|
Article Excerpt
ARTICLE CONTENTS
INTRODUCTION I. OUR EXTRACANONICAL CONSTITUTION A. Extracanonical Materials and Constitutional Functions 1. Constituting the Government 2. Conferring Rights on Individuals 3. Entrenching Structures and Rights Against Change B. Three Cases 1. The Statutory Safeguards of Federalism: Gonzales v. Oregon 2. The Clean Water Act as a Constitution: Rapanos v. United States 3. The Extracanonical Constitution of War Powers: Hamdan v. Rumsfeld C. Extracanonical Functions II. ENTRENCHMENT AND CONSTITUTIONAL CHANGE A. The Rule of Recognition Problem B. Extracanonical Mechanisms of Constitutional Change C. Relative Entrenchment III. THE FUNCTIONAL BOUNDARIES OF CONSTITUTIONAL LAW A. Doctrine 1. Two Federal Courts Puzzles 2. The Continuity of Interpretation B. Pedagogy and Scholarship CONCLUSION
INTRODUCTION
There is the notion that the primary source of information as to what our Constitution comes to, is the language of a certain Document of 1789, together with a severely select coterie of additional paragraphs called Amendments. Is this not extraordinary? (1)
My central claim in this Article is that the American "constitution" consists of a much wider range of legal materials than the document ratified in 1789 and its subsequent amendments. To clarify what I mean, it will help to begin with a thought experiment derived from comparative constitutional experience. It has long been said that the English have an "unwritten" constitution. This, however, is clearly untrue. As Adam Tomkins has pointed out, "notwithstanding its allegedly unwritten nature, much (indeed, nearly all) of the [English] constitution is written, somewhere." (2) The Magna Carta, the Bill of Rights of 1689, the Parliament Acts of 1911 and 1949, the European Communities Act of 1972, the Human Rights Act of 1998 (3)--these all form parts of the English constitution, and they are all written down. As Professor Tomkins explains, "[t]he unhappily misleading phrase, 'written constitution' really means 'codified constitution.' Thus, a written, or codified, constitution is one in which all the principal constitutional rules are written down in a single document named 'The Constitution.'" (4) That single codified document is what the English lack.
In a polity without a codified constitution, the content of "The Constitution" must be derived functionally, not formally. Matthew Palmer has described this perspective as "constitutional realism" that "seeks to identify the nature of a constitution through observing its operation in reality." (5) The functional perspective predates the realist movement, however. As early as 1908, A.V. Dicey defined English constitutional law to include "all rules which directly or indirectly affect the distribution or the exercise of the sovereign power in the state." (6) Hence, we know that the Parliament Acts of 1911 and 1949--defining the role of the House of Lords in the legislative process--are part of the English Constitution because of what they do, not because they have any formal markers that set them off from ordinary legislation. (7)
The thought experiment that I wish to propose involves thinking of the American constitutional order in the same way, despite the fact that we purport to have a codified constitution. It is possible to identify, in the abstract, certain functions that constitutions perform. In England, whatever laws actually perform those functions are considered part of "the constitution." (8) What if we thought of the United States' legal system in the same way? What would our "constitution" look like then? (9)
My descriptive claim is that much--perhaps even most--of the "constitutional" work in our legal system is in fact done by legal norms existing outside what we traditionally think of as "the Constitution." (10) A constitution generally does three primary things: It constitutes the government, that is, it establishes the various institutions of the government and sets out their powers and obligations. It identifies certain rights of individuals against that government. And (sometimes) it entrenches these structures against change, absent compliance with a difficult amendment procedure. A moment's reflection, however, reveals that under our modern institutional arrangements, the first two of these functions are no longer exclusively, or even primarily, performed by constitutional norms. (I shall have more to say about the third function--entrenchment--later on.) For virtually all practical purposes, the boundary between federal and state power is set by the terms of federal statutes; likewise, statutes and regulations play a far more significant role in regulating the separation of powers at the national level than do constitutional rules. Many of our most important individual rights--rights against discrimination based on age or disability, rights to welfare, medical care, and social security--stem from statutes rather than the Constitution. Even the basic electoral structure of our democracy is created and regulated by an assortment of nonconstitutional federal and state law rules.
Consider, for example, the Federal Communications Act. (11) That Act divides authority between the Congress and the Executive by delegating certain functions to an agency; (12) it further delegates some tasks to state governments while reserving others to federal authority. (13) The Act also confers both substantive and procedural rights on regulated entities and individuals. (14) From a functional point of view, the Communications Act might truly be described as a "constitution" in its own right. To be sure, the Act is not "entrenched" in the sense that it can only be modified by constitutional amendment. On the other hand, the broad range of important interests, both individual and commercial, that the Act balances and protects ensures that it is, as a practical matter, quite difficult to alter in any sort of fundamental way. (15)
It is time we recognized and thought systematically about the fact that much of the law that constitutes our government and establishes our rights derives from legal materials outside the Constitution itself. When lawyers talk about the Constitution being "open ended," they generally mean that constitutional norms themselves can be extended to cover unforeseen changes in technology or mores: the Fourth Amendment now covers wiretapping; (16) the Due Process Clause now covers abortion. (17) The more important sense of open endedness, however, lies in the extent to which the Constitution permits basic constitutive questions to be answered by subconstitutional norms. My point is emphatically not that the Constitution is irrelevant to most of today's legal problems. However, its relevance typically takes the form of a set of outside limits and a source of general constitutional values. The particular rules enshrined in the Constitution will themselves rarely have significant bite on our most important constitutional controversies. (18)
We can thus better understand our legal order if we decouple the constitutive function of a constitution from the entrenchment function. Other scholars, from Karl Llewellyn in the 1930s to Bruce Ackerman, William Eskridge, John Ferejohn, and many others today, have recognized that our political order is constituted by norms existing outside the canonical document. (19) But they have insisted on treating these extracanonical norms as "higher law," which puts their theories on a collision course with Article V and creates a great deal of pressure to develop an alternative rule of recognition to identify those norms that have achieved this higher status. If one is going to confer entrenched constitutional status on a norm that has not gone through Article V ratification--on the institutional innovations of the New Deal, for example--then one needs a highly determinate way to identify both which norms have achieved this status and what their precise content is. This rule of recognition problem has loomed large in critiques of alternative theories of constitutional change, as well as more general approaches to interpretation predicated on a "living constitution." (20)
My aim is more modest. I want to suggest that the set of norms that "constitutes" our government is in fact much broader than the set of norms that is constitutionally entrenched. A statute like the 1934 Act creating the Federal Communications Commission (FCC) may be constitutive of an institution without having any "higher law" status making it more difficult to change. (21) Decoupling constitutive function from entrenched status decreases the pressure to confine the class of constitutive enactments to a narrow and precisely defined category of norms. In fact, it would be fair to say that most laws have some constitutive aspects, to the extent that they create a government post, empower an institution, or confer a right. The fact that ordinary laws perform these functions is important, but it does not make them any less ordinary.
Decoupling the constitutive and entrenchment functions has important implications for constitutional law. The first is to offer a relatively simple account of constitutional change outside the Article V amendment process. The second, more doctrinal implication is to undermine sharp distinctions between constitutional claims and claims under statutes and regulations, as those distinctions are currently applied or proposed in areas like statutory construction, federal jurisdiction, and civil rights remedies. Finally, broadening the definition of "constitutive" norms beyond those that are formally entrenched ought to expand the jurisdiction of constitutional scholars, both as to what we teach and what we study.
Part I of this Article discusses three primary functions of constitutions--establishing the institutions of government, conferring rights on individuals, and entrenching these structures against easy change--and demonstrates that ordinary law frequently performs each of these roles. Part II argues for decoupling the constitutive and entrenching functions and explores the implications of that move for theories of constitutional change. Part III then traces the implications of this approach for constitutional doctrine and scholarship.
I. OUR EXTRACANONICAL CONSTITUTION
Exposition of my argument requires ready terms differentiating between the document generally designated as "the Constitution"--the one ratified in 1789, formally amended several times since, and passed out in handy pocketsize booklets by the Federalist Society--and those legal norms existing outside that document that nonetheless perform constitutional functions. It will not do to distinguish between "written" and "unwritten," because (as in England) the overwhelming bulk of the "constitution outside the Constitution" is, in fact, written down in statutes and regulations. Functionally speaking, one might distinguish the "entrenched" Constitution that can only be amended through the rigorous Article V procedure from various unentrenched norms that may be changed by other processes, including ordinary legislation. I want to suggest, however, that entrenchment is more multifarious than binary and that ordinary legislation performs important entrenching functions. For lack of a better term, I will refer to a "canonical" Constitution and an "extracanonical" constitution that exists alongside the canonical text.
I begin with an overview of the ways in which extracanonical materials perform crucial constitutional functions in Our system. I then develop some case studies in greater detail, focusing on three decisions from the Supreme Court's 2005 Term. Finally, I conclude this Part with a brief typology of extracanonical functions.
A. Extracanonical Materials and Constitutional Functions
To make the case that much of the "constitutional" work in our legal system is done by extracanonical norms, one first needs a catalog of constitutional functions. I want to focus on three such functions here: First, constitutions "constitute" the government by creating governmental institutions, prescribing procedures by which those institutions operate, and allocating powers and responsibilities among the various institutions thus created. (22) Second, constitutions typically confer certain rights on individuals as against government action. (23) Finally, many constitutions entrench certain institutional structures and individual rights by making those arrangements relatively difficult to change. (24)
I do not insist on this particular typology of constitutional functions. Others have described them somewhat differently. (25) I expect my argument could be replicated for just about any other function one might attribute to a constitution. If constitutions are meant to embody the basic aspirations and values of a society, for example, then it is easy to cite examples where those basic commitments are more readily found in statute. Our national commitments to environmental stewardship, intergenerational responsibility, and a free market economy are easier to discern in the Clean Water Act, (26) the Social Security and Medicare regimes, (27) and the Sherman Act (28) than in the Constitution itself. In any event, my point is not to develop an exhaustive definition of constitutional functions, but simply to identify enough key functions to test the hypothesis that these functions are often performed by ordinary law.
1. Constituting the Government
The first function of a constitution is to "constitute" the government. This includes creating governmental institutions, specifying their composition and methods for selecting officers, conferring powers upon them, establishing operational procedures, and drawing the boundaries of their jurisdictions. Article I of the U.S. Constitution, to take the most obvious example, vests "[a]ll legislative Powers herein granted ... in a Congress," (29) divides that Congress into two houses, specifies the composition of each house and the apportionment of representatives among the states, confers enumerated powers on the institution thus created, and prescribes procedures by which legislation may be enacted. Articles II and III perform similar functions for the executive and judicial branches, albeit in considerably less detail.
All this is elementary. My claim, however, is that massively complex legal systems like our own require a great deal of constituting, and relatively little of it is done by the canonical Constitution. Of the 2,677,999 civilian persons employed by the national government in 2006, (30) only 546 were Presidents, Vice Presidents, Supreme Court Justices, or members of Congress. The rest served in positions created not by the Constitution, but by federal statutes or regulations. (31) Most of these officers are selected and supervised according to legislation creating the modern civil service. (32) Many great institutions of national government--the vast administrative bureaucracies of the Environmental Protection Agency (EPA), the FCC, or the Social Security Administration, for example--are nowhere to be found in the canonical Constitution. These institutions produce a solid majority of federal law. (33) But these powerful agencies are created by their organic statutes, organized according to presidential directives, and regulated by the Administrative Procedure Act (APA) and a host of judge-made "common law" requirements. (34)
Consider, for example, the Federal Reserve, which sets U.S. monetary policy and consequently affects the lives of every American. As a former governor notes, "[t]he Federal Reserve... is often called the most powerful institution in America." (35) One might expect to find such a powerful institution described in the Constitution. The proposed European Constitution, for example, devotes considerable space to the powers and structure of the European Central Bank. (36) But our Federal Reserve was created by statute in 1913, and many of the rules by which it operates are found in regulations promulgated by the Fed's Board of Governors. (37) It is hard to imagine what American economic policy would look like without this critical institution, yet it is "constituted" by legal materials existing entirely outside the canonical Constitution.
Or consider what the canonical Constitution does not tell us about Congress, the institution the document discusses most extensively. The two most important questions concerning who can serve in Congress concern the content of the electorate: Who can vote in elections for Congress, and by what system? And how are the members of a state's delegation to the House of Representatives to be apportioned geographically? Article I punts the first of these questions to the states, (38) although their freedom is now circumscribed by the Fourteenth and Fifteenth Amendments and, more pervasively, the Voting Rights Act. (39) As a result, voter qualifications are controlled primarily by a combination of state and federal statutory rules; most significant, the "first past the post" system for choosing House members is entirely a creature of statute and convention. (40) Apportionment is, of course, controlled by a constitutional rule--one man, one vote (41)--but it is a rule with fairly tenuous grounding in the canonical text and which might therefore be better described as a product of judge-driven constitutional evolution. (42) And the most important issue of current controversy concerning apportionment--the legality and legitimacy of political gerrymandering-remains largely ungoverned by constitutional constraints. (43) Many observers have noted that much of the character of our politics derives from the ability of state legislatures to create an overwhelming proportion of "safe" congressional seats, (44) yet this vital constitutive dynamic remains largely outside the constraints imposed by the canonical Constitution.
Once Congress is elected, its operations are likewise framed largely by extracanonical materials. True, the "finely wrought" process of bicameral consideration and presentment to the President is set forth in Article I, Section 7, and the Court has been quite unwilling to countenance legislative modifications to that procedure. (45) But there is a great deal more to the legislative process than bicameralism and presentment, and none of it is in the Constitution. Congress is pervasively structured along the lines of our two dominant political parties, which were largely unanticipated by the Framers and accordingly left entirely out of the canonical document. (46) The progress of legislation, moreover, is dominated by the committee system, which is a creature not even of statute but of internal House and Senate rules. Even the basic principle that a bare majority in each house is sufficient to approve legislation, not to mention some of the exceptions to that rule in the Senate, does not appear in the Constitution. It exists simply as a matter of legislative convention. (47) To see the point most starkly, imagine if Congress were forced to operate with only the rules actually set forth in the canonical text of the Constitution. It is hard to imagine how it could possibly proceed.
In noting these facts, I do not mean to argue that these extracanonical institutions and constitutive rules are "unconstitutional." The Constitution clearly contemplates that additional officers will serve alongside those specifically identified in the text: it empowers Congress to create Article III courts other than the Supreme Court, provides rules for the appointment of principal and inferior officers, and authorizes the President to require the written opinions of the "heads of departments." (48) And it is impossible to imagine that the Framers could have intended to deny Congress the power to organize itself or select an appropriate voting rule for the passage of legislation. The Constitution is not a suicide pact, (49) and it is also not a nonstarter. My point is simply that the canonical Constitution leaves a very great deal of this essential work to be done by other legal materials.
We might identify a narrower sense in which a constitution "constitutes" a government-that is, it provides a "rule of recognition" by which we can tell what norms count as "law" within our legal system. The rule of recognition, as developed by H.L.A. Hart, "provides validity criteria that, directly or indirectly, determine the legal status of all other rules." (50) It is tempting to suppose that providing such a rule is, in fact, a key function of constitutions. Hence, the validity of a legal rule in the U.S. legal system depends on its having been promulgated pursuant to the lawmaking procedures laid out in Article I and on its consistency with the individual rights articulated elsewhere in the document. If the Constitution does not actually establish all the institutions of our government, perhaps it at least sets forth the criteria by which the validity of those institutions (and their works) can be judged.
This recognition function has the virtue of being sufficiently fundamental to fit our intuition about the special dignity of constitutions. But it does not withstand scrutiny. Most scholars seem to agree that a rule of recognition is a social fact, in the sense that it identifies the criteria that will cause the relevant class of officials to accept a norm as a rule of law. (51) Obviously one cannot refer to the criteria of Article I or the Supremacy Clause's simple statement that "[t]his Constitution ... shall be the supreme Law of the Land" (52) to determine that the Constitution is, in fact, supreme law; after all, any spurious document (even the one you are reading) might likewise affirm that it is the supreme law. If we have to resort to some prior criterion to determine that the document drafted in Philadelphia in 1787, and not some competitor document, is the supreme law, then that criterion would provide the ultimate rule of recognition- not the Constitution itself. (53)
One might concede this much and say that while a constitution may require a priori validation at the outset, once it is in place it provides the validity criteria for all subsequent legal norms within the legal system. (54) This is certainly true to some extent: we are well accustomed to saying that an otherwise valid legal rule will be invalid if it transgresses some principle in the Constitution. But compliance with the Constitution's criteria is sometimes not a necessary, and often not a sufficient, condition for legal validity in our system. (55) As Bruce Ackerman has demonstrated, it is very difficult to square the adoption of the Reconstruction Amendments with the formal requirements for ratification in Article V. (56) The authority of those amendments thus must stem from some combination of traditional acceptance and current agreement with the values they embody. Hence, our preconstitutional ultimate rule of recognition, predicated on social acceptance, seems to operate even after the initial act of constitutional founding. Satisfaction of the document's own terms, in other words, is not always a necessary condition for legal validity in our system. (57)
More frequently, satisfaction of the constitutional criteria is not a sufficient condition for legal validity. All state laws, for instance, must also satisfy whatever validity criteria are set out in the relevant state constitution. (58) Even if we confine the inquiry to federal legal rules, the majority of those rules are administrative in character rather than statutory. As such, they must satisfy various validity criteria set out in the Administrative Procedure Act and the organic statute of the relevant administrative agency. More fundamentally still, even federal statutes must satisfy validity criteria other than those set out in the Constitution. That document, after all, never sets out the voting rules that will govern statutory enactment. The basic criterion that approval requires a bare majority of each house is thus fixed by custom, not by the Constitution itself.
The inescapable conclusion is that, while satisfaction of constitutional criteria is an important component of a legal rule's validity, those criteria hardly capture the entire range of conditions that rules must satisfy in order to be valid within our legal system. The ultimate rule of recognition in our system--in any system, most likely--is something so basic that it transcends even the constitutional text. And to the extent that the Constitution does serve as a rule of legal validity, this constitutive function, like the others I have already discussed, is shared between the constitutional text and a variety of other rules and provisions in the system.
2. Conferring Rights on Individuals
Conferring rights fits somewhat uncomfortably with the other constitutional functions that I have discussed so far. The others involve "constitutive" rules--rules about rulemaking, if you will--that are relatively distinct from "substantive" rules like prohibiting murder or setting the permissible amount of a chemical in the drinking water. Rights, by contrast, often seem more substantive in their orientation-the right to bear arms or to obtain an abortion, for instance. This difference from the constitutive functions of constitutions may be more apparent than real, however. Rights perform a key constitutive function by setting the bounds of government power and constraining the exercise of government discretion. (59) The great bulk of our rights guarantees are procedural in nature, (60) and provisions like the Speech, Press, and Establishment Clauses can all be thought of as "constituting" a transparent and open political process as well as a public space for political, social, and religious debate free of governmental distortion. Even the right to bear arms has a constitutive dimension, to the extent that it was originally intended to create a military counterweight to potentially oppressive governments. (61)
In any event, individual human rights loom so large in our modern constitutional consciousness that it would be impossible to leave them off of any reasonable list of constitutional functions. But one also does not have to look far to see that many of our most important human rights are not part of the canonical Constitution. To begin with, some of our canonical rights are dependent on rights created elsewhere. One of the most important rights in the early Republic was the right against state impairment of contracts. (62) That right, however, only kicks in once state law has recognized contractual rights in the first place. (63) The same thing is true of property rights: the Federal Constitution says they cannot be "taken" without just compensation, but they are generally created in the first instance by state law. (64)
The more basic point, however, is that many rights that are fundamental for individuals in modern America are entirely creatures of statute. Notwithstanding the Court's interpretation of the First Amendment in Texas v. Johnson, (65) I am unlikely--to put it mildly-to exercise my right to burn an American flag. I do worry that I or someone close to me might one day be discriminated against on the basis of race, gender, age, or disability, and in that event I would look first to the Civil Rights Act of 1964, (66) the Age Discrimination in Employment Act, (67) or the Americans with Disabilities Act (68) for protection and relief rather than the Equal Protection Clause, even if the perpetrator is a state actor. (69) Even more obviously, American constitutional culture has generally been reluctant to recognize positive rights to housing, food, health care, or economic security, but we have created elaborate statutory entitlements to such benefits under the Social Security, Medicare, Medicaid, Aid to Families with Dependent Children, and unemployment assistance regimes. (70) One suspects that millions of our citizens value these entitlements considerably more highly than many or most of their canonical rights. The point, in any event, is not to establish that statutory rights are more important than canonical ones--just that many important individual rights derive from extracanonical sources. (71)
Finally, and more controversially, there are rights conferred by international law. Agreements like the North American Free Trade Agreement (NAFTA) and the Vienna Convention on Consular Relations confer on foreigners in our midst important rights against American governments. (72) With respect to our own citizens, the traditional assumption has been that international human rights are redundant with domestic constitutional protections, and the Senate has often attached reservations to human rights treaties designed to ensure that international rights sweep no more broadly than domestic constitutional rights within the domestic legal system....
|
