EVEN apart from its resemblance to esoteric debates in medieval Christian theology, the notion that a constitutional amendment (even more, a constitutional provision)1 might be unconstitutional seems almost absurd.2 On the one hand, the very idea that an amendment, proposed and ratified according to procedures set out in what appear to be the plain words of the Constitution, could be unconstitutional seems somehow incoherent, incompatible with what we sometimes take to be our “common sense” (or with widely shared positivist preconceptions) about how constitutions work.3 John Marshall may have put this sense best in Marbury v. Madison, where he held, “It cannot be presumed that any clause in the constitution is intended to be without effect.”4
On the other hand, a constitution without authoritative limits to the power to amend, including the power to amend its explicit provisions concerning the power to amend, carries within it the possibility of its own negation. That is, a constitution, which is to some extent a device for preserving certain states of affairs, might become a device for undermining the very states of affairs it is designed in part to preserve.
Each notion may lead to a kind of interpretive incoherence as well. Either Article V does not mean what it seems on its face to mean—that an amendment “shall be valid” when proposed and ratified according to its (Article V’s) terms—or it is an alternative (super)constitution unto itself and can, without harm to itself, swallow up the rest of the Constitution on which it seems to depend for its existence. Assuming, as I do, that coherence is a fundamental goal of constitutional interpretation, the question is how to render coherent what appears patently incoherent. Less abstractly, what is a political society free to accomplish by amendment to its constitutive document(s)? Or, to be more textually attentive, what is the relationship between Article V and the rest of the Constitution? Is Article V an integral part of the larger (whole) document or is the rest of the Constitution simply riding piggyback (perhaps temporarily) on Article V?
I explore those and other questions by examining an amendment that the U.S. Congress proposed and two (perhaps three) states ratified on the eve of the Civil War. The Corwin Amendment, which I refer to in the title to this chapter as the “Original” Thirteenth Amendment, was designed to protect slavery, perhaps perpetually. That amendment presents in historical guise an opportunity to investigate whether a written constitution can successfully fix substantive limits to formal constitutional change. Are there indeed limits to the amending power? If so, what are they and where do they come from?5 Had the Corwin Amendment been ratified, would it have been unconstitutional? (The fact that the amendment concerns what people now consider to be the easy moral question of whether slavery is a permissible practice actually permits us also to consider the relationship between constitutions and morality.)
In the chill of the winter of 1860–61, as the second session of the Thirty-sixth Congress of the United States convened and Abraham Lincoln was preparing for his inauguration, secessionist sentiment in the Deep South was beginning to reach a fever pitch. For many Southerners, the election of a “Black Republican” as president made John C. Calhoun appear a prophet. They worried that, through patterns of immigration and settlement, the resolution of disputes over the allocation of territories as “slave” or “free,” and the imperial character of Northern capital, the North had gradually gained virtual control of the national political apparatus.6 And Southerners consequently worried that not only were the stability and permanence of their society and institutions threatened, but also that the very character of constitutional government had been transformed into “a great national consolidated democracy” in which the national government could “claim, and practically maintain, the right to decide in the last resort, … the extent of its powers.”7 The fear, in sum, was that constitutional limitations had broken down. Some Southerners, wanting “the Union as it was,” which was presumably to say a creature of a compact among sovereign states,8 argued that the North had breached the compact by electing Lincoln.9
Although one might well argue that these Southerners misunderstood both the extent of differences between the two regions and the actual likely consequences of Lincoln’s election, their fears were certainly not constructed out of whole cloth. In substance, the nation’s future lay westward in the territories. Under virtually all of the plans for designating territories as slave or free,10 the potential size of slave territories was always much smaller than that of territories in which capitalism and wage labor were to be practiced to the exclusion of slavery.11 It was reasonable to expect that such an arrangement both would reinforce the North’s advantage in population and would eventually translate into the admission of a greater number of free states than slave states. Consequently, Northern interests could gain effective control of both houses of Congress. Lincoln’s election had demonstrated that they could gain control of the presidency. And although the Supreme Court was then populated with a majority of five Southerners and at least one “Copper-head” (a Northerner sympathetic to Southern interests), Northern control of the Congress and the presidency might eventually translate into a new antislavery majority on the Court. Thus, even without new interpretations of the existing Constitution that slavery was unconstitutional,12 a constitutional amendment prohibiting slavery in all the states was possible unless there was an enforceable substantive limitation on the nature of amendments.
Southern insecurity to one side, when Southerners began, as others had in previous years, to speak of and eventually threaten secession, many Northerners, too, miscalculated. They underestimated the intensity, popularity, and sincerity of secessionist sentiment in the South.13 Even after South Carolina seceded on December 20, 1860, followed in rapid succession by Mississippi, Florida, Alabama, Georgia, Louisiana, and Texas during the first month of 1861, some in the North argued that the states of the Deep South were bluffing and that some sort of concession would keep them in the Union and prevent states of the Upper South from seceding.14 On December 18, 1860, the Senate directed its Special Committee of Thirteen to investigate and report on the “agitated and distracted condition of the country, and the grievances between the slaveholding and non-slaveholding States.” Although the Committee was “not … able to agree upon any general plan of adjustment,”15 the House’s Select Committee of Thirty-three, investigating the country’s “perilous condition,” recommended the adoption of certain measures to mitigate the danger of overt hostility and to preserve, or reestablish, union.
One of those measures was a proposal to amend the Constitution.16 On March 2, 1861, absent the senators and members of Congress from the seven states of the Deep South that had seceded, an overwhelmingly Northern Congress approved a substitute resolution proposing the adoption of what would become known as the Corwin Amendment:
Resolved by the Senate and House of Representatives of the United States of America in Congress assembled, That the following article be proposed to the Legislatures of the several States as an amendment to the Constitution of the United States, which, when ratified by three-fourths of said Legislatures, shall be valid, to all intents and purposes, as part of the said Constitution, viz.:
“Article Thirteen.
No amendment shall be made to the Constitution which will authorize or give to Congress the power to abolish or interfere, within any State, with the domestic institutions thereof, including that of persons held to labor or service by the laws of said State.”17
President Lincoln acknowledged the pending amendment in his First Inaugural Address and said he supported it.18 Moreover, he chose personally to sign the resolution after Congress adopted it, the first time in our history that a president had so participated in the amendment process.19 Three states ratified the amendment—“Ohio and Maryland through their legislatures, and Illinois through a constitutional convention”20—all of which would eventually find themselves on the Northern side in the Civil War. The outbreak of military conflict and the secession of four more states in the Upper South rendered moot the amendment’s purpose of preserving Union by providing security to the South.
Suppose the Corwin Amendment had been ratified by the requisite number of states. Would it have been “constitutional”?21 That question depends for a coherent answer on a theory of the limits to constitutional change, which in turn requires corresponding theories of interpretation, of constitutionalism, and of the American constitutional regime.
After all, the resolution of debates over the existence and extent of limits to constitutional change is not simply a function of figuring out what the plain and explicit words of the document that we take to be our constitution mean. Even if we were to agree that our actions are or ought to be governed by particular marks on specified pages of paper, the explicitness of the marks—combined with an assumption that we take them as signifiers both of particular meanings and, perhaps more important, of the authority and possibility of interpretation itself—typically does not provide answers, but merely poses questions.22
Consequently, current debates over constitutional limitations have necessarily wrestled with the constructive tensions (or irreconcilable differences) between or within more abstract categories such as parts and whole(s), procedure and substance, means and ends, and internal and external perspectives. They have also implicated arguments among positivists, natural lawyers, and others over the source(s) of political and constitutional authority (sovereignty) and over the nature and authority of law. The very structure of the debates, therefore, reflects the tension between a desire to adhere to the constitutional text and a recognition that in important ways texts are not neatly self-contained but require something “outside” themselves to give them meaning. While the former desire presupposes, at least in its strongest form, that language can tell us with some degree of certainty what to do, the latter recognition implies that constitutional categories are not determinate.23 And as I have suggested, the debates also unavoidably rely on assumptions about the character of the Constitution or constitutionalism.
Those assumptions can come from a variety of sources, which in turn can locate their authority either in transcendental notions of morality or within the Constitution. I shall consider three types of sources that find currency among constitutional theorists. Two of those sources, natural law and platonic idealism, are largely “transcendental.” One, popular sovereignty, is largely “immanent.”24 While these sources do not exhaust the range of possible sources, they do seem to mark limits that are both recognizable and oppositional.
Walter F. Murphy nods definitively in the direction of natural law, or more accurately its cousin, natural rights. On the issue of limits to the amending power, his theory is in important respects similar to Lysander Spooner’s.25 Murphy argues that one can “retroduce” from the Constitution, in fact from any legitimate constitution, a hierarchical ranking of constitutional values, rights, and duties. Although the structure of values in a constitutional regime is potentially complex, the foundational value, the principle that underlies the entire constitutional structure and that motivates the constitutional enterprise, is “human dignity.” The hierarchy of values that grounds the constitutional text(s)—but that is itself grounded in the value of human dignity—informs the limits to “the possibility of legitimately amending the constitutional document so as to remove or materially restrict certain fundamental rights and severely cripple or even destroy the polity’s structure of values.”26
In Murphy’s hands, therefore, retroduction functions both as an interpretive device and as a vehicle for confirming the basic principles of natural rights, which form the basis for the Constitution’s normative theory. Hence, he can claim with some justification that his theory of natural rights is both immanent and transcendent. For him, the Constitution—including the Declaration of Independence, the Bill of Rights (especially the Ninth and Tenth Amendments), and the Preamble—becomes the positive embodiment of natural rights.27
In a later work, Murphy expands on his initial formulation. He argues that the plain words of Article V impose substantive limits on constitutional change, for they refer to “amendments” (that is, modifications), which are not equivalent to “creations of a new constitution.” Consequently, amendments must be either procedural in nature or aimed at clarifying the existing text. “Clarifications” are those substantive changes “necessary to make the Constitution intellectually coherent … [and to] increase the integrity of its underlying principles.” Those principles are not embodied in federalism, for federalism is merely a means, albeit an important one. Principles have to do with ends, and where means and ends conflict, the former must yield to the latter. The principles of the American constitutional regime consist of three values that are closely associated, through the Declaration of Independence, with natural rights: justice, liberty, and equality.28 To the extent that it perpetuates slavery, the Corwin Amendment modifies Murphy’s retroduced values of liberty, justice, equality, and individual natural rights and hence renders the Constitution incoherent.
Despite my worry that the Declaration and the Constitution might be incompatible,29 suppose we concede that slavery violates the principles of the Revolution and that those principles continued to hold after the constitutional founding. Do not the constitutional principles of the structure of sovereign authority, or more accurately the structure of the powers of the creatures of sovereign authority (which today we know as federalism), limit the reach of the principles of the Declaration? Murphy says no, because federalism does not rise to the level of a principle or an end. It is merely a means and is therefore always trumped by the more fundamental principles or ends of the Revolution.30 Under Murphy’s theory, individual rights always trump governmental authority. Because the Corwin Amendment is incompatible with fundamental principles of natural right, it dies.31
A second, explicitly transcendental, theory of limits to formal constitutional change is that of Sotirios Barber. It is based on an aspirational theory of the Constitution derived from the classical political theory of Plato and the moral theory of Kant. Like Murphy, Barber argues that, because “the word amendment ordinarily signifies incremental improvements or corrections of a larger whole,”32 amendments must be compatible with “the Constitution’s larger commitments—like democracy and the public and private attitudes conducive thereto, or human dignity, or freedom from ideological or religious impositions.” Article V provides textual support for its own limited domain both by noting that amendments are valid only “as Part of this Constitution”33 and by supplying explicit restrictions on permissible amendments.34
Barber also joins Murphy in exploiting a distinction between means and ends. Amendments are not always simply procedural in nature; moreover, procedural and substantive commitments are often difficult to distinguish. Procedural values are usually inextricably tied to substantive values,35 and some procedural values are important enough that they “seem to be valued as ends in themselves” apart from the substantive values they might imply. Nevertheless, and perhaps inconsistently, “the means should be sacrificed to the end.” “Claiming supremacy for a set of mere means would not only be irrational, it would also be unjust.”36
Two notions seem to follow from these contentions. First, if “the Constitution is a set of mere means,” which Barber believes follows from viewing Article V as imposing only procedural limitations on the amending process, “we cannot hold it intrinsically worthy or deny the possibility of its failure.” In order to hold the Constitution as worthy of being followed, therefore, we must “look upon [it] as a norm that combines the attributes of means and ends.” And because it is a norm that subordinates means to ends, it also allows holding certain provisions of the Constitution “irrational and unjust” (and presumably, therefore, unconstitutional).37
Second, Barber contends that in trying to determine what the Constitution means, we must acknowledge that there is a (single) meaning embodied in the language of the Constitution. This meaning, he argues, is not a function of immediate wants, individual interests, or assertions of will. Nor is it simply a function of historical practices and attitudes, especially “in easy questions concerning power and in questions involving what … [Marshall] called ‘the great principles of liberty.’” Instead, it is a function of constitutional aspiration, reason, justice, and tradition.38 As long as we can affirm that “the ways of the Constitution constitute our best current conception of the good society—our best understanding for now” and that the Constitution continues to “serve[] the ends for which it was established,” then that conception of the good society and those ends serve as substantive limitations on the amending process. Because we are not perfect, because we are willful and have wants and interests incompatible with reason, justice, and our true aspirations, the Constitution, through “relatively noncontroversial rules [and] a more or less stable structure of [governmental] offices” and through limitations on constitutional change, tries “to fortify the best in us … [by] restrain[ing] the worst in us.”39
Sotirios Barber, as a constitutional theorist sitting in the twentieth century, would have little trouble arguing that slavery violates his account of the Constitution’s “larger commitments,” his vision of what we ought to be doing and of what is best in us as a people, and his best conception of the good society. The question, however, is how he would treat the Corwin Amendment were he sitting in the middle of the nineteenth century. Although his is fundamentally a moral theory in important respects, it possesses a historical sensibility. He writes of “our best current conceptions of the good society,”40 which implicitly acknowledges a kind of moral progress rooted in historical circumstance. First, the word our might mitigate the pull of the platonic ideal or a moral reality insofar as it suggests the importance of collective conceptions of the good. Second, the word current seems to imply that in a historical context, people’s conceptions of the good can change through time.41
But even this conventionalist and historicist gloss on his theory might render the amendment unconstitutional by nineteenth-century standards. For example, Barber could point to Jefferson’s position on slavery.42 He might note that many, perhaps most, of the Founders, and maybe even most Americans in the late eighteenth century, thought that slavery would die out. He might argue that the fact that the word slavery never appears in either the Constitution or the Corwin Amendment provides a textual indication of moral shame about the institution. He might note that Lysander Spooner, Frederick Douglass, and others were employing constitutional theories in the nineteenth century that argued that slavery was unconstitutional. He might suggest that although the cotton gin made slavery profitable in the early nineteenth century, the shortsighted self-interest that secured slavery as a historical practice could not outweigh as a constitutional matter slavery’s incompatibility with Marshall’s “great principles of liberty” or “the nation’s highest aspirations.” Even if federalism were viewed as one of the Constitution’s larger commitments—that is, as a procedural value so important that it is valued as an end in itself—he might argue that federalism is inferior as a constitutional value to liberty and human dignity. As one scholar has said, people are simply not willing to die for federalism. For Barber, the Corwin Amendment neither fortifies the best nor restrains the worst in us. It would be unconstitutional.
Of course, one need not resort to transcendental sources for substantive limits to formal constitutional change. One might well locate such sources or standards within the Constitution. Such an enterprise would remain theoretical, but its method is largely interpretive. As already indicated, Murphy, perhaps even Barber, flirts with an attempt to derive his substantive constitutional theory from a method of interpretation. Murphy’s retroduction is an attempt to supply such a method, under which one interprets a constitution not simply by reading isolated clauses as if one were searching for a peg on which to hang a hat, but by interpreting individual clauses in light of the “inner unity” of the whole document. “As a unit, a constitution reflects certain overarching principles and fundamental decisions to which individual provisions are subordinate.”43
The most detailed exploration of “immanent” limits to constitutional change comes from William F. Harris II,44 who rejects many of the transcendental tendencies of Murphy’s and Barber’s theories. He rejects Murphy’s assumption that the essence of constitutionalism is bound up in natural rights, liberal theories of government, or notions of modern liberty and Barber’s version of reason as a transcendental category. He also rejects Barber’s contention that constitutionalism is fundamentally a normative or moral enterprise (or at least that it is a normative enterprise whose standards derive principally from “outside” the constitutional order). And he partly rejects the notion that substantive moral principles, even those retroduced from the constitutional document itself, operate as constraints on the amending power. For Harris, the Constitution can and may turn itself inside out. Sometimes.
Harris’s constitutionalism is firmly grounded in a vision of ancient liberty, in the liberty of a political society to govern itself. For constitutionalism is essentially the choice by a society to constitute itself politically by reference to a text. The crucial problem for Harris’s constitutionalism is the problem of authority: From where does a putatively constitutive text derive its authority? Harris locates the authority of the Constitution, indeed of any constitutional government, in the myth that “the people” are sovereign, that the people as a body represent “the ultimate source of constitutional authority.” That myth supplies the operative limits to change under Article V.45
The myth of popular sovereignty supplies such limits in two ways, one substantive and the other procedural. The substantive limit, deriving from the nature of Harris’s constitutionalism, asks “to what extent does the amendment cede popular sovereignty or permanently withdraw rights from the people.”46 The procedural limit refers to Article V’s procedures for adopting amendments and to one other possible procedure. Harris recognizes a hierarchy of procedures, each of whose authority is linked to the extent to which it is “approximate to the whole population.”47
The most minor “emendations” to the Constitution could be accomplished by the least stringent procedure. More fundamental alterations of constitutional arrangements, such as those that “implicate sovereignty” (as opposed to one or more of “the sovereign’s creations”) or that accomplish what amounts to “wholesale textual alteration,” would presumably require a more authoritative pedigree, a more stringent procedure. And “categorical substantive prohibitions,” including provisions that purport to be unamendable, would be unconstitutional if “accomplished through the Article V process.” Such provisions would “violate[] the status of the Constitution as a whole design” unless adopted according to the most stringent procedure available.48 The notion of a constitution as a whole design “lies behind the assumption that the Constitution itself may not be self-contradictory.” It would also seem to prop up claims that Article V is an integral part of the larger constitutional structure (as opposed to being a parallel, independent constitution) and that amendments are part of (and therefore cannot contradict) the meaning at the core—the hierarchy of procedures—of the Constitution. Even Article V could be amended, but only if such amendments “ratchet” Article V closer to Article VII’s procedures.49
The first four procedures, of course, are permutations of the possibilities provided under Article V. The fifth procedure, which in the case of the Constitution is equivalent to an amendment’s being adopted by Article VII’s procedures, derives from the nature of a constitution: It is “amendable … on the same terms as those by which it was given authority.” Consequently, even without an explicit textual procedure for amending the document, Harris’s interpretable Constitution is also “essentially” the amendable Constitution. And although under current liberal theories of constitutionalism the procedural and substantive standards for establishing limits to constitutional change are in tension (even conflict) with one another, Harris’s constitutionalism harmonizes substance (embodied in the notion of sovereignty) and procedure (embodied in the notion of pedigree).
For Harris, the question of the constitutionality of the Corwin Amendment is an easy one, but not because slavery is immoral, illiberal, or inconsistent with modern liberty, natural rights, human dignity, or any other substantive values implicit or explicit in the Constitution. Nor, presumably, is it unconstitutional because it “retro-ratchets” sovereignty in some form, since slaves were neither part of the sovereign nor members of the political community. Rather, it is unconstitutional because it purports to be unamendable, and the polity can adopt unamendable provisions only if it complies, at a minimum, with Article VII’s procedures.
Under all three of the theories considered in Section II, we get happy endings, at least from the standpoint of ethical sensibilities prevailing today.50 The possibility of perpetual slavery is detoured, either by an ingenious rationalization of the notion of sovereignty under Harris or by the better angels of our nature under Murphy and Barber. But constitutionalism is not merely about happy endings. Even if we grant the assumptions both that it is possible to give sufficient analytical content to such categories as human dignity or the good life to make them useful as moral guideposts and that constitutional inquiry is bound tightly to moral considerations, it remains the case that such moral considerations do not capture entirely the nature of the constitutional enterprise. To act constitutionally is not always to do the right thing.
Constitutions and the application of constitutional standards are not efforts to mimic, in Lockean or platonic fashion, the perfection (or imperfections) of nature or the moral reality that exists outside the cave. Nor are they, after Hobbes, necessarily efforts to ward off insurgent chaos. Instead they represent attempts, through written texts of manifold forms, to answer affirmatively Hamilton’s question about the Constitution of the United States: Whether it is possible to “establish[] good government from reflection and choice,” instead of through “accident and force.”51 At bottom, then, constitutionalism is concerned with the “constitution” of political societies by reference to written texts.52 It is the attempt to render political life comprehensible and therefore controllable through the written word. It describes and justifies a practice by which a two-dimensional blueprint ostensibly becomes realized in a three-dimensional world. It provides the foundation for the distribution of political authority in society. It need not have as its purpose or as a principal goal either a liberal theory of rights or a moral theory of decision making. It requires only that the actors in a political society justify (and perhaps envision) their actions through the arguably authoritative categories that the constitution supplies.
A constitution is a law that people, perhaps only mythically, have given to themselves. They might give it to themselves in two ways: one, by adopting it in whole or in part, that is, through initial ratification or through ratification by amendment; two, by interpreting and applying it in the ongoing life of the political society.53 In either case, one requisite standard for comprehensibility and hence for the legitimacy of the (or a) constitutional enterprise is coherence.
As a standard, coherence operates along two axes, one horizontal and one vertical. Horizontally, coherence has to do with the unity, wholeness, and sensibility—in short, the meanings—of the constitutive text(s) of a political society. Vertically, coherence concerns the political society’s “ethos,” those values or self-understandings that both inform the construction of constitutional meanings and serve, in light of the limited flexibility of language in a given time, culture, and circumstance, as criteria for judging the acceptability of those meanings. Horizontal coherence is principally a question of principle, informed in its construction by theory. Vertical coherence is largely a pragmatic concern that is primarily informed by experience, including established practices, extraconstitutional notions of morality, and even biases and predilections.54 The correspondence of the concerns represented by the two axes goes to the legitimacy of constitutional judgment. The closer that a constitution or a constitutional judgment comes to the intersection of the two axes, the more likely it will withstand questions about and challenges to its legitimacy.55
The justification for coherence takes two forms, corresponding to the axes along which it operates. First, and perhaps running along the vertical axis, coherence seems inextricably tied to the notion of comprehensibility. If we are to make a world through reflection and choice, then we must make it comprehensible. We must make it understandable. And if we are to make it comprehensible, then coherence is a valuable category. Second, and tending to run along the horizontal axis, I assume that coherence is linked to the notion of standards and hence to principle. It seems clear that a polity willing to forego coherence, to forego standards, stands ready to undercut the very legitimacy of government. It undermines institutional accountability. It breeds cynicism, even despair. In short, although government can exist and society will not crumble without standards, that government will cease to be “constitutional.”
But a potential challenge to horizontal coherence arises from Hume’s observation: “Speculative reasonings which cost so much pain to philosophers, are often formed by the world naturally, and without reflection: As difficulties, which seem insurmountable in theory are easily got over in practice.”56 In other words, the vertical axis often subverts the force of the horizontal. In our time, though, theory’s most serious challenge may come partly from within itself. For we live in an age in which visions of state-ordered harmony seem frightening, in which social and technological change seems either out of control or resistant to the application of available analytic or explanatory categories, in which many intellectuals have backed themselves into “totalizing” theories from which no exit seems possible, in which intelligent belief in external standards for behavior or practice seem implausible, impractical, or defunct.
A comprehensive defense of theory is beyond the scope of this essay, but a partial defense might rest on a rather simplistic and unsatisfying observation: Theory and experience need each other. Disconnected from experience, theory becomes a useless, airy nothing. And experience becomes simply incoherent, or at the very least meaningless, without theoretical categories of some sort to make them sensible.57 Hence, though they may ultimately be quixotic, the following assumptions seem sufficiently defensible that we are justified in holding onto them. First, both experiential and pragmatic considerations do matter for constitutionalism. Second, some form of constitutional theory is still possible in the twentieth century. And third, we can continue to use theory to draw meaningful boundaries with practical implications in an age that sometimes deems itself free of such illusions.
That said, one of the problems of the theories attributed to Murphy and Barber is that although they achieve a nice horizontal coherence, they may not be adequately coherent vertically, at least for understanding the Constitution and politics of the United States in the nineteenth century. They seem circumstantially disembodied, insufficiently sensitive to the role of vertical coherence as a regulator of constitutional meanings and hence of constitutional limits.58
But maybe Murphy and Barber are right. Maybe there are some amendments so at odds with the existing categories of the constitutional text or with interpreters’ perceptions of those categories or of themselves as a political society that the amendments simply render the text incoherent on some level and therefore virtually useless (at least on some matters) as a constituting document. And in such cases, either we throw out the entire document or we keep the document and throw out one or more of the portions that render it incoherent. Plainly, Murphy and Barber would have us throw out the Corwin Amendment as incompatible with the Constitution’s substantive values. The problem is that the Corwin Amendment does not seem to have rendered the Constitution incoherent, at least not in 1861.
The first reason, which cuts against Harris’s notion of popular sovereignty, has to do with the Constitution’s allocation of political authority. In nineteenth-century America, popular sovereignty (at least in the nationalist sense in which we think of it today) was as empirically empty as was the popular founding of the new constitutional order in 1789. In fact, there was a substantial body of opinion in the nineteenth century holding that the national political order was a creature, not of the people-as-sovereign, but of preexisting sovereign states. At the very least, notwithstanding Marshall’s attempts to assert national governmental primacy and a version of national popular sovereignty, the Constitution was generally held out to be a document under which states retained a large degree of discretion over their internal affairs and “domestic institutions.”
The second reason has to do with the status of slavery under the Constitution. Although it was possible, as Frederick Douglass and others demonstrated, to generate a horizontally coherent abolitionist reading of the Constitution, it was simply implausible under most accepted accounts of the text at the time to read the Constitution in that way. First, slaves as noncitizens were not part of the sovereign, popular or otherwise, that ratified the Constitution, either in the founding act of 1788–89 or in the continuous acts of interpretation and amendment. Second, slaves were often held to be exempt from most accounts of “the great principles of liberty” of the Declaration of Independence. Third, to speak of the Constitution of the mid-nineteenth century (without the gloss of the Declaration of Independence) as embodying human dignity or the best aspirations and traditions of the American people or even (less ambitiously) liberalism, at least on the question of slavery, seems somewhat anomalous.
Undoubtedly, there were (and are) those who viewed the Declaration of Independence as establishing for the new nation a universalistic theory of the fundamentality of the principles of individual liberty, justice, and equality. Similarly, there were (and are) those who viewed the Constitution, especially the Constitution of 1791, when the Bill of Rights was adopted, as the embodiment of that interpretation of the Declaration’s principles and as necessarily implying the abolition of slavery.59 It is equally plausible, however, as one scholar has suggested, that many Americans understood the Declaration of Independence as an articulation of principles of political self-determination and not of more radical and individualistic principles of modern liberty.60 Moreover, slavery was an important part of the American system that was to be self-determined.61 But even if we concede the libertarian and abolitionist potential of the Declaration, it is not clear that the Constitution fully (or even substantially) incorporated those principles. After the Revolution, as the new nation became threatened by economic stress and political fragmentation, the Constitution represented a tool as much for retrenchment, stabilization, and the consolidation of economic and political power as for the protection of individual liberties. If Locke was the spiritual progenitor of the Revolution, Hobbes could probably claim as convincingly to be the patriarch of American constitutional order.62
As Jefferson himself acknowledged, there was little dispute at the time of the founding that slavery was protected under the Constitution.63 Whatever the state of the American mind at the time of the Founding, however, the American ethos began to change over time in more than one direction where slavery was concerned. Pulling in one direction was abolitionism, whose seeds were sown in the moral fervor of the evangelical movement, took root, and flowered in the form of secular abolitionist, manumission, and antislavery societies that retained much of the moral fervor but often little of the transcendental religious baggage of Protestant evangelicalism. Many in those movements agreed with William Lloyd Garrison that the Constitution was a “covenant with death, and an agreement with hell.”64 Others, not wishing to make such a radical break with the existing constitutional order, devised and employed clever interpretive strategies to argue the unconstitutionality of slavery or to limit its reach in the country. For example, some incorporated accounts of natural law into constitutional interpretation.65 Some adopted alternative, quasi-positivist attacks on the constitutionality of slavery.66 Despite the changes, none of the antislavery interpretations of the Constitution that emerged in the 1840s and 1850s found substantial institutional support or widespread public acceptance, even as late as 1861.
If Lincoln is an accurate gauge of both Northern ethos and the outer limits of legitimate constitutional interpretation on the issue of slavery, two matters bear consideration. First, even Lincoln, who advocated the constitutional notion of perpetual union, sometimes referred to the nation as a “Confederacy,”67 which would seem to underscore the importance of federalism for understanding the nature of the American constitutional order. Second, although he opposed slavery in general on moral grounds and opposed its extension into the territories, he had never opposed its maintenance in those jurisdictions where it already existed. In his First Inaugural Address, when he “depart[ed] from [his] general purpose not to speak of particular amendments” to state his position on the Corwin Amendment, which was then pending in Congress, he stated that he did not object to making explicit that the individual states possess control over their domestic institutions, exclusive of federal interference, that he did not object to such a provisions being made irrevocable, and that such a provision was probably unnecessary inasmuch as it was already then “implied constitutional law.”68 But even if we assume that slavery and the Constitution were incompatible in some way and that the Corwin Amendment, in protecting slavery, rendered the constitutional order more incoherent than it might otherwise have been, it would still have been difficult to contend successfully that the amendment was unconstitutional. In fact, it will rarely be possible to argue successfully that any amendment is unconstitutional. There are four reasons why.
First, politics is invariably tied in large part to the existence of tensions—which sometimes threaten to erupt into outright incompatibilities—among people, groups, institutions, and governments. Any constitution worthy of the name will reflect those tensions to some extent, which demonstrates how fragile coherence is as a constitutional standard. Consequently, in order for an amendment to render the text incoherent, it would have to be a stark departure from a widely accepted meaning of the text to which it is attached. The Corwin Amendment was not such a radical departure.
Second, constitutional meanings are not static. The content of constitutional categories and their relation to one another change over time. Part of the reason for this goes back to the tensions between vertical and horizontal coherence. As a result, the adoption of an amendment might be evidence of a shift in what the Constitution means to certain members of the polity. If it is, an argument that the amendment renders incoherent the text as a whole is a much trickier proposition.
Third, given the relations among the Constitution’s institutional actors, any expression of sentiment sufficiently strong to produce a constitutional amendment on a matter so fundamental would stand little chance of being successfully flouted by an established institution of government. Even an institution that sees its role as distinctly countermajoritarian—as the Supreme Court has at various times in the country’s history—would have difficulty withstanding the force of supermajorities strong enough to ratify an amendment. There is some chance that the Court, or some other institution, could successfully stand up to such a force, but it is surely miniscule.
Finally, substantive constitutional principle is almost always constrained by experience or practice. But in times when popular will is expressed in fairly concrete form, as it arguably is when an amendment is ratified, the constraint is a strong one. The capacity of a constitution (or more accurately, in light of considerations related to its particular allocation of political authority, the capacity of the Constitution) to enforce substantive limits on formal constitutional change is weak.
Still, one might like to hold open the theoretical possibility of challenging on substantive grounds the constitutional legitimacy of some amendments. Given that impulse, there are four related conditions pertinent to sustaining such a challenge. First, how explicitly is the value at stake embodied in the text of the Constitution? In general, though not always, the more explicit the value, the better the chance for a successful challenge. Second, what is the relation of the value at stake to other constitutional (or, because of the pull of the vertical axis, even extraconstitutional) values? The stronger the ties to other values, the greater the chance of a successful challenge. Third, how strongly does the amendment reflect an expression of public will? That is, by what procedure was the amendment proposed and ratified? Harris’s hierarchy of procedures is pertinent here. Fourth, how recently was the amendment ratified? This last consideration—time—cuts more than one way. On the one hand, recent ratification might indicate that public opinion on the issue remains strong, and passage of time might provide the polity an opportunity for reflection on and reconsideration of an amendment adopted in haste or in a time of political fervor. On the other hand, time can strengthen the inertial force of habit.
Under the “right” conditions, then, an institutional interpreter might be able to nullify an amendment that contradicts substantive constitutional values, once the better angels of whatever human nature there may be have spoken. But the confluence of the right conditions will probably be rare.
As for the Corwin Amendment, constitutional theory can guarantee no satisfying way out, no happy ending. Even if there had been a general textual commitment to the abolition of slavery, its pull on political society would have been weak, at least in the United States in the nineteenth century. The popular sovereign would have prevailed, not merely because it held the public sword, but also because it embodied something of the public will.69
Special thanks to Walter Murphy, Donald Kommers, and Judith Failer for comments on earlier drafts of this essay.
1 Walter F. Murphy, drawing on a decision of the Federal Constitutional Court of West Germany, explores the possibility of an unconstitutional provision of an original (unamended) constitution in Murphy, “An Ordering of Constitutional Values,” Southern California Law Review 53 (1980): 703, 755; Murphy, “Slaughter-House, Civil Rights, and Limits on Constitutional Change,” American Journal of Jurisprudence 32 (1987): 1.
2 William F. Harris II calls the notion “either a riddle, a paradox, or an incoherency.” The Interpretable Constitution (Baltimore: Johns Hopkins University Press, 1992), p. 169.
3 That is not to mention the possibility that the articulation and application of standards for limits to the amendability of a constitution might amount to an arrogation of institutional authority that potentially runs counter to processes of democratic decision making. For two accounts of democratic theory that address this problem, if it is a problem, see William Rehnquist, “The Notion of a Living Constitution,” Texas Law Review 54 (1976): 693; and John Hart Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge: Harvard University Press, 1980).
4 Marbury v. Madison, 5 U.S. (1 Cranch) 137, 174 (1803).
5 This essay does not deal directly with two other questions related to the issue. First, what are the nature and source(s) of limits, if any, to the amendability of those textual provisions (in the American Constitution, Article V and perhaps other provisions) that purport to regulate the adoption of amendments? Second, who may constitutionally decide questions about the appropriateness of amendments to a constitution (or the Constitution)?
6 An early articulation of this fear is present in John C. Calhoun’s “Address in the Senate of the United States, on the Subject of Slavery” (March 4, 1850) (Princeton Microfiche Collection), pp. 2–4, 8.
7 Ibid.
8 Again, Calhoun was the most consistent and intelligent proponent of such a notion of state sovereignty. See, e.g., ibid., pp. 7, 9; and Calhoun, “Speech on the Importance of Domestic Slavery” (January 10, 1838), in Eric McKitrick, ed., Slavery Defended: Views of the Old South (1963), pp. 16–17.
9 James M. McPherson, Battle Cry of Freedom: The Civil War Era (New York: Oxford University Press, 1988), p. 239.
10 Note, for example, the Northwest Ordinance of 1787, the Missouri Compromise of 1820, and even Stephen Douglas’s popular sovereignty, which found expression in the Kansas-Nebraska Act of 1854.
11 This is especially true given the admission of California in 1850 and Oregon in 1859 as free states. And it was probably true despite the Supreme Court’s pronouncement in Dred Scott v. Sandford, 60 U.S. (19 How.) 393 (1857).
12 Interpretations such as those of Lysander Spooner and eventually of Frederick Douglass.
13 McPherson, Battle Cry of Freedom, pp. 229–31.
14 Ibid., pp. 235, 255–57.
15 “Report of the Committee of Thirteen” (December 31, 1860), Rep. No. 288, Index to the Reports of the Committees of the Senate of the United States for the Second Session of the Thirty-Sixth Congress.
16 Ibid. As originally proposed, the amendment read as follows: “Article XII [sic]. No amendment of this Constitution having for its object any interference within the States with the relation between their citizens and those described in section second of the first article of the Constitution as ‘all other persons,’ shall originate with any State that does not recognize that relation within its own limits, or shall be valid without the assent of every one of the States composing the Union.”
17 U.S. Statutes at Large, vol. 12, 1859–1863, ed. G. Sawyer (1863), p. 251.
18 Lincoln may have understood the amendment to protect slavery only where it then existed, but the amendment’s plain words do not suggest such a limitation.
19 Journal of the House of Representatives of the United States: Being the Second Session of the Thirty-Sixth Congress, p. 486. For an explanation of why the president’s approval has been deemed, both before and after Lincoln, unnecessary, see H. Ames, The Proposed Amendments to the Constitution of the United States During the First Century of its History (New York: B. Franklin, 1970), pp. 196, 295–96; Hollingsworth v. Virginia, 3 U.S. (Dall.) 378 (1798).
20 Ames, Proposed Amendments, p. 196.
21 The Corwin Amendment poses the problem fairly starkly, because either that amendment is itself unconstitutional for violating some substantive account of what the Constitution means, or any subsequent amendment purporting to delegate to Congress authority over slavery is unconstitutional for violating the Corwin Amendment. Some choice is unavoidable.
22 Hence, when Vile argues that the only limits on amendments to the Constitution are the procedural (and two substantive) limits expressed in Article V, he justifies his position in part by considering the consequences of the alternative position(s) for American constitutional government. Vile, “Limitations on the Constitutional Amending Process,” Constitutional Commentary 2 (1985): 373. And when Taft offers a similar argument, he justifies it by exploring the theoretical implications of both the Tenth Amendment’s recognition of “powers not delegated” and the nature of constitutional government. Taft, “Amendment of the Federal Constitution,” Virginia Law Review 16 (1930): 647.
23 Charles A. Black notes the significance of the generality and ambiguity of important provisions of the Constitution. See Structure and Relationship in Constitutional Law (Englewood Cliffs, N.J.: Prentice-Hall, 1969).
24 The categories “immanence” and “transcendence” are borrowed from William F. Harris II, “Bonding Word and Polity: The Logic of American Constitutionalism,” American Political Science Review 76 (1982): 34. They refer, respectively, to the location of values within or outside the Constitution. For reasons discussed more fully below, the categories are simultaneously useful and illusory, which may in turn suggest that the very notion of the Constitution is problematic.
25 Spooner argued that “the constitution of the United States must be made consistent throughout; and if any of its parts are irreconcilable with each other, those parts that are inconsistent with liberty, justice and right, must be thrown out for inconsistency.” The Unconstitutionality of Slavery (1860; reprinted, New York: Burt Franklin), p. 94.
26 Murphy, “An Ordering of Constitutional Values,” pp. 703, 745, 754.
27 Ibid., p. 757. For reasons that space will not permit me to explicate here, I suspect that Murphy’s theory of natural rights is prior to the Constitution, both temporally and theoretically. Moreover, despite Murphy’s ingenious interpretive claims, the theory has but a tenuous grounding in the Constitution of 1787 (or even of 1791). Consequently, I consider the theory to be transcendental.
28 Murphy, “Slaughter-House,” pp. 12, 17–18. Murphy goes on to supply criteria for determining what is “in” the Constitution. Among other things, he is concerned with justifying our treating the Declaration of Independence as part of the Constitution (pp. 19–20). For reasons having to do with both the pedigree of the Declaration and its incompatibility with values readily traceable to the Constitution, I am inclined to dispute the inclusion of the Declaration in the Constitution.
29 See n. 28, above.
30 One might speculate whether slavery itself might have been a fundamental constitutional principle (apart from questions concerning the allocation of governmental authority). A number of Garrisonian abolitionists in the middle of the nineteenth century pressed such a claim. Although he does not speak explicitly to the issue, Murphy would probably reject the proposition that slavery could be a fundamental constitutional principle, or at least that it could be a legitimate one.
31 Murphy offers a second criterion for determining whether an amendment to the Constitution might be unconstitutional: Does the amendment overturn certain compromises that made nationhood possible? “Slaughter-House,” pp. 12, 17–18. It is easy to argue that slavery was such a compromise. Indeed, the U.S. Supreme Court indirectly confirmed this notion when it declared the fugitive slave clause of the Constitution “a fundamental article, without the adoption of which the Union could not have been formed.” Prigg v. Pennsylvania, 16 Peters 539 (1842). As I read Murphy’s argument, however, the Corwin Amendment would not survive even if slavery had been fundamental to establishing nationhood. As I read him, natural right prevails in all circumstances, though Murphy has informed me that he disagrees with my interpretation of his theory.
32 Sotirios Barber, On What the Constitution Means (Baltimore: Johns Hopkins University Press, 1984), p. 43.
33 The language quoted is from Article V. The emphasis supplied is mine. Barber’s argument here, as he would probably concede, demonstrates the weakness of relying exclusively on textual pegs to support one’s position. His contention that this clause in Article V means that amendments must be part of (i.e., consistent with) the existing meaning of the whole document is plausible, but only in the context of corresponding theories of interpretation and the Constitution, which he later supplies. In the absence of such theories, an equally plausible interpretation of the same clause is that any amendment adopted by Article V’s procedures is valid as part of the Constitution, regardless of its (the amendment’s) content.
34 Interestingly, Barber notes only one such provision of Article V: the “equal Suffrage” provision. He does not mention the provision that temporarily prohibits amendments restricting the importation of slaves or imposing some forms of direct taxation.
35 Barber suggests that “substantive values,” at least those worthy of being followed, are values that are compatible with our long-term “aspirations,” as opposed to mere “immediate wants.” Barber, On What the Constitution Means, p. 48.
36 Ibid., pp. 43–46.
37 Ibid., pp. 49, 53, 54.
38 Barber understands “tradition” not as history or past practices, but as “a theory of what we stand for as a people.” What we stand for is fundamentally a moral imperative, “what we ought to have been doing and should continue to do whether we have, in fact, done it or will do it.” Tradition is “a normative theory of what has always been and therefore still is best in us as a people.” Ibid., pp. 84, 85. Similarly, “reason” for Barber is a transcendental category. Reasons are not rationalizations. Hence, reason is not immanent in the “logic” of a discourse; nor is it justified by reference to prevailing myths (including religion). Instead, reason is aimed at “truth itself.”
39 Ibid., pp. 34, 48, 49, 51, 57, 79.
40 Emphasis added.
41 Barber would recoil at the notion that his constitutional theory is conventionalist or historicist.
42 See, e.g., Thomas Jefferson, “Answers and Observations for Demeunier’s Article on the United States in the Encyclopedia Methodique” (1786), and Notes on the State of Virginia (1787), in Merrill Peterson, ed., Thomas Jefferson: Writings (New York: Viking, 1984), pp. 592, 298.
43 Murphy, “An Ordering of Constitutional Values,” p. 755, quoting from the Southwest Case, 1 BVerfGE 14 (1951).
44 Harris, The Interpretable Constitution.
45 Ibid., pp. 241, 250–51.
46 Two points deserve mention here. First, the notion of permanence is important, for one grounding principle of Harris’s constitutionalism is (ironically) that there shall be no orthodoxy. Hence, an “entrenching” constitutional provision may, under certain circumstances, be unconstitutional. Second, the rights that Harris refers to are apparently not the individual rights of modern liberty, but the (collective) rights necessary for self-government. Hence, the Ninth and Tenth Amendments, in which Murphy sees evidence of the natural (individual) rights that provide the foundation for his own constitutionalism and for much of liberal political theory, Harris sees as the textual embodiment of a “residual constitution-making power.” Ibid., p. 253.
47 This hierarchy, from lowest to highest, is: (1) proposal by the Congress with ratification by state legislatures; (2) proposal by the Congress with ratification by state conventions; (3) proposal by a national convention with ratification by state legislatures; (4) proposal by a national convention with ratification by state conventions; and (5) proposal and adoption by “the same … [procedures] by which … [the Constitution] was given authority in the first place.” Ibid., pp. 229–30, 235–36, 250–55.
48 This presumably refers to Article VII’s procedures.
49 Harris, The Interpretable Constitution, pp. 235–36, 242–49, 259–61.
50 The notion of “happy endings” is borrowed from Sanford Levinson, Constitutional Faith (Princeton: Princeton University Press, 1988), p. 87.
51 Hamilton, The Federalist, No. 1 (1787).
52 This account of constitutionalism is controversial. Walter Murphy, for example, derides it as a form of “constitutionism” that overlooks Israel, New Zealand, and Great Britain as constitutionalist societies. He also criticizes it for failing to incorporate certain substantive transcendental standards, without which Nazism, Stalinism, and Maoism become constitutionally permissible. I concede that my version of constitutionalism might exclude certain political societies—including some that are liberal with limited governments—that might claim to be constitutional and might even be attractive. I do not concede that my version necessarily sanctions Hitler or Stalin or Mao, though constitutionalism does permit a multitude of sins.
53 The distinction between these two categories is sometimes a blurry one, for ratification requires an element of interpretation, and interpretation amounts, in effect, to a kind of ratification.
54 I use the term experience in the broadest sense. And ethos here is also clearly an experiential notion, although it may be given content by what we tend to think of as matters of morality, even perhaps transcendental morality. I suspect that any moral sensibilities that make their way into the consciousness of (parts of) a society are largely, if not exclusively, functions of human experience rather than divine intercession or cosmic rationality. But in any event, in order for morality, transcendental or otherwise, to play an effective and legitimate role in the construction or evaluation of ostensibly authoritative constitutional meanings, it must first make itself either manifest or covertly effective throughout a significant portion of the population. Once it becomes so manifest or effective, regardless of its initial source, it appears ultimately to be conventional in application.
55 At the “micro” level, this correspondence is exemplified by Frederick Douglass’s decision to abandon an interpretation of the Constitution that was horizontally coherent but inconsistent (vertically) with his moral sensibilities in favor of a different interpretation that was also (but differently) horizontally coherent but also consistent (vertically) with his moral sensibilities.
56 David Hume, A Treatise of Human Nature (1739) ed. L. A. Selby-Biggs and P. H. Nidditch, 2d ed. (Oxford: Clarendon Press, 1978), p. 572.
57 Note Einstein’s suggestion to Heisenberg that “whether you can observe a thing or not depends on the theory which you use. It is the theory which decides what can be observed.” Quoted in Michael Polanyi, “Genius in Science,” Encounter 38 (1972): 48.
58 The presence of that defect (if it is a defect) is understandable in light of the fact that Murphy and Barber are trying to construct substantively principled ways of understanding the Constitution in our time, but part of the aim of this essay is to illuminate the limits to principle, or perhaps to suggest an alternative account of principle.
59 See Don E. Fehrenbacher, The Dred Scott Case: Its Significance in American Law and Politics (New York: Oxford University Press, 1978), pp. 16–17. See also Winthrop Jordan, White over Black: American Attitudes Toward the Negro, 1550–1812 (Chapel Hill: University of North Carolina Press, 1968).
60 Ely, Democracy and Distrust, pp. 77–101.
61 David Brion-Davis, The Problem of Slavery in the Age of Revolution: 1770–1823, (Ithaca: Cornell University Press, 1975), p. 256.
62 These claims are intentionally contentious and admittedly underdeveloped. The idea that the Constitution was a repudiation of the Revolution can be traced to several sources. Charles A. Beard was a proponent of the notion. See Beard, An Economic Interpretation of the Constitution of the United States (New York: Macmillan, 1913), pp. 19–51. Although Beard has accumulated a boatload of detractors, Gordon S. Wood seems to affirm that aspect of Beard’s thesis, even if he rejects Beard’s radical materialism. Wood, The Creation of the American Republic, 1776–1787 (Chapel Hill: University of North Carolina Press, 1969). Even Forrest McDonald, who sets himself up as a critic of Beard, concedes the influence of material interest, economic consolidation, and fiscal stabilization in the constitutional founding. McDonald, We the People: The Economic Origins of the Constitution (Chicago: University of Chicago Press, 1958); McDonald, E Pluribus Unum: The Formation of the American Republic, 1776–1790 (Boston: Houghton Mifflin, 1965).
63 See T. Jefferson, “Letter to John Holmes” (April 22, 1820), “Answers and Observations for Demeunier’s Article on the United States in the Encyclopedie Methodique” (1786), and Notes on the State of Virginia (1787), in Writings, pp. 1434, 592, 289.
64 W. Merrill, Against Wind and Tide: A Biography of Wm. Lloyd Garrison (Cambridge: Harvard University Press, 1963), p. 203.
65 See, e.g., Robert Cover, Justice Accused: Antislavery and the Judicial Process (New Haven: Yale University Press, 1975), pp. 8–30; William Nelson, “The Impact of the Anti-slavery Movement upon Styles of Judicial Reasoning in Nineteenth Century America,” Harvard Law Review 87 (1974): 513; William Wiecek, The Sources of Antislavery Constitutionalism in America, 1760–1848 (Ithaca: Cornell University Press, 1977).
66 See, e.g., Lysander Spooner, The Unconstitutionality of Slavery (1860).
67 Lincoln, “Reply to Illinois Delegation” (March 5, 1861), in The Collected Works of Abraham Lincoln, ed. R. Basler (New Brunswick, N.J.: Rutgers University Press, 1953), p. 275.
68 Lincoln, “First Inaugural Address—Final Text” (March 4, 1861), in ibid., pp. 262, 270.
69 If text and theory cannot constrain, however, there is a happy counterfactual ending to oppose the unhappy one that I have proposed. Had Henry Stanton been able to gain sufficient support for his proposed amendment abolishing slavery, even a constitution that was arguably proslavery would have been little protection against the amendment’s enforcement.