. Chatman shows the rule still operating in the same way, with a procedural ground in place of a substantive one. The Georgia courts there had examined the defendant's jury-selection claim under Batson while deciding whether res judicata barred it, which cast doubt on the independence of the procedural ground, and because no clear statement established that independence, the Court followed Long and confirmed its jurisdiction.41 The rule applies whether the disputed state ground is substantive, as in Long, or procedural, as in Foster.42
The presumption depends on the trigger, so it does not reach every opinion that mentions federal law. If the decision does not appear to rest primarily on, or to be interwoven with, federal law, no plain statement is needed for the state ground to bar review.43 Students lose points when they apply the presumption to an opinion that merely mentions federal law in passing, because the first question is always what the opinion rests on.
The roadside-search opinion and the ordinance opinion differ in what their reasoning is made of. Where federal decisions carry the analysis, the page must say that the state ground stands alone; where state decisions carry it, the authorities the court chose already say so.
Step two does the real work, and a bare citation to the state constitution does not take the place of the statement in step four, as the two citations in Long's own opinion show.
State Courts After Long
The plain statement rule tells the Supreme Court when to presume that it has jurisdiction, but the rule is aimed at state judges, because they write the opinions it reads and so decide which side of the line a decision lands on. What the rule asks of them depends first on a question that comes before it: whether the added protection a state court wants to give comes from federal law or from state law.
The Supreme Court's reading of the federal Constitution fixes one meaning that every court must apply, and a state court has no power to move that meaning in either direction. A state court that construes the Fourth Amendment to give individuals fewer rights than the Court has prescribed is wrong, and so is one that construes it to give more.46 In Oregon v. Hass the Court said that a state may impose greater restrictions on police activity as a matter of its own law than the Court thinks federal standards require, but may not impose those greater restrictions as a matter of federal constitutional law when the Court has specifically refrained from imposing them.47 The reason follows from what the federal guarantee is for. If each state court could enlarge the Fourth Amendment, the same clause would mean different things in different states, and the need for uniform federal law that Long itself stressed would go unmet.48 Supreme Court interpretations therefore operate as a floor, and the only route to a higher ceiling runs through the state's own constitution and statutes.49
That division tells a state court where its added protection must come from. The court has to rest the protection on state constitutional or statutory law, and its interpretation of that law has to be entirely independent of federal law.50 Michigan v. Long then adds a drafting requirement for opinions that lean on federal decisions or interweave state and federal law, because the mere existence of a state ground no longer insulates the judgment in that setting. A court that cites federal precedents the way it would cite those of any other jurisdiction must make a plain statement that the cases are used only for guidance and do not compel its result, and a court that gives alternative holdings must indicate clearly and expressly that its decision also rests on bona fide separate, adequate, and independent state grounds.51 The statement is how a state court places a decision on the state side of the line, since it supplies on the face of the opinion what the Supreme Court would otherwise have to guess.
When a state court omits that placement, the consequences fall on its more protective rulings. A ground that adds rights is presumably adequate, so independence is the only open question, and the presumption answers that question against the state court whenever the opinion leaves it unclear.52 Long itself shows the sequence. The Michigan Supreme Court had given motorists more protection against vehicle searches than the Fourth Amendment, as the Supreme Court understood it, required, and the Court found no independent state ground, held that the state court had erred in reading the Fourth Amendment to grant that extra protection, and upheld the search under Terry.53 The presumption therefore makes it easier for the Court to review and reverse decisions that over-read federal rights. Students lose this issue when they assume a state court may be more generous in construing the federal Constitution itself, because the freedom to add rights belongs to state law alone.
The Long facts and this hypothetical differ in one variable, the place the opinion says the protection comes from, and that variable alone decides whether the Court may reach the Fourth Amendment question. The same variable decides who can correct the ruling. A decision grounded in federal law can be reversed by the Supreme Court and cannot be undone by state voters, while a decision grounded in state law is beyond the Court's reach and stays open to correction through the state political process, where state judges are more accountable and state constitutions are typically easier to amend.55 A state judge who would prefer not to answer for an unpopular protection can no longer attribute it to federal compulsion without inviting reversal, which pushes the judge to claim the protection openly under state law.
State courts accepted the invitation. The Oregon Supreme Court issued an opinion on the day Long was decided declaring that, when it cites federal opinions in interpreting Oregon law, it does so because it finds their views persuasive and not because it considers itself bound by its understanding of federal doctrine.56 The court expected counsel and judges to refer to federal decisions even when debating an undecided state-law question, so the declaration worked as a standing plain statement covering its later citations. It was written in a case the Supreme Court had remanded after finding no adequate and independent state ground, and on remand the intermediate court specified that its decision rested on state law before the Oregon Supreme Court affirmed. Other states have echoed Oregon's sentiments.57 Long's own state court answered in the same direction. On remand the Michigan Supreme Court adhered to its view that the search violated the state constitution, so Long's conviction was overturned and the Fourth Amendment holding did not change the outcome for him.58 Whether that result shows the presumption producing advisory opinions is a criticism taken up in the next subsection.
Steps two and three check each other, because the statement must be genuine. A recital that the state ground stands alone carries no weight if the opinion's reasoning shows that the state clause was read only to mirror the federal one, and the Court may disregard it.59
Criticism and Remand Practice
The presumption announced in Michigan v. Long governs state opinions that leave the independence of a state ground unclear, yet it has drawn sustained criticism, and the Court has not treated it as an inflexible command.62 The objections fall into three groups. The first is that the presumption produces federal rulings that change nothing for the parties. The second is that it extends the Court's reach to cases in which a state has simply protected its own citizen more generously than federal law requires. The third is that the rule was built to reach a particular result and has been applied unevenly since. Each is an argument advanced by a dissent, a partial concurrence, or later critics, and each draws a reply, so they are presented here as positions to weigh.
The advisory-opinion criticism starts from a sequence that the presumption makes possible. When an opinion looks primarily federal and shows no clear state ground, the Court takes jurisdiction and decides the federal question.63 If it reverses and remands, the state court remains free to reinstate its earlier judgment by stating, more clearly than before, that the judgment rests on state law.64 As long as the newly clarified ground is also adequate, meaning that it does not conflict with federal law, the reinstated judgment binds the parties exactly as the first one did, so the federal ruling has altered no one's rights or obligations.65 That result collides with the premise of the whole doctrine, which is that the Court corrects judgments that wrongly decide federal rights and may not issue opinions that leave the same judgment standing.66
Michigan v. Long itself supplies the standard example. The Court asserted jurisdiction, held the interior search justified under Terry, and reversed; on remand the Michigan Supreme Court adhered to its view that the search violated the state constitution, Long's conviction stayed overturned, and the Fourth Amendment ruling did not influence the final disposition.67 The Court has sometimes invited exactly this response. In Florida v. Powell it took care to say that its decision did not trench on the Florida Supreme Court's authority to impose additional protections against coerced confessions under the state constitution, and in Marmet Health Care Center v. Brown it invited the state court on remand to consider whether its decision could rest on purely state contract principles.68 An invitation does not always produce a reinstatement, because on remand in Powell the Florida Supreme Court reconsidered its state-law ground and concluded that the Florida constitution conferred no rights beyond those the U.S. Constitution provides.69 State courts in Iowa and New York, in Racing Association of Central Iowa v. Fitzgerald and People v. P.J. Video, Inc., did take the reinstatement route after reversals. Justice Stevens, dissenting in Washington v. Recuenco, objected that the Washington Supreme Court could reinstate the same judgment on remand because it chose, as a matter of state law, to give the petitioner more rights than federal law gave him.70
Defenders of the presumption answer that whether a ruling is advisory should not depend on what the state court does afterward. A ruling on a federal question that the Court found fairly interwoven with state law, and therefore potentially dispositive, is not rendered advisory because later proceedings leave the outcome unchanged.71 Long itself said that the plain statement requirement does not authorize advisory opinions, since the Court merely assumes there is no independent state ground when the opinion does not make one clear and it fairly appears that the state court rested primarily on federal law.72 The disagreement is therefore over timing. The critics measure the ruling by its effect after the state court has had a second chance to speak, while the defenders measure it by what the opinion showed when the Court acted, and a state court can control that second measure by writing the plain statement the first time.
The second criticism comes from Justice Stevens, joined by Justice Brennan, in dissent. Stevens found the Court's rejection of the remand and state-law-examination approaches defensible, but he would have kept the historical presumption that the Court declines jurisdiction when it is unclear whether an independent state ground supports the judgment, the rule stated in Lynch v. New York ex rel. Pierson.73 His objection rested on what federal review is for. Michigan had arrested one of its citizens and its supreme court had decided to release him, no one claimed that the state had mistreated him, and the outcome therefore offended no federal interest, because Michigan had simply given its citizen greater protection than another state might give or than the Court might require nationwide.74 In his view the Court's primary role in reviewing state courts is to make sure that persons who seek to vindicate federal rights have been fairly heard.75 Overprotection in one state, he added, does not freeze the law or set a precedent for the nation, since it restricts police behavior in one state in the interest of individual liberty.76
Stevens also attacked the majority's only stated justification, the need for uniformity in federal law. That need is equally present when a state ground is plainly independent and adequate, so it proves too much, and he called it an "ungovernable engine."77 The Court's power is to correct wrong judgments and not to revise opinions, and he asked whether the Court would show respect for Finland by convening a special sitting to declare that its decision to release an American citizen rested on a misunderstanding of American law.78
The majority replied that this view reaches well beyond cases with possible state grounds, since taken at face value it would bar review of a decision resting entirely on federal grounds whenever no federal right is endangered. The majority also observed that state courts handle the great bulk of criminal litigation and necessarily create a considerable body of federal law when they apply federal constitutional standards, which gives the Court its own interest in how they do it.79 Supporters add that overprotection is less harmless than the dissent assumed. Expanding individual liberties can impede law enforcement goals that many would support, and a state court that attributes its result to federal compulsion shifts political responsibility away from itself, whereas the Long rule requires the court to take responsibility openly and exposes the result to correction through the state political process.80 Justice Sotomayor has since taken up Stevens's position, objecting in Kansas v. Carr when the Court reversed a Kansas Supreme Court decision that had vacated three death sentences because of procedural errors the Kansas court viewed as violating the Eighth Amendment.81
Justice Blackmun joined Justice O'Connor's opinion for the Court and filed a separate concurring opinion.82 Justice Stevens dissented with Justice Brennan joining, and Justice Marshall took no part in the decision.83 The third criticism is that the Court defined its jurisdiction expansively in a case in which it plainly wanted to reach the merits, and critics doubted that it would stay true to its goal of a consistently applied body of law on state grounds of doubtful independence.84 Later decisions give that doubt some support, because the Court has occasionally returned to vacating and remanding for clarification and has sometimes declined to apply the presumption even where a state ground arguably lacked independence from a federal claim.85 Other decisions show the framework still in use. In Foster v. Chatman the Georgia courts had considered the jury-selection claim while evaluating res judicata, and with no clear statement establishing the independence of the procedural bar, the Court followed Long and confirmed its jurisdiction.86
The remand practice that the Court never fully abandoned has an express home in Long. A footnote reserves the point: there may be circumstances in which clarification is necessary or desirable, and the Court will not be foreclosed from taking the appropriate action.87 The categories of remand that existed before Long show what those circumstances look like. The Court asked for clarification where state law changed after the state court's judgment in a way that might affect the case, where the record did not clearly reflect the facts underlying the federal issue, and where the actual ground of decision was unclear; the first two survive, and the third has been dispensed with only sometimes.88 The reason to keep the remand available is that the presumption exists to avoid guesswork about a state court's intent, and when the guesswork is unusually hard, a brief return to the state court may resolve the uncertainty more reliably than a presumption does.
The 2000 presidential election litigation shows the remand option in use. In Bush v. Palm Beach County Canvassing Board the Florida Supreme Court had held, under state law, that counties had an extended but limited time to submit recount returns, and the Republican candidate petitioned for review even though the decision rested on state law.89 The petition argued that the state ground was inadequate because it rested on state constitutional principles in tension with Article II, section 1, which assigns the manner of choosing electors to a state's legislature, and that it was not independent because it reflected a novel reading of state law shaped by 3 U.S.C. § 5.90 Whether either federal question was truly present was unclear from the opinion, since the Florida court may have relied solely on state statutes and cited § 5 only in a footnote that may have played no role.91 Long would have allowed review on that record, but the Court instead found "considerable uncertainty as to the precise grounds for the decision" and declined for the time being to review the federal questions.92 The remand halted the recount without any ruling on the merits, and a day after the Florida court clarified its decision, the Court held the recount procedure unconstitutional in Bush v. Gore.93
Placing Long beside the Florida litigation shows the difference between a default and a command. In Long the opinion's reasoning ran through federal cases, no plain statement appeared, and the presumption supplied jurisdiction. The Florida opinion was written on state-law foundations, and the doubt was whether federal provisions had shaped a state-law holding. In both the independence of the state ground was open to question, and the lesson is that the presumption directs the usual answer without removing the Court's discretion to ask. Students lose points when they treat the presumption as an absolute rule that forbids every remand, because Long itself reserved clarification for appropriate cases.94
Step four is the one most often left out, and step five is where the critics' objection and the defenders' reply meet: the critics see a ruling that changed nothing, and the defenders see a ruling that was potentially dispositive when the Court made it.