I. The purpose of our constitutional practice

What is the purpose of our constitutional practice? That is the first, orienting question. Constitutional practice, rather than Constitution, because the authority of the written Constitution is derivative of its place within our practice. The text is in service of our practice; our practice is not in service of the text... at least not until we arrive at the conclusion that our practice is best served by an understanding that connects us to the text in this dominating way. If the text of the Constitution is appropriately understood as directing the behavior of actors in our political community, it is because the complex practice of constitutionalism —which includes that text— makes that behavior morally obligatory for such actors. This is a familiar observation, perhaps, but no less important for its familiarity.

To be sure, on any understanding of our constitutional practice the written Constitution has an important place. The text is important in a number of ways, not least of which is that, absent serious regard for the text, the people in their role as constituent sovereign have no way of insisting on a change of constitutional course. But the place of the written Constitution in our constitutional practice is defined secondarily, from outside the text, not commanded by the text. Like moral readers of our constitutional practice, originalists of all stripes have to argue from outside the text. This makes the new originalist turn to the theory of language come too early; the place of such theory is necessarily subordinate to a convincing moral account of our practice as a whole.

The purpose I would claim for our constitutional practice is that it is justice-seeking.¹ It aims at better aligning our institutions, policies, laws and the political community as a whole, with the requirements of justice. It aims, that is, at our making moral progress as a political community.

II. Delegatory constitutional provisions

I want to turn now to a feature of the written Constitution that is morally salient to its role in our practice. Many provisions of the Constitution do not constitute encompassing commitments. For a provision of the Constitution to be “encompassing”, for our purposes, it must contain or constitute a full and concrete description of the state of affairs to which it refers. A Court which undertook to enforce an encompassing constitutional provision would have work to do —choices to make— but those choices would be among implementing strategies, not target states of affairs; encompassing provisions fully describe the pertinent constitutional targets. Many —perhaps most— of the liberty-bearing provisions of our Constitution are not encompassing in this way.

Actually, no legal provision can be fully encompassing; there will always need to be some normative engagement with the text to establish its meaning. But let us agree that some provisions reduce the amount of normative engagement to just this side of the vanishing point. This is true, for example, of the Constitution’s stipulation that “neither shall any person be eligible to th[e] Office [of the President] who shall not have attained to the Age of thirty five Years”. Other provisions, in contrast, leave broad space that plainly requires normative engagement to fill in. This is true, for example of the First Amendment’s free exercise clause: “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof...”. For practical purposes, we can regard provisions like the age requirement of the President as encompassing (or at least largely encompassing) and provisions like the free exercise clause as non-encompassing.

The new, language-centered, originalists implicitly disfavor non-encompassing constitutional provisions. Consider, for example, the following new originalist typology of what we have called non-encompassing constitutional provisions: A non-encompassing provision may be vague (possessed of problematic boundaries at its margins); it may be ambiguous (susceptible to more than one meaning); it may be contradictory (in direct conflict with another provision), or it may leave gaps.² There are two things to note about this list: The first is that its operative terms are not flattering; “vague”, “ambiguous”, “contradictory”, and riven with “gaps”, are far from terms of praise or even dispassion. They read like a list of the deadly faults of failed expression. On this understanding, the Constitution ought to have consisted of provisions that read “We must have State of Affairs X”, where State of Affairs X were a legible and encompassing description of a state of affairs favored or demanded by the Constitution. To describe a constitutional provision as vague, ambiguous, contradictory, or gap-riven, is to imply that it is a failed attempt to call out an encompassing state of affairs. On this view the numerous non-encompassing constitutional provisions are unhappily incomplete communications of the constitutional will, well deserving of the these perjorative descriptions.

But to lament the existence of constitutional provisions that do not provide legible descriptions of encompassing states of affairs is to badly misunderstand a crucial element of our constitutional practice. In an early work, and to a somewhat different end, Ronald Dworkin invoked the example of a military figure who is ordered by a superior officer to select the best members of the company under their command for an important mission. That order clearly is not encompassing; an encompassing order would include the names of the soldiers who were to be charged with the mission. But it not vague, ambiguous, contradictory, or gap-ridden; it is delegatory. Properly understood as a delegation, the order is complete; the superior officer has nothing to add, and the recipient of the order does not need anything further to fully understand his or her instruction. As a complete instruction, the order is altogether successful in communicating exactly what the superior officer wants its recipient to do. In like fashion, the delegatory provisions of the Constitution are complete and successful, notwithstanding their non-encompassing nature. A delegatory constitutional instruction hands off responsibility and authority; in order to do that, it must be non-encompassing.

Consider a constitutional provision like the free exercise clause. The language is somewhat unfamiliar, but we conclude that the best reading is something like “Congress must not constrict religious liberty”. That is surely non-encompassing. But it is best read as instructing responsible constitutional actors within its purview to fashion a jurisprudence aimed at assuring that the federal government’s behavior is consistent with religious liberty. That will require the embrace of at least a rough theory of religious liberty; workable doctrine derived from and aimed at implementing that theory; and the application of that doctrine to individual cases (this last mostly by lower courts but with occasional intervention by the Supreme Court to adjust, refine and enforce the doctrine). Ronald Dworkin and others have described the interpretive process of moving from a general target like religious liberty to an intermediate conceptual structure as the move from concept to conception.³ In the case of concepts as abstract as “religious liberty”, an intermediate conceptual structure surely is necessary to the judicial enterprise of fashioning implementing doctrine. In principle, I think, there is no reason to exclude the possibility that there could be more than one intermediate level, nested conceptual structures bridging the gap from the abstract basal concepts to concrete doctrines and case outcomes; there is nothing magical or exclusive about the notion of a conception. The important point for our purposes is that the non-encompassing nature of an abstract constitutional instruction of this sort is not a failure of expression; it is a complete and successful passing on of responsibility and authority for the generation of encompassing content —it is a delegation.

I anticipate a reaction to this observation that takes issue with the proposition that the Free Exercise Clause is should be read as a delegation of responsibility and authority. But text and context press forcibly for that interpretation. “Congress shall make no law ...prohibiting the free exercise [of religion].” If we conclude that this is equivalent to “Congress shall make no law constricting religious liberty,” that with the charge to follow and enforce the Constitution comes the concomitant responsibility of providing concrete content consistent with that delegatory instruction. There is nothing surprising about this understanding; indeed it is hard to imagine what could be the content of a contrary interpretation. If, in our military example, the commanding officer handed his subordinate a written order, with a description of the mission, followed by the instruction “We need the six most qualified members of your company for this mission”, the delegation of responsibility would be plain. So too is it plain in the case of the delegatory provisions of the Constitution.

Many of the most important provisions of the Constitution —particularly the liberty— bearing provisions are delegatory. This not an arbitrary circumstance or simply a matter of drafter’s taste. By design the Constitution is obdurate to change, and demanding of a broad national consensus for enactment and amendment. These structural features of the ratification and amendment encourage delegatory commitments to abstract principles of justice: Generality conduces to the necessary broad consensus; and thoughtful participants in the process of textual composition will see the virtue of abstraction in substantive commitments intended to endure for generations. Further, the use of delegatory pronouncements invites the partnership of contemporary constitutional actors, including and especially courts; thoughtful drafters should see that as a benefit.

The prevalence of delegatory provisions in our written Constitution does not without more insist on the moral reading of our constitutional practice. After all, we began by insisting that it is the best view of our practice that should drive our approach to the text, not the text that should drive our practice. But the delegatory provisions do serve to deprive opponents of the moral reading of our constitutional practice of support from the text of the Constitution. Actually, the problem faced by text-centered originalists runs deeper than that; having turned to the text for guidance, they find the text laced with delegatory instructions. Once we look past the perjorative treatment of delegation tacitly embedded in the linquistic typology of new originalism, we can see the delgatory provisions of the Constitution as calls for the active judgment of responsible governmental actors, not as failed attempts at more precise instructions. Think again of our soldier, commanded to find the best members of his or her squadron to undertake a mission. Were the soldier to demur on the grounds that this choice belonged to the superior officer, they would be disobeying the order. This means that reflective originalists need non-textual reasons to oppose the moral reading; and further, that it will be very hard indeed for reflective originalists to resist some version of the moral reading of our practice.

III. Courts

If, as I argued at the outset, the question is not how the text directs our practice, but rather, what our practice should make of the text, we need to understand what virtues and liabilities flow from the judicial undertaking to discharge the judgmental responsibility of the numerous delegatory provisions in the text. We need to think about courts: How they work and what advantages and disadvantages they offer with regard to process and outcome. This is a complex question which is crucially important to the question of whether our justice-seeking constitutional practice is advantaged if constitutional courts robustly embrace the moral view of our constitutional practice, and with it, the written Constitution, and its many delegatory provisions. This is a question which is all too easily neglected by approaches to our constitutional practice that make the text and linguistic analysis the center of reflective attention from the outset.

Our view of courts is crucial because the moral view of our constitutional practice gives them substantial authority and responsibility to set the concrete shape of important liberties, realms of equality, and normative elements in our structure of governance (most notably, the distribution of authority between the federal government and the states). I hold our constitutional courts in relatively high regard. I want to say a little about why, in part because the reasons why we should esteem courts in this context give us reasons in turn to argue for a view of the best shape of constitutional adjudication. Remember, our practice is justice-seeking and justice is both our purpose and the lens through which we consider our practice itself.

Here is a rough sketch of the case for the robust role of courts in our constitutional practice: Our constitutional courts are common law courts; they are, that is, reason-giving and precedent-drawn (We could say precedent-bound, but something less absolute and less mechanical —on the order of Dworkin’s “gravitational pull”⁴— is apt; hence precedent-drawn.) Common law courts are obliged to give reasons grounded in general principles for the decision in the case before them; and those reasons need to account as well both for what is valued in past outcomes, and, for attractive outcomes in future, imaginable cases. The process of making all this line up is the institutional equivalent of seeking reflective equilibrium. A simple example: Think of a judge deciding whether members of the American Nazi Party have a right to march and spread their vile message of hate in Ferguson, Missouri. How do the reasons and outcome to which she is attracted in the case before her square with her court’s view of the appropriate outcome in Selma, Alabama decades ago? And how would they seem in a future case, as applied, say, to war protestors in a military town in Texas in time of active national military activity? (There may or may not be convincing moral grounds for distinguishing among claims of free speech in these cases... the point is that they place a burden of articulate justification on the judge.)

There are several virtues to this institutional form of reflective equilibration. First, it conduces to fairness. As despised as the American Nazi Party is, as a First Amendment constitutional protagonist, the Party potentially stands in the shoes of Martin Luther King and future anti-war protestors. Second, reflective equilibration is a staple of practical reasoning about normative questions; it gives courts a judgmental capacity not shared by other institutions of government. And third, in principle, constitutional courts have the distinct democratic virtue of giving each claimant the opportunity to invoke the regime of principle to which the Court is on an ongoing basis committed, without reference to the number of votes or number of dollars that support their claim. So it is possible for an accused terrorist, with a limited education in a foreign land, to clash with the Secretary of Defense… to clash and to prevail in the Supreme Court. This is an important and democracy-enhancing form of equality —we can call it deliberative equality.

These are reasons to think that constitutional courts are well-suited for the role they are called upon to play by the moral reading of our constitutional practice. That role is one in which —especially with regard to the Constitution’s liberty-bearing provisions— the text speaks at a relatively high level of moral abstraction, and courts have a significant role in responding to the text’s moral calls with more concrete substantive understandings and still more concrete implementing doctrine. Courts do this in the name of our constitutional practice, as the most conspicuous and settled delegees of the authority conferred on the text by our practice and delegated by the text to responsible constitutional actors.

Respect for the tug of prior decisions, and, more generally, for the processes of common law adjudication, is important at every turn of this account of the virtues of the constitutional judiciary. But in the broader debate between originalist and moral understandings of our constitutional practice, this salient institutional feature of our practice may sometimes be lost to view. For the staunch originalist, what grounds an appropriate constitutional outcome is the enacted text and various facts about the world at the moment of enactment. The discovery of and respect for those textual and contextual facts is central and the process of adjudication secondary and subordinate. This places respect for precedent at risk.

The moral understanding of our constitutional practice should lead more naturally to concerns about the structure and virtue of our common law judicial practice. But even here there are potential diversions. A moral reader of our constitutional practice may well be drawn, as is James Fleming in Fidelity to Our Imperfect Constitution, to Ronald Dworkin’s interpretive approach, which entails a balance of sorts between fit and value.⁵ Dworkinian interpretation enters the project of understanding the law at a number of different conceptual levels. Two are important for our present purposes. First, the project of understanding law in general, or our constitutional practice more specifically is an interpretive project; and second, what courts do when they make judgments about the content of the law is itself an interpretive enterprise. Without more, the general fit/value elements of interpretation concedes a role of considerable importance to what is already in place… we are interpreting, not inventing. Hence the element of fit, which has both a threshold minimum requirement and continues beyond that threshold to exert important force over the shape of the interpretive outcome. But it is important to recognize that judicial interpretation is appropriately understood to include a distinct and robust element of regard for past adjudicated outcomes, beyond the naturally backward looking element of fit in interpretative endeavors generally. In part, that regard grows out of the nature of law generally; in part, it is particular to the nature of adjudication in a common law system. If the efficacy and legitimacy of our constitutional practice depends on the common law protocols that shape constitutional adjudication, moral readers of our constitutional practice need to attend to the virtues and liabilities of courts and constitutional adjudication. Affording too little weight to precedent may undermine claims on behalf of the delegatory strategy of our written Constitution and the robust role adjudication plays in the moral view of our constitutional practice.

IV. Legislatures

Courts are not the only delegees of constitutional authority and responsibility. Legislatures can, should, and do play an important part in constitutional justice-seeking. This is famously true of Congress, which has enacted prominent national legislation barring discrimination in employment, in housing, schools, and in some “public accommodations” that have a connection to interstate commerce or are supported by state action, like movie theaters, restaurants, bowling alleys and gas state stations.⁶

But federal anti-discrimination legislation is in various ways limited.⁷ Much of the slack in addressing social injustice is taken up by state and local laws that insist on equal access to a wide swath of commercial enterprises. These “public accommodation” laws are remarkably widespread; they exist in 45 states, in the District of Columbia, and in a number of major cities. All bar discrimination on race, sex, nationality and religion, and about half bar discrimination on grounds of sexual orientation and/or gender identity.⁸

Some federal anti-discrimination legislation is understood to fall under Congress’s authority to enforce the guarantees of the reconstruction amendments, most notably the abolition of slavery in the Thirteenth Amendment, and the equal protection clause of the Fourteenth Amendment. Much of the federal legislation, however, while unmistakenly pointed at issues of social justice, is understood to draw its authority from the interstate commerce and taxing and spending clauses of the Constitution. The extensive web of state and local public accommodation laws find their authority in state constitutions and home rule provisions. But this fractured and awkward⁹ attribution of authority should not obscure the crucially important role all these legislative enactments play in our constitutional practice.

Since the Civil War, a central preoccupation of our constitutional project has been the overcoming of structural failures of equal membership. The Constitution has come to be understood as insisting on the equal treatment and equal stature of groups who have been systematically denied both: racial and religious minorities, women, and, only recently, gays and lesbians. But the adjudicated Constitution has been truncated its its reach; it stops at the shaggy and incoherent boundary of state action. The rough idea, familiar to even a casual observer of our constitutional tradition, is this: The liberty-bearing commitments of the Constitution only address governmental behavior. The regulation of private behavior lies wholly in the domain of legislatures, with the singular exception of the Thirteen Amendment’s abolition of slavery.

This picture is simple and familiar. It is also incoherent in a deep and obvious way: In the modern state, law is everywhere. The pertinent regime of law may be permissive or highly restrictive, but it is always there, as a background condition. When a restaurant owner refuses to serve someone because of their race, gender or sexual identification; when an employer declines to hire or promote an applicant or employee; when a shopkeeper refuses to serve or sell; when a property owner declines to sell or rent —you get the drift— the simple truth is that the odious refusal is either legal or illegal. The absence of state prohibition is the presence of state permission. Further, the state will have to own up to bite of its vacant law when secondary questions of enforcement arise, as when, for example, the owner of a restaurant asks the police to escort an unwanted patron from their premises.

The point is not that all behavior is state behavior; the restaurant owner, the employer, the shopkeeper and the property owner are not state actors. But when government permits them to discriminate, government has acted; and when government enforces its permission, then too, of course, government has acted. It makes no sense to say that the government hasn’t acted in these circumstances, and no sense to think that the Constitution is inapt. How the Constitution should regard these cases is an open and important question; in practice, our constitutional tradition has waivered erractically on the point, most famously in Shelley v. Kraemer,¹⁰ where the judicial enforcement of private, racially-restrictive covenants on the sale of real estate was held to violate the equal protection clause of the Fourteenth Amendment.

The behavior of the state in permitting, enforcing —and quite possibly encouraging— odius private choices is by no means innocent of the constitutional vice of those choices. This is particularly so in light of the genuinely awful history behind each of these patterns of structural injustice: Government underwrote slavery, of course; government subjected women to numerous disabilities and vulnerabilities; and government punished gays and lesbians for daring to be themselves. Government ought to regard itself as responsible for enduring patterns structural injustice in both senses of that word: It is to a substantical extent the cause of these patterns; and it ought to see itself as obliged, in the name of justice and the Constitution, to undo the enduring consequnces of the patterns it played a deep, deplorable, and thoroughly discredited role in forging.

We can make sense of the state action doctrine, however, if we think of it as enforcing a radical division of institutional labor with regard to the amelioration of structural injustice. Pursuant to that division, legislatures bear ameliorative authority and responsibility, largely to the exclusion of the judiciary. Even if we take a strong view of the constitutional obligation of government to work to undo structural injustice, we have good reason for seeing courts as poorly suited to the enterprise, and good reason, in turn, for seeing this duty as falling on legislatures. The obligation to undo structural injustice is too porous for the judiciary to fulfill: What targets are best and appropriate? What machinery will be efficacious? What level or levels of government should assume what part of the burden? These and others like them are questions of strategy and responsibility that are far better suited to legislative judgment.

It may be helpful to think of the constitutional obligation of government to undo structural injustice as analogous to what some philosophers call imperfect moral duties. Imperfect moral duties are first and foremost insistent and demanding duties; but they are duties with diffuse and complex entailments, and carry with them a good deal of discretion with regard to specifics. Imperfect duties are driven by principles or values, not rules; and and they stipulate ends, not means. That is just what we have here: The constitutional value is equal membership, and, like so many of the demands of constitutional justice, the required end is the extirpation of an extreme injustice: the systematic diminution of stature and regard we have called structural injustice. But the means involve a wide range of choices, choices which legislatures are far better situated to make.

Seen this way, the puzzling legal landscape of social justice in the United States begins to make sense. Legislatures at all levels of government have contributed to a far-reaching web of anti-disrimination laws, making the undoing of structural injustice one of the most constant and pervasive themes of our “Republic of Statutes”.¹¹ Courts, in contrast, have seen the requirement of state action as a bar to spontaneous application of the Constitution to constitutionally-odious behavior by non-state actors. There have been exceptions to this bar, but they involve unusual cases where judicial intervention is consistent with the norm of legislative responsibility: There are a handful of cases in which state or local governments have distorted their ordinary political processes in ways that block legislative efforts to dismantle structural injustice; the Court has acted in those cases to restore state and local legislative authority.¹² Several other cases involve state acceptance of non-state behavior that clearly threatens to entrench structural injustice; these are cases where no state committed in good faith to the enterprise of dismantling structural injustice would countanence the conduct in question. The Court in these cases has directly interdicted the threatening conduct in the name of the Constitution.¹³

Where courts, for institutional reasons, cannot fulfill the responsibilities of constitutional delegation, legislatures can; and they have in fact done so. The result is not perfect: Five states have no public accommodation statutes at all; and there has been no serious response from either the judiciary or popular politics to this glaring gap. The statutes in the remaining 45 states are themselves incomplere and to some degree works-in-progress. But state and local legislation has gone a considerable distance towards the creation of a credible response to the challenge of structural injustice. The amelioration of structural injustice is a vivid example of the division of constitutional responsibility between courts and legislatures, and, further, of the rich complexity of constitutional practice under our delegatory Constitution.

V. Fidelity to our constitutional practice of justice-seeking delegation

Fidelity is often treated as the governing virtue of constitutional actors. And there is something very attractive about fidelity. It bespeaks constancy to an something worthy of such faithfulness, even in the face of temptations to waiver. Fidelity mistakenly might be identified with some form of originalism: On this sort of view, what the faithful constitutional actor is to do is dig into text or history, assign an enactment-centered meaning to the written Constitution, and stick by the instructions thus derived through thick and thin. If that is what constitutional fidelity meant, then fidelity would be virtue in marriage, but far from that in our constitutional practice.

But it makes great sense to see constitutional actors —judges, legislators, and for that matter, members of our political community in their official capacity as voters— as being committed on an ongoing basis to the substantive values of our constitutional tradition and to their roles in working towards the achievement of those values. This fidelity to role and value is deeply challenging. In the place of a fictive set of a enactment-centered, concrete and detailed instructions, responsible constitutional actors have to assume intersecting responsibilities to the rule of law and to moral progress. Constitutional fidelity in our world demands judgment, moral imagination, and courage; and the temptations to deviate from faithfulness are many and great. But constitutional fidelity offers us as a people the opportunity to be a better version of ourselves. That ought to be reward enough.