Precedent and the Semblance of Law
settled versus right: A Theory of precedent. By Randy J. Kozel.[1] Cambridge University Press. 2017. PP. x + 180. $99.99 (hardcover), $34.99 (paper).
Stephen E. Sachs[2]
Americans disagree about the Constitution. Often they disagree deeply. The Supreme Court sometimes purports to “resolve [an] intensely divisive controversy,” calling both sides “to end their national division”—but it rarely succeeds.[3] Five-four majorities shift with the political winds, whether on abortion, the death penalty, or campaign finance.[4] At the level of theory, the disagreement is just as bad: different Justices deploy different arguments from originalism, pragmatism, common-law constitutionalism, or other approaches still more exotic or idiosyncratic (assuming they draw on any theories at all). Justice Scalia once suggested that his colleagues didn’t even agree “on the basic question of what we think we’re doing when we interpret the Constitution.”[5] If the leaders of American law are so torn on these questions, are we sure that there really are right answers? Does America truly have any constitutional law?
Into this divided world comes Randy Kozel’s Settled Versus Right. Like its author, the book is insightful, thoughtful, and kind; it’s deeply committed both to describing and to improving the world that it sees. Its argument and approach are worth thinking deeply about, even for—indeed, especially for—those who might disagree with its conclusions. But despite its upbeat tone, the book paints rather a dark picture of current law and the current Court. It depicts a society whose judges are, in a positive sense, lawless—not because they disregard the law, but because they are without law, because they have no shared law to guide them. What they do share is an institution, a Court, whose commands are generally accepted as law by lower courts (and, at least sometimes, by the Justices themselves).
So Settled Versus Right makes the best of what we’ve got. Instead of a legal system that shifts with each new appointment, it seeks a body of precedent that’s stable and impersonal (p. 11), the way law ought to be. It hopes to reorient judicial culture around a “second-best” stare decisis (pp. 100-103): one that deliberately enforces precedents that are “badly reasoned”[6] or “flagran[tly]” wrong, so as to avoid reopening our disagreements on interpretive method or theory (pp. 103-104). If we can’t agree on rules of law, or on the theories that ought to generate these rules, at least we can compromise on preserving whatever our legal institutions have done before.
Though this compromise is well-argued, it may fail to satisfy both sides. On the one hand, if we do still have some constitutional law, this law may take a view on our rules of stare decisis—a view that might not be the same as Kozel’s. The second-best theory is openly revisionary, rather than merely trying to capture our existing legal practice (including our practice of precedent). Without a firm grounding in existing law, its pursuit of stability and impersonality may yield a system that’s more lawlike than lawful. It may be a mere semblance of law, the way Kant saw “love of honor and outward propriety” as mere “semblances of morality,”[7] sharing only an obedience to “strict laws of conduct for their own sake.”[8] On the other hand, if our disagreements really have deprived us of any real law to apply—leaving judges to advance their values as best they can—then there are many other important values to consider, beyond maintaining the system’s lawlike nature through stable and impersonal rules. A judge might wish to use her office to protect the environment, prevent violent crime, and so on; the second-best theory can’t tell us where stability and impersonality rank on that list.
Perhaps there’s another way forward. Rather than patching up a broken system, we might use Kozel’s analysis to illuminate ways of deepening our existing areas of agreement on rules and theories of law. In this project, stare decisis might aid us, if we see it as a fallback and not as a foundation-stone. Precedent often serves as a provisional doctrine, supplying us with stand-in answers when we’re unsure of the real ones—requiring us to act as if a court has decided a case correctly, but not necessarily to treat the court’s decision as itself establishing the standard of correctness.[9] Perhaps precedent is supposed to be a mere semblance; perhaps that’s its proper role, letting us debate the contours of our actual law without requiring a thousand judicial flip-flops along the way. If so, then expanding our agreement on the law might indeed involve a cultural change, but of a somewhat different sort. We ought to take the law rather more seriously, and courts and judges rather less so. Once we do, we might find that our world is a lot less lawless than we think.
i
The problem with “[t]he prevailing approach to precedent,” Kozel writes, is that it presumes “a greater degree of agreement about constitutional theory than actually exists” (p. 6). Our legal system features “pervasive disagreement over constitutional interpretation,” without “anything approaching consensus” as to “the proper method” (p. 6). This disagreement is often “principled” (p. 17): the disputants are reasonable people arguing in good faith, and we can’t show (on shared standards) that one side is obviously correct. But even if we could, it wouldn’t help. The combination of legal disagreements, party alignments, multimember courts, and staggered appointments means that our system will inevitably be “characterized by interpretive pluralism” (p. 98).
The unfortunate result is that our constitutional practice will “vacillat[e]” along with “changes in judicial personnel” (p. 18). Whether, say, Citizens United v. FEC[10] “remains the law of the land” will turn on whether new appointments make “the four-justice dissent into a five-justice majority,” even as “the Constitution itself will not have changed a bit” (p. 5). In what Kozel describes as “the worst-case scenario”—a scenario that seems not so far off—the very “ideal of constitutional law as stable and enduring” might “give way to the notion that judicial identity and legal meaning are one and the same,” and that the law is whatever judges say it is (p. 99).
In its extreme form, this disagreement could undermine the case that there is any constitutional law on the matters in dispute. The judges who write concurrences and dissents seem to think they’re making legal arguments, and that some of their arguments are right—that one side’s answer is legally better than the other’s. But how is this possible? On the most commonly accepted account of law, what makes a legal rule valid—or a legal answer good—is the actual practice of Americans and their attitudes thereto.[11] If Americans are so fundamentally divided that they lack any unified legal practice, then how can one side have the right of the argument legally, and not just prudentially or morally?
This is what Matthew Adler calls “The Puzzle”: on the “contested questions” of constitutional theory, are there any “legally correct answers”?[12] If there’s some deep consensus at work, why don’t we see more evidence of it in practice?[13] And if there isn’t one, why do judges act like they’re disagreeing about the law (say, “[w]hether there really is a federal power to regulate marijuana”), when their disagreement itself shows the questions at issue to be “legally indeterminate”?[14] As Adler asks, “[w]hat would it take for one side in this debate to be correct and the other incorrect? How is that even possible given the very fact of debate?”[15] This, in turn, poses what he calls “the ‘Meta-Puzzle’: Why has the Puzzle failed to bother constitutional theorists? . . . Why have theorists not spent more time reflecting upon the status of the arguments they are making?”[16]
One way to explain all this disagreement is to explain it away, as a giant mistake—to adopt an “error theory” of constitutional law.[17] On almost all legal questions, Brian Leiter notes, American society displays “massive and pervasive agreement”: most disputes stay out of court, most filed cases never reach trial, and so on.[18] For these disagreements, there really are right legal answers, grounded in the shared practices of the American legal community. But, Leiter argues, there are also some deep questions on which “the practice of officials breaks down, and the ‘law’ is up for grabs.”[19] To the error theorist, studying such questions is like studying astrology or the Loch Ness monster: the field is useless, because its subject doesn’t really exist.[20] But when judges proceed to address these questions, they don’t always notice that the ground has run out from underneath them. Misled by their experience of agreement elsewhere, they keep right on going, as if legal answers to the deep questions are still out there.[21] So long as the judges don’t look down, they can keep running, Wile-E.-Coyote-like, over empty air.
Kozel’s response to these alleged gaps in the law is to start building bridges. Whatever the right answers might be—if there are any—we can at least find the answers that courts gave in the past. Enforcing those answers today, through an enhanced doctrine of precedent, makes the practice of American law more stable over time, and it prevents the results in constitutional cases from varying with the personalities of the individual judges deciding them.
At the same time, Kozel would shape the enforcement of past precedent to present circumstances. For stare decisis to “make good on its promise” (p. 103), it has to be made safe for pluralism. We need a doctrine that’s immune to “disputes over interpretive philosophy” and capable of “transcend[ing] methodological disputes” (pp. 103, 129). That means “limiting or excluding” any features of stare decisis that might depend “on methodological and normative commitments that vary from judge to judge” (p. 107). Some traditional features of the doctrine, like “procedural workability, factual accuracy, and reliance expectations,” can be applied just as well “by judges across the methodological spectrum” (p. 103). Yet others, like “jurisprudential coherence, flagrancy of error, and a precedent’s perceived harmfulness,” inevitably vary with a judge’s legal commitments and will have to be dropped (p. 104).
This may not be the best doctrine of stare decisis, or the one we’d choose if we agreed on the legal underpinnings. But a second-best theory of stare decisis, Kozel argues, is better shaped for the “second-best world” in which we live (p. 100).
ii
The second-best theory is a reasonable response to a legally divided world. But is it a lawful response? “Stare decisis is, at base, a legal doctrine” (p. 171): it claims to generate legal reasons, telling judges and officials to use their powers in particular ways. If a judge is going to decide a case by relying on stare decisis—say, imprisoning someone for helping to publish a political book within 60 days of an election[22]—then there needs to be legal warrant for that decision. This suggests that the legal system may already have rules on point.
Yet it also presents a dilemma. If there’s already law on point, then a second-best theory might be inconsistent with that law. No matter how lawlike a theory might be, or how helpful in smoothing over disagreements, a judge might still be legally forbidden to use it. Alternatively, if there’s no law on point—if the use of second-best stare decisis really is an open choice for judges—then there might be many other reasons, having nothing to do with precedent or courts, for those judges to choose differently. Stare decisis can’t “shake off the trappings of constitutional theory,” producing a view-from-nowhere “that operates independently of any interpretive methodology” (p. 61), because it needs those trappings to assert a general claim of priority over other values. It can’t stand outside the law and act with the force of law at the same time.
a
Consider the first horn of the dilemma—that we may have law on point. Courts sometimes call stare decisis “a principle of policy,”[23] but they don’t really mean it. (If a Justice were convinced that a defendant’s conduct were protected by the First Amendment, it’d be no argument that affirming the conviction might still prove useful for policy reasons—by promoting economic growth, say, or easing the strain on judicial dockets.) Instead, stare decisis is treated as a separate legal rule, like laches or waiver, that justifies courts in reaching different decisions than the merits might otherwise suggest.[24] This is particularly visible for vertical stare decisis: if a district court disregards the decision of its regional court of appeals, that’s treated as a breach of its legal obligations, whether or not it has the better argument on the merits.
But if stare decisis is a legal doctrine, then we need to know how it fits in with the rest of American law. That ineluctably brings up questions of legal theory. For example, some originalists reject stare decisis outright: to enforce a precedent is to ignore the actual Constitution in favor of what judges have said about it.[25] Other originalists find stare decisis to be a valid consequence of their theories, and not what Scalia called a “pragmatic exception.”[26] Maybe a common-law rule of precedent, to be applied when the law is otherwise uncertain (and the past decision not “demonstrably erroneous”),[27] was part of the “original law” of the Founding, which courts must continue to enforce until that law is properly changed.[28] On either of these accounts, the second-best theory might be forbidden, as a departure either from the Constitution or from the Founders’ law more generally. So any complete defense of the lawfulness of second-best stare decisis would have to establish that these originalist accounts are wrong. (And the same goes for contrary accounts of precedent based on other high theories—pragmatism, Ackermanian “constitutional moments,” and so on.)
Kozel describes these “[i]ssues of constitutional legitimacy” as “complicated and fascinating,” but he doesn’t “dwell on them,” because stare decisis is “uncontroversial in modern judicial practice”: not a single Justice “has challenged the lawfulness of stare decisis” (p. 9). This is a perfectly respectable positivist argument, and it operates at a different level of generality than the high-theory arguments discussed above. It suggests that American social practice is unified enough to support a law of stare decisis, in the same way that it supports a federal tax code or state-law property rights—regardless of whether it also supports more abstract legal conclusions about originalism, common-law constitutionalism, or other issues found at what Leiter calls “the pinnacle of the pyramid.”[29]
If we do have a law of stare decisis, though, we need to know what it says. The book discusses various arguments for the lawfulness of stare decisis (pp. 54-59), but these arguments don’t necessarily apply to the second-best theory in particular. Instead, the doctrine that’s uncontroversial in modern practice is the existing one, in which a precedent may sometimes be abandoned for being “badly reasoned.”[30] That’s why, for example, the book describes “the prevailing approach to stare decisis” as still “tethering a decision’s continued vitality to the perceived gravity of its offenses” (p. 102). That’s also why it argues for “revis[ing] the current doctrine” and adopting the second-best theory instead (p. 175), limiting our consideration of a precedent’s wrongness to truly extraordinary circumstances (pp. 121-123).
So we still need to explain why a judge, bound by our existing law of stare decisis, should enforce a precedent that’s not only wrong, but (per the existing rules) legally suitable for overruling. There may be good reason to revise the rules of precedent, of course, just as there’s good reason to revise the tax code. Unless those reasons are legal reasons, though, judges might not be legally free to act on them.
b
One reason to revise the doctrine might be that current law is in dispute. Unlike Leiter, Kozel doesn’t advance an error theory of constitutional debate: he describes modern interpretive disagreements as “principled” (p. 17), suggesting that there still remains something real for us to disagree about. But given the degree of disagreement we have, no one theory can hope to conquer anytime soon, even with truth on its side. “[T]o accommodate the reality of interpretive pluralism,” the partisans of each theory will have to “mediate their theoretical commitments,” saving whatever benefits of their theories can still be saved (pp. 170-171). And the most important benefits, the ones that make the second-best theory best of all, are the benefits of the rule of law. Through its “commitment to continuity and impersonality,” stare decisis also “embodies . . . a commitment to the rule of law rather than the rule of men and women,” offering reliable answers even when men and women disagree (p. 106).
At one level, this account makes perfect sense. To the extent that our “interpretive methodology” rests on “[n]ormative commitments” (p. 64), the on-the-ground fact of disagreement might change how those normative commitments are best satisfied. And whatever reasons one might have for caring about the rule of law (as judges plainly do) are reasons for caring about stare decisis as well.
But there are two kinds of reasons in play here: legal reasons and reasons to follow the law. Both the rule of law in general and our own legal regime in particular confer all kinds of benefits on society, which judges as well as other people have reason to pursue. But doing what the legal system requires, and doing whatever maximizes its benefits under current circumstances, are two very different things. Legal systems are human institutions and share in human frailty; their aims may be ill-chosen or inconsistent, or they may achieve those aims imperfectly. As a result, what one has legal reason to do isn’t always the same as what one has good reason to do, even if one usually has good reason to follow the law. So long as there really is law to follow, many defenses of otherwise-attractive doctrines might be ruled out by the very legal system whose values they serve.
Take, for example, the value of judicial accuracy. A judge who’s “realistic about the limits of individual reason” and “optimistic about the soundness of collective wisdom” might choose to defer to past decisions, in the hopes of increasing her percentage of correct decisions over time (p. 57). By choosing what seems like the worse result in a specific case, she might better satisfy her general obligation “to render decisions according to law” (p. 57). Assuming that this is true, though (and that the past decisions aren’t systematically unreliable),[31] it still presumes that the law allows this kind of consequentialism—maximizing law-adherence over time, rather than respecting the side constraints imposed in each individual case. In fact, the law often rules out tradeoffs like these. A lone juror in a criminal case, convinced within the limits of her own reason that her doubts of guilt are reasonable, may be legally required to dissent from her fellow jurors and to vote to acquit. This might be true even if compromising on a lesser charge would help avoid a retrial and likely conviction on the greater charge: the system demands individual honesty, not maximal accuracy overall.[32] Judges, too, might have similar obligations; perhaps they’re charged to vote correctly by their own lights, and not to accord more than persuasive authority to other judges or other courts. Either way, these are system-specific questions that we can’t answer in the abstract. They depend on what the law says, and not on what would make the judicial system the best it can be.
The same is true for values advanced by particular constitutional theories. Originalism, for example, has been defended for reasons ranging from “popular sovereignty” to “welfare maximization” to others further afield (p. 66). Perhaps, in our fallen world, these values are better advanced by a regime of super-strong stare decisis—which, like originalism, would require judges “to apply predefined, publicly accessible legal rules,” and which would restrict future changes to the process of Article V (pp. 166-167). But this resemblance turns out to be superficial, for anyone committed to originalism as law will not be indifferent as to which publicly accessible legal rules get super-strong enforcement. The law of the Founding and “whatever maxes out our normative commitments” aren’t the same thing.
If interpretive methodology really rested only on normative commitments, then it might make sense to determine which precedents to overrule by asking normative questions—like “why it is important for the law to be correct in the first place” (pp. 64, 92). But if the correctness of a precedent is a matter of positive law, then that question is extremely hard to make sense of—like asking why it’s so darn important for chess players to make valid moves. An external observer might well ask why judicial accuracy is important, or why anyone should care about playing chess the usual way; but inside the system, so to speak, there’s no discussing it. That chess players just do have to make valid moves is a side constraint on “chessical” reasoning, just as deciding cases correctly might be a side constraint on American legal reasoning. That’s why it matters so much whether precedent can legally require what would otherwise be the incorrect disposition—as it does for the district judge obliged to follow the Fourth Circuit. Maybe a court with absolute discretion to deny certiorari—or, indeed, a judge reviewing her daily calendar—might care about “which mistakes are in greatest need of correction” (p. 92); but it’s precisely because these are extralegal decisions that they can rely on such plainly extralegal factors.
In fact, a legal system does more than advance particular values. It advances them in particular ways, with attention to means as well as ends.[33] Law is a “constructed or ‘artificial’ normative system[],”[34] one that “screen[s] off from a decisionmaker factors that a sensitive decisionmaker would otherwise take into account.”[35] Deciding which values, including rule-of-law values, will be sacrificed to which others “is the very essence of legislative choice”—and as was once said of legislative intent, it “frustrates rather than effectuates” such aims to assume that whatever advances legal values must be the law.[36]
Deferring to precedent, by “enhancing continuity and avoiding disruption,” might well promote rule-of-law values (p. 88). But it can’t promote the rule of law unless it’s actually consistent with law, unless it would actually help the law to rule. And we can’t make that determination so long as we’re officially agnostic about what the law requires. Enforcing some other rule that isn’t law, but acts like it, is only enforcing law’s semblance—no closer to real law than honor and propriety are to good morals.
c
All this assumes that there’s some law to be had. So what of the other horn of the dilemma: what if Leiter’s right, and there isn’t really any law on point?
If judges can’t base their decisions on legal norms, they’ll have to use some other set of norms instead. In that case, it helps that the book’s case for stare decisis isn’t “value-free,” but “unmistakably normative”; its argument “privileges the impersonality and continuity of law” as values “accepted by judges across the philosophical spectrum” (p. 106). Yet while the book is broadly compatible with an error-theory account, actually adopting that account might entail a far greater change to our practice, greatly weakening any argument for privileging these norms in particular. Rule-of-law values aren’t the only values out there, and they lose any special claim to prominence as soon as we step outside the process of enforcing law.
Kozel’s isn’t exactly an error-theory account, but error theory provides a useful lens for reading it. If our practice is so divided that we fail to share any deep norms of constitutional law—if our interpretive commitments are nothing more than “individual beliefs,” “preferences,” “inclinations,” or “predilections” (pp. 11, 16, 105, 130)—then it’s easy to see why high theory should be subordinated to other concerns. And if there isn’t any law on high-theory questions, then we can easily strike these questions from the list of traditional factors (like reliance interests or changing facts) that might be relevant to reconsidering precedent. Suppose that a five-Justice majority overrules a Sherman Act case based on its economic views,[37] “empirical observations that do not depend on methodological or normative commitments” (p. 111). The abrupt change might be disruptive, but at least the Justices’ disagreement is about something real, which expert judgment might eventually resolve. By contrast, if the same five-Justice majority overrules campaign-finance cases based on its reading of the First Amendment—well, that’s essentially a matter of opinion, based on “criteria that have no independent content unless they are situated within a particular methodological framework” (p. 110).
If the methodological debates are fruitless, though, it seems odd to describe the ongoing disagreements as “principled” (p. 17), rather than “well-meaning but deeply confused.” Recognizing our error ought to make it go away. If there’s no real law to apply, there’s no point arguing about particular frameworks; we might as well argue about the underlying values instead, conducting our formerly legal debates in ordinary normative terms.
And at that point, it’d be necessary to explain why stability and impersonality are the only norms that count. Second-best stare decisis “implicitly gives less salience to other values” (p. 135); it considers them only in “exceptional situations” (p. 122), as a Scalia-like “pragmatic exception”[38] or “safety valve” to keep disgruntled judges from bolting (p. 123). In “the ordinary course,” the theory chooses to “exclud[e]” consideration of a precedent’s “substantive effects,” whether good or bad (p. 121). That choice needs more defense once we depart from our familiar practice of enforcing preexisting law.
Stability and impersonality do have a natural connection to the rule of law. If you happen to have a legal rule on point, it ought to be applied consistently, even at different times or by different people. And it’s equally natural for a legal system to care more about rule-of-law values like impartial application than about other considerations of substance. Law succeeds at “screening off” those substantive factors from the decisionmaker precisely because it’s an artificial system, one that won’t always line up with our actual moral obligations.[39]
If there’s no law on point, though, there’s little reason to expect these other factors to stay screened off. The fact that stability and impersonality are widely appreciated—that they’re something “all judges can recognize” (p. 109)—doesn’t mean that they’re the most valuable things to have. Sometimes legal systems treat other things as more important: a legislature might prohibit “child neglect” through general standards rather than hyperspecific proscriptions, making sure to cover the waterfront at the cost of tolerating some variance in application.[40] Given a free hand, judges might similarly advance values that trade off with stability or impersonality: establishing justice, preserving domestic tranquility, protecting endangered species, lowering tax rates, and so on. Without law on point, judges should presumably pursue whatever’s actually the best thing to do—with stability and impersonality receiving only as much weight as they deserve.
Even from the standpoint of stability and impersonality, adhering to mistaken precedent can carry real substantive costs. The prisoner with a valid claim to his freedom may find it cold comfort to be “treated consistently with past litigants,” or to learn that his prison term may “enhance a court’s credibility, and with it the status of the judiciary as a steady, impersonal institution” (p. 40). It’s unclear whether courts gain more credibility by adhering to past results or by reasoning from grounds (or arriving at outcomes) that are themselves broadly perceived as consistent with law. And while second-best stare decisis may be a shield against future change, it’s also a sword in the hand of any group of five current Justices, whose badly reasoned opinions can be superseded in the future only by constitutional amendment. In practice, the Constitution might then be precisely “what five justices say it is” (p. 18)—if not today’s five, then yesterday’s, with the entrenched practice turning on which of the Warren or the Rehnquist Courts was the first to get its say. The arbitrary “preferences and personalities of a particular group of justices assembled at a particular moment in time” (p. 42) might then be the sole determinant of many constitutional outcomes.
None of this would matter, legally at least, if second-best stare decisis were required by law. But if there’s no law on point, judges might well prefer to do something else.
iii
In our system, precedent may well be merely a semblance of law. That sounds like a bad thing, but it isn’t. Law, too, is only a semblance. It imposes artificial rules of conduct that sometimes (and only sometimes) advance our actual moral goals. Because people disagree about the right answers to moral questions, we use legal rules to generate fake answers, “on which society (mostly) agrees and which allow us (mostly) to get along.”[41] In the same way, when people disagree about the right answers to legal questions, we use devices such as waiver rules, statutes of limitations, or doctrines of res judicata or adverse possession to generate stand-in answers in place of the real ones. (Whether A really owned Blackacre doesn’t matter, so long as B occupied it openly and notoriously for a long enough time.)
Viewing precedent as provisional in this way suggests a different response to our deep and persistent disagreement. Instead of focusing on the matters on which we disagree, laboring to reconcile the outcomes of scattershot Supreme Court decisions, we should look to the elements of the legal system that are nonetheless shared and see what they entail. We don’t need to construct theories of law to match the Court’s opinions. Instead, we can take the opinions as the Court describes them, as fallible efforts to implement external standards. Doctrines of precedent, like those of waiver or preclusion, might then supply us with answers when we disagree about those standards’ content or application. Treating precedent as a fallback, rather than the system’s foundation, is not only quite useful in practice; it might also point the way toward further unity on matters of theory.
On Kozel’s account, “[d]isagreements over interpretive theory are here to stay” (p. 106). If that’s true, and if the best anyone can do is to count to five, then we can’t say much about the law beyond how it’s practiced on the ground. In such a world, Kozel argues, “a change in the composition of the Court is a change in the fabric of the law” (p. 103). It’s not merely our “perception” (p. 106), but the “reality . . . that constitutional change happens through judicial appointments, not formal amendments” (p. 17).
Yet this picture may be too dark. Despite all our disputes, there’s a curious degree of agreement on the “official story”[42] of the American legal system about what, exactly, precedent is supposed to do. The Supreme Court doesn’t take itself to establish the law for all time, much less to change the law with each new opinion. It agrees that its individual decisions can be wrong at the time—legally wrong, and not just proven wrong by subsequent events.[43] In Lawrence v. Texas,[44] for example, the Court declared that Bowers v. Hardwick[45] “was not correct when it was decided,”[46] and also that it had been fully “authoritative” until its overruling;[47] authority as precedent and actual correctness were two different things.[48] In saying this, the Court was merely reflecting the conventional view, as described by Caleb Nelson: “[a]ll modern lawyers” would understand a claim that “[t]he Constitution plainly establishes Rule X, but the Supreme Court has interpreted it to establish Rule Y instead, and the Court is not going to overrule that interpretation.”[49]
In other words, precedent seems to play a derivative role—not as a source of constitutional law, but as a proxy or heuristic for the law. To both originalists and nonoriginalists alike, foundational cases like Marbury or M‘Culloch are “revisitable, at least in principle, by the Supreme Court,” in a way that the Constitution’s text is not.[50] Even those who argue that we’ve evolved a common-law constitution—in which precedent, and not the text, serves as the primary raw material for interpretation—accept that their approach is not one “we usually associate with a written constitution, or indeed with codified law of any kind.”[51]
To the extent we share a theory of precedent, then, it’s one that takes precedent as a useful semblance, and not as the real thing. Certainly that’s its role in lower courts; the law of the circuit, like the law of the case, stands in for the actual law without replacing it.[52] When people talk about “the law of the Fourth Circuit,” we know what they mean; a district court in Maryland is obliged to act as if the Fourth Circuit is always right on the law. Yet if someone argued that a federal statute really had different legal content in Maryland and in Delaware, we’d call them crazy; that’s just not how the American legal system works.
Because Supreme Court precedent isn’t limited to particular parties or a particular geographic circuit, it’s easier to confuse “the law of the Supreme Court” with the underlying law.[53] But we don’t have to think of the Court as the Oracle of One First Street, delivering infallible pronouncements from on high. We could think of it as a group of fallible human beings who are trying, not always successfully, to apply the law to the facts. And we could take the precedent it produces the way we say we take it—as something explicitly provisional, as the Court’s best guess at the law given the time it had and the arguments it heard.[54]
If that’s right, it suggests a different way to approach our remaining disputes over method. Rather than cautioning the Court against “revisiting its constitutional decisions, lest it be viewed as announcing the conclusions of five individuals rather than the judgment of an enduring institution” (p. 27), we could distinguish the fallible conclusions of individuals (today’s or yesterday’s) from the judgment of an enduring law. Legal theories wouldn’t be expected to “explain the changes that have happened” in judicial doctrine[55]—changes that might themselves result from unrelated contingencies, like shifting political coalitions or the untimely resignation of individual Justices.[56] Instead, they would articulate and expound standards for what makes doctrinal changes correct, about which there may be rather more agreement than we have about the doctrines themselves.[57]
Diminishing the status of past decisions, rather than committing to leave them in place, might also help avoid the confusion sown by arbitrary departures from legal standards, whether through faint-heartedness or the occasional “pragmatic exception.” Lower-court judges may be obliged to assume that the Supreme Court always knows best, and to stare at a pointillist mess of precedent until it resolves into a simple picture. Even the Justices might sometimes be bound, by common-law rules in place from the Founding, to take an inflated opinion of the wisdom of their predecessors.[58] But we aren’t so bound, and we can accept that actual courts will sometimes err. The goal would be a legal culture that maintained a healthy skepticism of judges, but a healthy confidence in the law—one that truly “reinforc[ed] the prevalence of the rule of law over the rule of individual women and men” (p. 16).
In pursuing greater consistency in theory, this strategy needn’t tolerate inconsistency on the ground. If Nelson’s account is right, stare decisis can lawfully act as a fallback in cases of uncertainty—even when the prior precedent is potentially mistaken, but not yet “demonstrably erroneous.”[59] Not only is this fallback theory closer to what the Court already does, but it gets us most of the way to the stable and impersonal practice sought by Kozel. Where the law is “too vague or ambiguous” (p. 59), or “too sparse and abstract” (p. 45), precedent-as-fallback creates “a presumption against discarding the accumulated wisdom of past generations” (p. 39)—something that can “fill the gaps” and “constrain a judge’s discretion where enacted law leaves off” (p. 43).
The fallback and second-best theories come apart, though, when different Justices rely on different theories, such that each finds the other in demonstrable error. To Kozel, this is a stalemate: “Fostering agreement around a particular interpretive methodology is a tall order in our pluralistic legal culture” (p. 107). Though we might “urge a justice to subordinate her personal views” in a particular case, it’s “unreasonable (and unrealistic) to request that she adopt, for all intents and purposes, an interpretive methodology that is not her own”—to make an “extraordinary sacrifice without a sufficient return” (p. 154). Yet we ask judges to set aside their personal views all the time, whenever we require them to decide in accordance with law. If there’s any law on interpretive questions, judges should be expected to apply it, no less than they’re expected to apply the tax code.[60]
Distinguishing precedent from the underlying law—separating semblance from substance—may help us see our way through to these areas of agreement. This process needn’t require any individual Justices “to take the dramatic step of disavowing their constitutional philosophies” (p. 107). A single Justice’s pursuit of the best understanding of the law, like a single Justice’s adherence to precedent, is “a means of strengthening a doctrine that provides value to the constitutional system more broadly”—providing an example to others “irrespective of the conduct of one’s judicial peers” (pp. 130-131). Besides, constitutional theory is a long game: offer a better-grounded theory to the next generation, and in a few years the consensus will change. To paraphrase Max Planck, maybe constitutional theory, like physics, advances one funeral at a time.[61]
iv
At its heart, Settled Versus Right is a case for compromise: a generous attempt at peacemaking in not-always-peaceful times. At the same time, any argument for compromise—for seeking common ground with those who disagree—will always be tested by each side’s own lights. One can compromise on significant matters to achieve significant gains; yet the compromise still has to be, under the circumstances, the right thing for the compromiser to do.[62]
As a moral matter, Kozel’s compromise might well be a good one. But law, the mere semblance of morality, interposes its own rules and may forbid us from taking the deal. However great the need for bridges across our legal divides, it may be that no lasting structure can be built on what’s merely the semblance of law.
[1]. Professor of Law, Notre Dame Law School.
[2]. Professor of Law, Duke University School of Law. © Stephen E. Sachs 2018. The author is grateful for comments from William Baude, Randy Kozel, Amanda Schwoerke, and the attendees of the Settled Versus Right Book Talk at the University of Richmond, as well as for the excellent research assistance of Patrick Butler and Adam Griffin.
[3]. Planned Parenthood of Se. Pa. v. Casey, 505 U.S. 833, 866–67 (1992) (plurality opinion).
[4]. See Gonzales v. Carhart, 550 U.S. 124 (2007) (overruling Stenberg v. Carhart, 530 U.S. 914 (2000)) (abortion); Roper v. Simmons, 543 U.S. 551 (2005) (overruling Stanford v. Kentucky, 492 U.S. 361 (1989)) (death penalty); Citizens United v. FEC, 558 U.S. 310 (2010) (overruling McConnell v. FEC, 540 U.S. 93 (2003)) (campaign finance).
[5]. Jennifer Senior, In Conversation: Antonin Scalia, N.Y. Mag. (Oct. 6, 2013), http://nymag.com/news/features/antonin-scalia-2013-10/.
[6]. Pearson v. Callahan, 555 U.S. 223, 234 (2009).
[7]. Immanuel Kant, Idea for a Universal History with a Cosmopolitan Intent, in Perpetual Peace and Other Essays 29, 36 (Ted Humphrey trans., Hackett Publ’g Co. 1983) (1784).
[8]. Christine M. Korsgaard, Creating the Kingdom of Ends 57 (1996).
[9]. See Stephen E. Sachs, Finding Law, 107 Cal. L. Rev. (forthcoming Apr. 2019) (manuscript at 41–43), http://ssrn.com/id=3064443 [hereinafter Sachs, Finding Law].
[11]. Brian Leiter, Explaining Theoretical Disagreement, 76 U. Chi. L. Rev. 1215, 1221–22 (2009).
[12]. Matthew D. Adler, Interpretive Contestation and Legal Correctness, 53 Wm. & Mary L. Rev. 1115, 1117 (2012).
[17]. See Leiter, supra note 11, at 1224.
[20]. See Richard Joyce, Moral Anti-Realism, in Stanford Encyclopedia of Philosophy (Edward N. Zalta ed., Winter 2016), https://plato.stanford.edu/archives/win2016/entries/moral-anti-realism/#ErrThe.
[21]. See Leiter, supra note 11, at 1232.
[22]. See 2 U.S.C. §§ 437g, 441b (2006) (recodified at 52 U.S.C. §§ 30109, 30118); Citizens United v. FEC, 558 U.S. 310, 321, 337–38 (2010).
[23]. Payne v. Tennessee, 501 U.S. 808, 828 (1991) (quoting Helvering v. Hallock, 309 U.S. 106, 119 (1940)).
[24]. See, e.g., Stephen E. Sachs, The “Unwritten Constitution” and Unwritten Law, 2013 U. Ill. L. Rev. 1797, 1832.
[25]. See, e.g., Gary Lawson, The Constitutional Case Against Precedent, 17 Harv. J.L. & Pub. Pol’y 23 (1994); Michael Stokes Paulsen, The Intrinsically Corrupting Influence of Precedent, 22 Const. Comment. 289 (2005).
[26]. Antonin Scalia, Response, in A Matter of Interpretation: Federal Courts and the Law 140 (Amy Gutmann ed., 1997) (emphasis omitted).
[27]. Caleb Nelson, Stare Decisis and Demonstrably Erroneous Precedents, 87 Va. L. Rev. 1, 1 (2001) [hereinafter Nelson, Stare Decisis].
[28]. On “original-law originalism” and stare decisis, see Stephen E. Sachs, Originalism as a Theory of Legal Change, 38 Harv. J.L. & Pub. Pol’y 817, 863–64 (2015); see also William Baude, Is Originalism Our Law?, 115 Colum. L. Rev. 2349, 2358–61 (2015); Stephen E. Sachs, Constitutional Backdrops, 80 Geo. Wash. L. Rev. 1813, 1863–66 (2012) [hereinafter Sachs, Backdrops].
[29]. Leiter, supra note 11, at 1228.
[30]. Pearson v. Callahan, 555 U.S. 223, 234 (2009).
[31]. See generally William Baude & Ryan Doerfler, Arguing with Friends, 117 Mich. L. Rev. (forthcoming 2018), http://ssrn.com/id=2985032.
[32]. See Comm. on Fed. Criminal Jury Instructions of the Seventh Circuit, Pattern Criminal Jury Instructions of the Seventh Circuit § 7.03, at 103 (2012), http://www.ca7.uscourts.gov/pattern-jury-instructions/7th_criminal_jury _instr.pdf.
[33]. Cf. John F. Manning, The New Purposivism, 2011 Sup. Ct. Rev. 113, 115–16 (discussing attention to means in statutory interpretation).
[34]. Mitchell N. Berman, Of Law and Other Artificial Normative Systems, in Dimensions of Normativity: New Essays on Metaethics and Jurisprudence (D. Plunkett et al. eds., forthcoming 2018) (manuscript at 3), http://ssrn.com/id=3042669.
[35]. Frederick Schauer, Formalism, 97 Yale L.J. 509, 510 (1988).
[36]. Rodriguez v. United States, 480 U.S. 522, 526 (1987).
[37]. See, e.g., Leegin Creative Leather Prods. v. PSKS, Inc., 551 U.S. 877 (2007).
[38]. Scalia, supra note 26, at 140.
[39]. Schauer, supra note 35, at 510.
[40]. See Timothy Endicott, The Value of Vagueness, in Philosophical Foundations of Language in the Law 14, 24–28 (Andrei Marmor & Scott Soames eds., Oxford Univ. Press 2011) (2005).
[41]. William Baude & Stephen E. Sachs, The Law of Interpretation, 130 Harv. L. Rev. 1079, 1096 (2017).
[42]. See Mikolaj Barczentewicz, The Illuminati Problem, Rules of Recognition, and What Distinguishes Law from Non-Law?, 38 Oxford J. Legal Stud. (forthcoming 2018) (manuscript at 1), http://ssrn.com/id=2915564.
[43]. Cf. Christopher R. Green, Constitutional Truthmakers, 32 Notre Dame J.L. Ethics & Pub. Pol’y (forthcoming 2018) (manuscript at 26–27), http://ssrn.com/id=2901157 (describing the Court’s own conception of its opinions).
[46]. Lawrence, 539 U.S. at 578.
[48]. See Stephen E. Sachs, The “Constitution in Exile” as a Problem for Legal Theory, 89 Notre Dame L. Rev. 2253, 2280–81 (2014) [hereinafter Sachs, Constitution in Exile].
[49]. Caleb Nelson, A Critical Guide to Erie Railroad Co. v. Tompkins, 54 Wm. & Mary L. Rev. 921, 937 (2013).
[50]. Neil S. Siegel, Jack Balkin’s Rich Historicism and Diet Originalism: Health Benefits and Risks for the Constitutional System, 111 Mich. L. Rev. 931, 947 (2013) (book review).
[51]. David A. Strauss, Common Law Constitutional Interpretation, 63 U. Chi. L. Rev. 877, 885 (1996).
[52]. Sachs, Finding Law, supra note 9 (manuscript at 41–42).
[54]. See Amy Coney Barrett, Stare Decisis and Due Process, 74 U. Colo. L. Rev. 1011, 1071 (2003) (noting that the quality of argument can vary across cases); see also Antonin Scalia, Originalism: The Lesser Evil, 57 U. Cin. L. Rev. 849, 852 (1989) (describing an opinion that took “almost three years and seventy pages,” and if “done perfectly it might well take thirty years and 7,000 pages”).
[55]. David A. Strauss, Do We Have a Living Constitution?, 59 Drake L. Rev. 973, 976 (2011).
[56]. See Sachs, Constitution in Exile, supra note 48, at 2285–86.
[58]. See Sachs, Backdrops, supra note 28, at 1863–66.
[59]. Nelson, supra note 27, at 1.
[60]. See generally Baude & Sachs, supra note 41.
[61]. See Max Planck, A Scientific Autobiography, in Scientific Autobiography and Other Papers 13, 33–34 (Frank Gaynor trans., Philosophical Library 1949) (1948).
[62]. See David Archard, Moral Compromise, 87 Phil. 403, 410–11 (2012).