[Paleopsych] Ha'va'd L. Rev. Symposium: Changing Images of the State

[Premise Checker]

Symposium: Changing Images of the State: The Rise and Fall of the 
Administrative State
Ha'va'd Law Review, 94.4
107 Harv. L. Rev. 1231

Gary Lawson

Associate Professor, Northwestern University School of Law. B.A. 1980, 
Claremont Men's College; J.D. 1983, Yale Law School. I am grateful to Robert W. 
Bennett, Steven G. Calabresi, Cynthia R. Farina, Patricia B. Granger, Daniel 
Polsby, Martin H. Redish, Jennifer Roback, Marshall Shapo, and the participants 
at colloquia at Cornell Law School and Northwestern University School of Law 
for their insightful comments and suggestions.

SUMMARY:
    ... The post-New Deal administrative state is unconstitutional, and its 
validation by the legal system amounts to nothing less than a bloodless 
constitutional revolution. ... Furthermore, I am aware of only one instance in 
the Reagan-Bush era in which the Justice Department formally opposed 
legislation on such grounds; a 1986 opinion from the Office of Legal Counsel 
stated that Congress did not have the enumerated power to enact a national 
lottery. ... Significantly, that power to execute the laws is vested, not in 
the executive department of the national government, but in "a President of the 
United States of America." ... Indeed, most contemporary scholars believe that 
Congress may vest discretionary authority in subordinate officers free from 
direct presidential control, and early American history and practice reflect 
this view to a considerable extent. ... This overlap between the executive and 
judicial functions is not surprising; under many pre-American conceptions of 
separation of powers, the judicial power was treated as an aspect of the 
executive power. ... Furthermore, if Professor Ackerman is correct that the 
Reconstruction Amendments were invalid under formal constitutional rules of 
ratification, the obvious conclusion seems to be that both the Reconstruction 
Amendments and the modern administrative state are unconstitutional. ... 
According to McCutchen, the administrative state is here to stay, and even a 
very weak theory of precedent ratifies this result. ...

---------------

   [*1231] The post-New Deal administrative state is unconstitutional, 1 and its 
validation by the legal system amounts to nothing less than a bloodless 
constitutional revolution. 2 The original New Dealers were aware, at least to 
some degree, that their vision of the national government's proper role and 
structure could not be squared with the written Constitution: 3 The 
Administrative Process, James Landis's classic exposition of the New Deal model 
of administration, fairly drips with contempt for the idea of a limited 
national government subject to a formal, tripartite separation of powers. 4 
Faced with a choice between the administrative state and the Constitution, the 
architects of our modern government chose the administrative state, and their 
choice has stuck.

There is a perception among some observers, however, that this post-New Deal 
consensus has recently come under serious legal attack, [*1232] especially from 
the now-departed Reagan and Bush administrations. 5 But though debate about 
structural constitutional issues has clearly grown more vibrant over the past 
few decades, 6 the essential features of the modern administrative state have, 
for more than half a century, been taken as unchallengeable postulates by 
virtually all players in the legal and political worlds, including the Reagan 
and Bush administrations. The post-New Deal conception of the national 
government has not changed one iota, nor even been a serious subject of 
discussion, since the Revolution of 1937. 7

Part I of this Article sketches, in purely descriptive fashion, some of the 
most important ways in which the modern administrative state, without serious 
opposition, contravenes the Constitution's design. 8 Many elements of this 
design remain poorly understood even after more than two centuries, and my 
brief discussion here is unlikely to be satisfying. Nonetheless, my discussion 
at least touches on such important issues as the scope of Congress's 
legislative powers, the contours and constitutional source of the nondelegation 
doctrine, the character of the unitary executive created by Article II, and the 
extent to which administrative adjudication is inconsistent with Article III.

Part II briefly ponders some possible responses to the enormous gap between 
constitutional meaning and constitutional practice. For those of us for whom 
the written Constitution (as validly amended) is the only Constitution, 9 the 
seemingly irrevocable entrenchment of the [*1233] post-New Deal structure of 
national governance raises serious doubts about the utility of constitutional 
discourse.

I. THE DEATH OF CONSTITUTIONAL GOVERNMENT

The United States Congress today effectively exercises general legislative 
powers, in contravention of the constitutional principle of limited powers. 
Moreover, Congress frequently delegates that general legislative authority to 
administrative agencies, in contravention of Article I. Furthermore, those 
agencies are not always subject to the direct control of the President, in 
contravention of Article II. In addition, those agencies sometimes exercise the 
judicial power, in contravention of Article III. Finally, those agencies 
typically concentrate legislative, executive, and judicial functions in the 
same institution, in simultaneous contravention of Articles I, II, and III.

In short, the modern administrative state openly flouts almost every important 
structural precept of the American constitutional order.

A. The Death of Limited Government

The advocates of the Constitution of 1789 were very clear about the kind of 
national government they sought to create. As James Madison put it: "The powers 
delegated by the proposed Constitution to the federal government are few and 
defined." 10 Those national powers, Madison suggested, would be "exercised 
principally on external objects, as war, peace, negotiation, and foreign 
commerce," 11 and the states would be the principal units of government for 
most internal matters. 12

The expectations of founding-era figures such as James Madison are instructive 
but not controlling for purposes of determining the Constitution's original 
public meaning: the best laid schemes o' mice, men and framers gang aft a-gley. 
The Constitution, however, is well designed to limit the national government 
essentially to the functions described by Madison.

Article I of the Constitution vests in the national Congress "[a]ll legislative 
powers herein granted," 13 and thus clearly indicates that the national 
government can legislate only in accordance with enumerations [*1234] Article I 
then spells out seventeen specific subjects to which the federal legislative 
power extends: such matters as taxing and borrowing, interstate and foreign 
commerce, naturalization and bankruptcy, currency and counterfeiting, post 
offices and post roads, patents and copyrights, national courts, piracy and 
offenses against the law of nations, the military, and the governance of the 
nation's capital and certain federal enclaves. 15 Article IV further grants to 
Congress power to enforce interstate full-faith-and-credit requirements, to 
admit new states, and to manage federal territories and property. 16 Article V 
grants Congress power to propose constitutional amendments. 17

This is not the stuff of which Leviathan is made. None of these powers, alone 
or in combination, grants the federal government anything remotely resembling a 
general jurisdiction over citizens' affairs. The Commerce Clause, for example, 
is a grant of power to regulate "Commerce . . . among the several States," 18 
not to regulate "all Activities affecting, or affected by, Commerce . . . among 
the several States." The Commerce Clause clearly leaves outside the national 
government's jurisdiction such important matters as manufacturing (which is an 
activity distinct from commerce), 19 the terms, formation, and execution of 
contracts that cover subjects other than the interstate shipment of goods, and 
commerce within a state's boundaries.

Nor does the Necessary and Proper Clause, 20 which the founding generation 
called the Sweeping Clause, grant general legislative powers to the national 
government. This clause contains two significant internal limitations. First, 
it only validates laws that "carry[] into Execution" other granted powers. To 
carry a law or power "into Execution" means to provide the administrative 
machinery for its enforcement; it does not mean to regulate unenumerated 
subjects in order to make the exercise of enumerated powers more effective. 21 
Second, and more fundamentally, laws enacted pursuant to the Sweeping Clause 
must be both "necessary and proper" for carrying into [*1235] execution 
enumerated powers. As Patty Granger and I have elsewhere demonstrated at 
length, the word "proper" in this context requires executory laws to be 
distinctively and peculiarly within the jurisdictional competence of the 
national government -- that is, consistent with background principles of 
separation of powers, federalism, and individual rights. 22 Thus, the Sweeping 
Clause does not grant Congress power to regulate unenumerated subjects as a 
means of regulating subjects within its constitutional scope. 23

Nor does the power of the purse give Congress unlimited authority, though here 
the limits are a bit fuzzy. The Constitution contains a Taxing Clause, 24 but 
it does not contain a "spending clause" as such. 25 Nevertheless, Congress 
acquires the power to spend from two sources. First, the Sweeping Clause 
permits Congress to pass appropriations laws -- provided that such laws 
"carry[] into Execution" an enumerated power and are "necessary and proper" for 
doing so. Second, as David Engdahl has pointed out, Congress's power "to 
dispose of and make all needful Rules and Regulations respecting the Territory 
or other Property belonging to the United States" 26 seems to provide a general 
spending power. 27 But a general power to spend is not a general power to 
regulate. Even if Congress can impose whatever conditions it pleases on the 
receipt of federal funds, those conditions are contractual in nature, a fact 
that limits both their enforceability and their scope. 28 Moreover, it is not 
obvious that the Property Clause gives Congress wholly unlimited power to enact 
spending conditions, though identifying any limits on such power would require 
careful consideration of the meaning of the phrase "dispose of," the 
relationship between the Property Clause and the Sweeping Clause, and the 
viability of a general doctrine of unconstitutional conditions.

Admittedly, some post-1789 amendments to the Constitution expand Congress's 
powers beyond their original limits. For example, the Thirteenth and Fifteenth 
Amendments authorize Congress to enforce prohibitions against, respectively, 
involuntary servitude 29 and racially discriminatory voting practices; 30 the 
Fourteenth Amendment [*1236] gives Congress power to enforce that Amendment's 
numerous substantive constraints on states; 31 and the Sixteenth Amendment 
permits Congress to impose direct taxes without an apportionment requirement. 
32 These are important powers, to be sure, but they do not fundamentally alter 
the limited scope of Congress's power over private conduct.

Of course, in this day and age, discussing the doctrine of enumerated powers is 
like discussing the redemption of Imperial Chinese bonds. There is now 
virtually no significant aspect of life that is not in some way regulated by 
the federal government. This situation is not about to change. Only twice since 
1937 has the Supreme Court held that a congressional statute exceeded the 
national government's enumerated powers, 33 and one of those holdings was 
overruled nine years later. 34 Furthermore, both cases involved the direct 
regulation of state governments in their sovereign capacities. To the best of 
my knowledge, the post-New Deal Supreme Court has never invalidated a 
congressional intrusion into private affairs on ultra vires grounds; instead 
the Court has effectively acquiesced in Congress's assumption of general 
legislative powers. 35

The courts, of course, are not the only, or even the principal, interpreters of 
the Constitution. Under the Constitution, it is emphatically the province and 
duty of the President to say what the law [*1237] is 36 -- and hence to veto 
bills that contravene constitutional limits. During their twelve years in 
office, however, the Reagan and Bush administrations made no serious attempt to 
resuscitate the doctrine of enumerated powers. I do not know of a single 
instance in which President Reagan or President Bush vetoed or even opposed 
legislation on the ground that it exceeded Congress's enumerated powers. 
Furthermore, I am aware of only one instance in the Reagan-Bush era in which 
the Justice Department formally opposed legislation on such grounds; a 1986 
opinion from the Office of Legal Counsel stated that Congress did not have the 
enumerated power to enact a national lottery. 37

Thus, the demise of the doctrine of enumerated powers, which made possible the 
growth of the modern regulatory state, has encountered no serious real-world 
legal or political challenges, and none are on the horizon.

B. The Death of the Nondelegation Doctrine

The Constitution both confines the national government to certain enumerated 
powers and defines the institutions of the national government that can 
permissibly exercise those powers. Article I of the Constitution provides that 
"[a]ll legislative Powers herein granted shall be vested in a Congress of the 
United States, which shall consist of a Senate and House of Representatives." 
38 Article II provides that "[t]he executive Power shall be vested in a 
President of the United States of America." 39 Article III specifies that 
"[t]he judicial Power of the United States, shall be vested in one supreme 
Court, and in such inferior Courts as the Congress may from time to time ordain 
and establish." 40 The Constitution thus divides the powers of the national 
government into three categories -- legislative, executive, and judicial -- and 
vests such powers in three separate institutions. To be sure, the Constitution 
expressly prescribes some deviations from a pure tripartite scheme of 
separation, 41 but this only underscores the role of [*1238] the three Vesting 
Clauses in assigning responsibility for governmental functions that are not 
specifically allocated by the constitutional text.

Although the Constitution does not contain an express provision declaring that 
the Vesting Clauses' allocations of power are exclusive, 42 it is a mistake in 
principle to look for such an express declaration. 43 The institutions of the 
national government are creatures of the Constitution and must find 
constitutional authorization for any action. Congress is constitutionally 
authorized to exercise "[a]ll legislative Powers herein granted," the President 
is authorized to exercise "[t]he executive Power," and the federal courts are 
authorized to exercise "[t]he judicial Power of the United States." Congress 
thus cannot exercise the federal executive or judicial powers for the simple 
reason that the Constitution does not vest such power in Congress. Similarly, 
the President and the federal courts can exercise only those powers vested in 
them by the Constitution: the general executive and judicial powers, 
respectively, plus a small number of specific powers outside those 
descriptions. Thus, any law that attempts to vest legislative power in the 
President or in the courts is not "necessary and proper for carrying into 
Execution" constitutionally vested federal powers and is therefore 
unconstitutional. 44

Although the Constitution does not tell us how to distinguish the legislative, 
executive, and judicial powers from each other, 45 there is [*1239] clearly 
some differentiation among the three governmental functions, which at least 
generates some easy cases. Consider, for example, a statute creating the 
Goodness and Niceness Commission and giving it power "to promulgate rules for 
the promotion of goodness and niceness in all areas within the power of 
Congress under the Constitution." If the "executive power" means simply the 
power to carry out legislative commands regardless of their substance, then the 
Goodness and Niceness Commission's rulemaking authority is executive rather 
than legislative power and is therefore valid. But if that is true, then there 
never was and never could be such a thing as a constitutional principle of 
nondelegation -- a proposition that is belied by all available evidence about 
the meaning of the Constitution. Accordingly, the nondelegation principle, 
which is textually embodied in the command that all executory laws be 
"necessary and proper," constrains the substance of congressional enactments. 
Certain powers simply cannot be given to executive (or judicial) officials, 
because those powers are legislative in character.

A governmental function is not legislative, however, merely because it involves 
some element of policymaking discretion: it has long been understood that some 
such exercises of discretion can fall within the definition of the executive 
power. 46 The task is therefore to determine when a statute that vests 
discretionary authority in an executive (or judicial) officer has crossed the 
line from a necessary and proper implementing statute to an unnecessary and/or 
improper delegation of distinctively legislative power. While I cannot complete 
that task here, the core of the Constitution's nondelegation principle can be 
expressed as follows: Congress must make whatever policy decisions are 
sufficiently important to the statutory scheme at issue so that Congress must 
make them. Although this circular formulation may seem farcical, 47 it 
recognizes that a statute's required degree of specificity depends on context, 
takes seriously the well-recognized distinction between legislating and 
gap-filling, and corresponds reasonably well to judicial application of the 
nondelegation principle in the first 150 years of the nation's history. 48 If 
it does not precisely capture [*1240] the true constitutional rule of 
nondelegation, it is a plausible first approximation. 49

In any event, it is a much better approximation of the true constitutional rule 
than is the post-New Deal positive law. The Supreme Court has not invalidated a 
congressional statute on nondelegation grounds since 1935. 50 This has not been 
for lack of opportunity. The United States Code is filled with statutes that 
create little Goodness and Niceness Commissions -- each confined to a limited 
subject area such as securities, 51 broadcast licenses, 52 or (my personal 
favorite) imported tea. 53 These statutes are easy kills under any plausible 
interpretation of the Constitution's nondelegation principle. The Supreme 
Court, however, has rejected so many delegation challenges to so many utterly 
vacuous statutes that modern nondelegation decisions now simply recite these 
past holdings and wearily move on. 54 Anything short of the Goodness and 
Niceness Commission, it seems, is permissible. 55

   [*1241] The rationale for this virtually complete abandonment of the 
nondelegation principle is simple: the Court believes -- possibly correctly -- 
that the modern administrative state could not function if Congress were 
actually required to make a significant percentage of the fundamental policy 
decisions. Judicial opinions candidly acknowledge this rationale for permitting 
delegations. For example, the majority in Mistretta v. United States 56 
declared that "our jurisprudence has been driven by a practical understanding 
that in our increasingly complex society, replete with ever changing and more 
technical problems, Congress simply cannot do its job absent an ability to 
delegate power under broad general directives." 57 When faced with a choice 
between the Constitution and the structure of modern governance, the Court has 
had no difficulty making the choice.

Contrary to conventional wisdom, neither did the Reagan and Bush 
administrations. Neither President Reagan nor President Bush ever vetoed or 
opposed legislation on the express ground that it violated the nondelegation 
doctrine. Nor, to my knowledge, did the Reagan-Bush Justice Departments ever 
formally make such an objection to proposed or actual legislation. 58

Thus, the demise of the nondelegation doctrine, which allows the national 
government's now-general legislative powers to be exercised by administrative 
agencies, has encountered no serious real-world legal or political challenges, 
and none are on the horizon.

C. The Death of the Unitary Executive

Article II states that "[t]he executive Power shall be vested in a President of 
the United States of America." 59 Although the precise contours of this 
"executive Power" are not entirely clear, 60 at a minimum it includes the power 
to execute the laws of the United States.  [*1242] Other clauses of the 
Constitution, such as the requirement that the President "take Care that the 
Laws be faithfully executed," 61 assume and constrain this power to execute the 
laws, 62 but the Article II Vesting Clause is the constitutional source of this 
power -- just as the Article III Vesting Clause is the constitutional source of 
the federal judiciary's power to decide cases. 63

Significantly, that power to execute the laws is vested, not in the executive 
department of the national government, but in "a President of the United States 
of America." 64 The Constitution thus creates a unitary executive. Any 
plausible theory of the federal executive power must acknowledge and account 
for this vesting of the executive power in the person of the President.

Of course, the President cannot be expected personally to execute all laws. 
Congress, pursuant to its power to make all laws "necessary and proper for 
carrying into Execution" the national government's powers, can create 
administrative machinery to assist the President in carrying out legislatively 
prescribed tasks. But if a statute vests discretionary authority directly in an 
agency official (as do most regulatory statutes) rather than in the President, 
the Article II Vesting Clause seems to require that such discretionary 
authority be subject to the President's control. 65

This model of presidential power is not without its critics. Indeed, most 
contemporary scholars believe that Congress may vest discretionary authority in 
subordinate officers free from direct presidential control, 66 and early 
American history and practice reflect this view to a considerable extent. 67 
Nonetheless, the Vesting Clause inescapably [*1243] vests "the executive Power" 
directly and solely in the person of the President. Accordingly, scholars 
sometimes deny that the Article II Vesting Clause is a grant of power to the 
President to execute the laws, 68 but none has yet adequately rebutted the 
compelling textual and structural arguments for reading the Vesting Clause as a 
grant of power 69 -- a grant of power specifically and exclusively to "a 
President of the United States."

Thus, the important question is what form the President's power of control over 
subordinates must take in order to ensure a constitutionally unitary executive. 
There are two evident possibilities. First, the President might be thought to 
have the power personally to make all discretionary decisions involving the 
execution of the laws. On this understanding, the President can step into the 
shoes of any subordinate and directly exercise that subordinate's statutory 
powers. 70 Second, one might think that, although the President cannot directly 
exercise power vested by statute in another official, any action by that 
subordinate contrary to presidential instructions is void. 71 Either 
alternative is plausible, though the latter is perhaps more consistent with 
Congress's power under the Sweeping Clause to structure the executive 
department. 72

   [*1244] Congress and the President have fought hard in recent years over 
control of the federal administrative machinery, and the courts have 
adjudicated such disputes in some high-profile cases. 73 Significantly, 
however, neither of the two possible constitutional mechanisms of presidential 
control has played a role in those battles. No modern judicial decision 
specifically addresses the President's power either directly to make all 
discretionary decisions within the executive department or to nullify the 
actions of insubordinate subordinates. Instead, debate has focused almost 
exclusively on whether and when the President must have unlimited power to 
remove subordinate executive officials. That is an interesting and important 
question, but it does not address the central issue concerning the executive 
power. Even if the President has a constitutionally unlimited power to remove 
certain executive officials, that power alone does not satisfy the Article II 
Vesting Clause. If an official exercises power contrary to the President's 
directives and is then removed, one must still determine whether the official's 
exercise of power is legally valid. If the answer is "no," then the President 
necessarily has the power to nullify discretionary actions of subordinates, and 
removal is therefore not the President's sole power of control. If the answer 
is "yes," then the insubordinate ex-official will have effectively exercised 
executive power contrary to the President's wishes, which contravenes the 
vesting of that power in the President. A presidential removal power, even an 
unlimited removal power, is thus either constitutionally superfluous or 
constitutionally inadequate. 74 Congress, the President, and the courts [*1245] 
have accordingly been spending a great deal of energy arguing about something 
of relatively little constitutional significance.

The death of the unitary executive cannot be traced to the New Deal revolution. 
The First Congress, in the so-called Decision of 1789, engaged in one of the 
most spirited and sophisticated debates on executive power in the nation's 
history, but did not once focus on a presidential power to make discretionary 
decisions or to veto actions by subordinates. 75 Moreover, many Attorneys 
General in the nineteenth century affirmatively denied that the President must 
always have the power to review decisions by subordinates. 76 The absence of a 
functioning unitary executive principle, however, may well have made the 
Revolution of 1937 possible. Judging from the political conflict that is often 
generated by disputes between Congress and the President, it is at least 
arguable that Congress would never have granted agencies their current, 
almost-limitless powers if Congress recognized that such power had to be 
directly under the control of the President. 77

Although the Reagan and Bush administrations often fought hard to defend their 
views of the proper role of the President, they did not directly assert their 
power to invalidate discretionary actions of subordinates or to make 
discretionary executive decisions when statutes confer power directly on 
subordinates. Opinions of the Office of Legal Counsel from the Reagan-Bush era 
have sometimes insisted that congressional attempts to place executive 
authority beyond presidential supervision are unconstitutional, 78 but neither 
President Reagan nor President Bush ever made either of the two plausible 
conceptions of [*1246] the unitary executive the focal point of a separation of 
powers dispute. The unitary executive has met its fate almost as meekly as have 
the principles of enumerated powers and nondelegation.

D. The Death of the Independent Judiciary

Article III provides that "[t]he judicial Power of the United States, shall be 
vested in one supreme Court, and in such inferior Courts as the Congress may 
from time to time ordain and establish." 79 The judges of all such federal 
courts are constitutionally guaranteed tenure during good behavior as well as 
assurance that their salaries will not be diminished during their time in 
office. 80 One of the principal functions of administrative agencies is to 
adjudicate disputes, yet "administrative adjudicators plainly lack the 
essential attributes that Article III requires of any decisionmaker invested 
with 'the judicial Power of the United States.'" 81 Is adjudication by 
administrative agencies therefore another instance of abandonment of a 
fundamental constitutional principle?

Maybe. Administrative adjudication is problematic only if it must be considered 
an exercise of judicial power. But an activity is not exclusively judicial 
merely because it is adjudicative -- that is, because it involves the 
application of legal standards to particular facts. Much adjudicative activity 
by executive officials -- such as granting or denying benefits under 
entitlement statutes -- is execution of the laws by any rational standard, 82 
though it also fits comfortably within the concept of the judicial power if 
conducted by judicial officers. 83 This overlap between the executive and 
judicial functions is not surprising; under many pre-American conceptions of 
separation of powers, the judicial power was treated as an aspect of the 
executive power. 84

Agency adjudication is therefore constitutionally permissible under Article III 
as long as the activity in question can fairly fit the definition of executive 
power, even if it also fairly fits the definition of judicial power. Some forms 
of adjudication, however, are quintessentially judicial. The conviction of a 
defendant under the criminal laws, for [*1247] example, is surely something 
that requires the exercise of judicial rather than executive power. Although it 
is difficult to identify those activities that are strictly judicial in the 
constitutional sense, perhaps Justice Curtis had the right answer in Murray's 
Lessee v. Hoboken Land & Improvement Co. 85 when he suggested that the Article 
III inquiry merges with questions of due process: if the government is 
depriving a citizen of "life, liberty, or property," 86 it generally must do so 
by judicial process, which in the federal system requires an Article III court; 
87 but if it is denying a citizen (to use discredited but useful language) a 
mere privilege, it can do so by purely executive action. Wherever the line is 
drawn, however, at least some modern administrative adjudication undoubtedly 
falls squarely on the judicial side. Most notably, the imposition of a civil 
penalty or fine is very hard to distinguish from the imposition of a criminal 
sentence (especially when the criminal sentence is itself a fine). If the 
latter is judicial, it is difficult to see why the former is not as well.

Some scholars believe that administrative adjudication is constitutionally 
permissible as long as the administrative decisions are subject to Article III 
appellate court review that is "adequately searching" 88 and "meaningful." 89 
And there's the rub. An agency's interpretation of a statute that it 
administers receives considerable deference under current law. 90 More 
fundamentally, agency fact-finding is generally subject to deferential review 
under numerous statutes that expressly require courts to affirm agency factual 
conclusions that are supported by "substantial evidence." 91 This kind of 
deferential review arguably fails to satisfy Article III. Article III certainly 
would not be satisfied if Congress provided for judicial review but ordered the 
courts to affirm the agency no matter what. That would effectively vest the 
judicial power either in the agency or in Congress. There is no reason to think 
that it is any different if Congress instead simply orders courts to put a 
thumb (or perhaps two forearms) on the [*1248] agency's side of the scale. I do 
not make this claim with full confidence (and thus do not emphasize the Reagan 
and Bush administrations' failure to advance it), but it seems to me that 
Article III requires de novo review, of both fact and law, of all agency 
adjudication that is properly classified as "judicial" activity. Much of the 
modern administrative state passes this test, but much of it fails as well.

E. The Death of Separation of Powers

The constitutional separation of powers is a means to safeguard the liberty of 
the people. 92 In Madison's famous words, "[t]he accumulation of all powers, 
legislative, executive, and judiciary, in the same hands, whether of one, a 
few, or many, and whether hereditary, self-appointed, or elective, may justly 
be pronounced the very definition of tyranny." 93 The destruction of this 
principle of separation of powers is perhaps the crowning jewel of the modern 
administrative revolution. Administrative agencies routinely combine all three 
governmental functions in the same body, and even in the same people within 
that body. 94

Consider the typical enforcement activities of a typical federal agency -- for 
example, of the Federal Trade Commission. 95 The Commission promulgates 
substantive rules of conduct. The Commission then considers whether to 
authorize investigations into whether the Commission's rules have been 
violated. If the Commission authorizes an investigation, the investigation is 
conducted by the Commission, which reports its findings to the Commission. If 
the Commission thinks that the Commission's findings warrant an enforcement 
action, the Commission issues a complaint. The Commission's complaint that a 
Commission rule has been violated is then prosecuted by the Commission and 
adjudicated by the Commission. This Commission adjudication can either take 
place before the full Commission or before a semi-autonomous Commission 
administrative law judge. If the Commission chooses to adjudicate before an 
administrative law judge rather than before the Commission and the decision is 
adverse to the Commission, the Commission can appeal to the Commission. If the 
Commission ultimately finds a violation, then, and only then, the affected 
private party can appeal to an Article III court. 96 But the agency decision, 
even before the bona fide Article III tribunal, [*1249] possesses a very strong 
presumption of correctness on matters both of fact and of law.

This is probably the most jarring way in which the administrative state departs 
from the Constitution, and it typically does not even raise eyebrows. The 
post-New Deal Supreme Court has never seriously questioned the 
constitutionality of this combination of functions in agencies. 97 Nor, to the 
best of my knowledge, did Presidents Reagan or Bush ever veto or object to 
legislation on this ground.

II. WHAT IS TO BE DONE?

The actual structure and operation of the national government today has 
virtually nothing to do with the Constitution. There is no reasonable prospect 
that this circumstance will significantly improve in the foreseeable future. If 
one is not prepared (as I am) to hold fast to the Constitution though the 
heavens may fall, what is one supposed to do with that knowledge?

One option, of course, is to argue directly that the Constitution, properly 
interpreted in accordance with its original public meaning, is actually 
flexible enough to accommodate the modern administrative state. But although 
some of the claims I make in Part I with respect to Articles II and III may 
ultimately prove to be wrong in some important respects, the most fundamental 
constitutional problems with modern administrative governance -- unlimited 
federal power, rampant delegations of legislative authority, and the 
combination of functions in administrators -- are not even remotely close 
cases. The Commerce Clause does not give Congress jurisdiction over all human 
activity, 98 and the Sweeping Clause does not give Congress carte blanche to 
structure the government any way it chooses. 99

   [*1250] A second option is to insist that the administrative state can be 
reconciled with the Constitution if only we reject the methodology of original 
public meaning. I cannot enter here into a discussion of interpretative theory, 
but for those of us who believe that "a dash of salt" refers to some 
identifiable, real-world quantity of salt, 100 originalist interpretivism is 
not simply one method of interpretation among many -- it is the only method 
that is suited to discovering the actual meaning of the relevant text. 101

A third option, pursued at length by Bruce Ackerman, is to argue that the 
Constitution has been validly amended, through means other than the formal 
process of Article V, in a fashion that constitutionalizes the administrative 
state. 102 Professor Ackerman claims that the [*1251] ratifications of the 
original Constitution and the Reconstruction Amendments were knowingly 
"illegal" under then-governing formal norms for the ratification of fundamental 
law. 103 The New Deal, he contends, reflected a similarly self-conscious 
rejection of the formal mechanisms for constitutional change. According to 
Professor Ackerman, if the formally deficient "ratifications" of the 
Constitution and the Reconstruction Amendments are legally valid, it is 
difficult to see why the same cannot be true of the formally deficient 
"ratification" of the New Deal structure of governance via the 1936 election 
and the concomitant Revolution of 1937.

I cannot here do justice to Professor Ackerman's elegant and still-growing 
edifice, so I will content myself with some preliminary thoughts. For purposes 
of constitutional interpretation, the creation of the Constitution is the legal 
equivalent of the Big Bang; the Constitution, whatever its normative 
significance may be, is an irreducible fact from which constitutional 
interpretation proceeds. Accordingly, from an interpretative, as opposed to a 
justificatory, standpoint, irregularities in the Constitution's ratification 
validate further irregularities only if the original irregularity reflects a 
background principle that was then incorporated into the Constitution and the 
subsequent irregularity conforms to that principle. Professor Ackerman's 
proposed method of constitutional amendment does not follow the form of the 
background principle employed by the original ratifiers/usurpers. 104 
Furthermore, if Professor Ackerman is correct that the Reconstruction 
Amendments were invalid under formal constitutional rules of ratification, the 
obvious conclusion seems to be that both the Reconstruction Amendments and the 
modern administrative state are unconstitutional.

Professor Ackerman's response is that the formally deficient ratifications of 
the Reconstruction Amendments, which occurred under the regime of the 
Constitution of 1789, "provide us with 'historic [*1252] precedents[]' [for 
such ratifications] which we are no more justified in ignoring than Marbury v. 
Madison." 105 But if precedent is a validating concept under the Constitution, 
why not invoke precedent more straightforwardly? This suggests a fourth option 
for dealing with the modern administrative state: conclude, with Henry 
Monaghan, that because "[p]recedent is, of course, part of our understanding of 
what law is," 106 the administrative state's firm entrenchment through 
precedent constitutes legal validation. I have elsewhere argued, however, that 
the use of horizontal precedent in federal constitutional interpretation is 
itself forbidden by the Constitution. 107

Those who believe in some form of precedent have the fifth option, ingeniously 
advanced in a recent manuscript by Peter McCutchen, 108 of seeking "a form of 
constitutional damage control." 109 According to McCutchen, the administrative 
state is here to stay, and even a very weak theory of precedent ratifies this 
result. 110 But our goal, his theory continues, should be to approximate the 
"first-best" world as nearly as we can from within a state of constitutional 
disequilibrium. As McCutchen puts it:

Where unconstitutional institutions are allowed to stand based on a theory of 
precedent, the Court should allow (or even require) the creation of 
compensating institutions that seek to move back toward the constitutional 
equilibrium. The Court should allow such institutions even where the 
compensating institutions themselves would have been unconstitutional if 
considered standing alone. 111

For example, the legislative veto, standing alone, is plainly unconstitutional 
because it violates the Article I presentment requirement. 112 But the 
legislative veto helps compensate for widespread, unconstitutional delegations 
to agencies. A first-best world would have neither delegations nor legislative 
vetoes, but a world with both delegations [*1253] and legislative vetoes is 
closer to the correct constitutional "baseline" than is a world with only 
delegations. 113

If there is any proper role for precedent in constitutional theory, McCutchen 
is probably right: if an incorrect precedent creates a constitutional 
disequilibrium, it is foolish to proceed as though one were still in an 
equilibrium state. As discussed above, however, I do not believe that there is 
any proper role for horizontal precedent in constitutional theory. 114

There remains a sixth option: acknowledge openly and honestly, as did some of 
the architects of the New Deal, that one cannot have allegiance both to the 
administrative state and to the Constitution. If, however, one then further 
follows the New Deal architects in choosing the administrative state over the 
Constitution, one must also acknowledge that all constitutional discourse is 
thereby rendered problematic. The Constitution was a carefully integrated 
document, which contains no severability clause. It makes no sense to agonize 
over the correct application of, for example, the Appointments Clause, the 
Exceptions Clause, or even the First Amendment when principles as basic to the 
Constitution as enumerated powers and nondelegation are no longer considered 
part of the interpretative order. What is left of the Constitution after 
excision of its structural provisions, however interesting it may be as a 
matter of normative political theory, simply is not the Constitution. One can 
certainly take bits and pieces of the Constitution and incorporate them into a 
new, hypothetical document, but nothing is fostered other than intellectual 
confusion by calling that new document the Constitution. 115

Modern champions of the administrative state, however, seem loathe to abandon 
the sheltering language of constitutionalism. But tactical considerations 
aside, it is not at all clear why this is so. Perhaps instead of assuming that 
the label "unconstitutional" should carry normative weight, the constitutional 
problems of the administrative state can lead us to ask whether it should carry 
any weight -- with judges or anyone else. After all, the moral relevance of the 
Constitution is hardly self-evident. 116

   [*1254] And at that point, the humble lawyer must plead incompetence. 
Questions about the Constitution's normative significance, as with all 
questions about how people ought to behave, are distinctively within the domain 
of moral theory. A legal scholar qua legal scholar can tell us, as a factual 
matter, that one must choose between the Constitution and the administrative 
state. He or she can tell us that the architects of the New Deal chose the 
administrative state and that that choice has been accepted by all institutions 
of government and by the electorate. 117 But only the best of moral 
philosophers can tell us which choice is correct. 118

FOOTNOTES:


n1 I use the word "unconstitutional" to mean "at variance with the 
Constitution's original public meaning." That is not the only way in which the 
word is used in contemporary legal discourse. On the contrary, it is commonly 
used to mean everything from "at variance with the private intentions of the 
Constitution's drafters" to "at variance with decisions of the United States 
Supreme Court" to "at variance with the current platform of the speaker's 
favorite political party." These other usages are wholly unobjectionable as 
long as they are clearly identified and used without equivocation. The usage I 
employ, however, is the only usage that fully ties the words "constitutional" 
and "unconstitutional" to the actual meaning of the written Constitution. A 
defense of this claim would require an extended essay on the philosophy of 
language, but I can offer some preliminary observations: consider a recipe that 
calls for "a dash of salt." If one were reading the recipe as a poem or an 
aspirational tract, one might seek that meaning of "dash" that is aesthetically 
or morally most pleasing. But if one is reading it as a recipe, one wants to 
know what "dash" meant to an informed public at the time the recipe was written 
(assuming that the recipe was written for public consumption rather than for 
the private use of the author). Of course, once the recipe is understood, one 
might conclude that it is a bad recipe, either because it is ambiguous or, more 
fundamentally, because the dish that it yields simply isn't very appealing. But 
deciding whether to try to follow the recipe and determining what the recipe 
prescribes are conceptually distinct enterprises. If the Constitution is best 
viewed as a recipe -- and it certainly looks much more like a recipe than a 
poem or an aspirational tract -- application of the methodology of original 
public meaning is the appropriate way to determine its meaning.

n2 Cf. Cass R. Sunstein, Constitutionalism After the New Deal, 101 HARV. L. 
REV. 421, 447-48 (1987) (noting that the New Deal "altered the constitutional 
system in ways so fundamental as to suggest that something akin to a 
constitutional amendment had taken place").

n3 See 1 BRUCE ACKERMAN, WE THE PEOPLE 44 (1991); Sunstein, supra note 2, at 
430.

n4 See JAMES M. LANDIS, THE ADMINISTRATIVE PROCESS passim (1938).

n5 This perception is evident more from the quantity and tone than from the 
specific content of recent discussions of the structural Constitution, but a 
few scholars have stated the point expressly. See Alfred C. Aman, Jr., 
Introduction, 77 CORNELL L. REV. 421, 427 (1987) (claiming that structural 
issues "of fundamental importance" "are again up for grabs"); Sunstein, supra 
note 2, at 509 (noting that "[t]he last three decades have seen a growing 
rejection of the New Deal conception of administration").

n6 See Geoffrey P. Miller, From Compromise to Confrontation: Separation of 
Powers in the Reagan Era, 57 GEO. WASH. L. REV. 401, 401 (1989).

n7 Modern debates about the scope and structure of the national government tend 
to concern such relatively peripheral matters as the removability of 
administrative officials, see, e.g., Morrison v. Olson, 487 U.S. 654, 682-83 
(1988); Bowsher v. Synar, 478 U.S. 714, 721-34 (1986), or the national 
government's power directly to regulate state governments, see, e.g., New York 
v. United States, 112 S. Ct. 2408, 2419-32 (1992); Garcia v. San Antonio Metro. 
Transit Auth., 469 U.S. 528, 537-47 (1985); National League of Cities v. Usery, 
426 U.S. 833, 852-55 (1976), overruled by Garcia, 469 U.S. at 531 (1985).

n8 Cynthia Farina has aptly described this explicitly non-normative project as 
an exercise in "legal archaeology."

n9 It is possible to maintain that the phrase "the Constitution of the United 
States" refers not to the text of a specific document, but refers instead, in 
the fashion of England's unwritten constitution, to a set of practices and 
traditions that have evolved over time. As a matter of practical governance, 
such unwritten practices are surely more important than the instructions 
contained in the written Constitution, but this Article is concerned solely 
with the written texts that have been submitted to and ratified by the American 
electorate. Cf. Akhil Reed Amar, Our Forgotten Constitution: A Bicentennial 
Comment, 97 YALE L.J. 281, 282-85 (1987) (noting discrepancies between the 
document produced by the constitutional convention and the document ratified by 
the electorate).

n10 THE FEDERALIST No. 45, at 292 (James Madison) (Clinton Rossiter ed., 1961).

n11 Id.

n12 See id. at 292-93. In my favorite passage from The Federalist, Madison 
boldly proclaimed that the federal revenue collectors "will be principally on 
the seacoast, and not very numerous." Id. at 292.

n13 U.S. CONST. art. I, § 1 (emphasis added).

n14 This understanding is expressly confirmed by the Tenth Amendment, which 
declares that "[t]he powers not delegated to the United States by the 
Constitution, nor prohibited by it to the States, are reserved to the States 
respectively, or to the people." Id. amend. X.

n15 See id. art. I, § 8, cls. 1-17.

n16 See id. art. IV, §§ 1, 3.

n17 See id. art. V.

n18 See id. art. I, § 8, cl. 3.

n19 See Richard A. Epstein, The Proper Scope of the Commerce Power, 73 VA. L. 
REV. 1387, 1388-89, 1393-95 (1987).

n20 U.S. CONST. art. I, § 8, cl. 18 (providing that Congress shall have the 
power "[t]o make all Laws which shall be necessary and proper for carrying into 
Execution the foregoing Powers, and all other Powers vested by this 
Constitution in the Government of the United States, or in any Department or 
Officer thereof").

n21 See Epstein, supra note 19, at 1397-98.

n22 See Gary Lawson & Patricia B. Granger, The "Proper" Scope of Federal Power: 
A Jurisdictional Interpretation of the Sweeping Clause, 43 DUKE L.J. 267, 
335-36 (1993).

n23 See id. at 331-32. But see DAVID E. ENGDAHL, CONSTITUTIONAL FEDERALISM IN A 
NUTSHELL 18-19 (2d ed. 1987) (arguing that Congress may regulate unenumerated 
subjects in the course of implementing enumerated powers); David E. Engdahl, 
The Spending Power 9-14 (April 14, 1993) (unpublished manuscript, on file with 
the Harvard Law School Library) (same).

n24 See U.S. CONST. art. I, § 8, cl. 1.

n25 See Engdahl, supra note 23, at 29-32 (demonstrating that the Taxing Clause 
is not a proper source of a federal spending power).

n26 U.S. CONST. art. IV, § 3, cl. 2 (emphasis added).

n27 See Engdahl, supra note 23, at 30-31.

n28 See id. at 37-63.

n29 See U.S. CONST. amend. XIII, § 2.

n30 See id. amend. XV, § 2.

n31 See id. amend. XIV, § 5.

n32 See id. amend. XVI. Other amendments also grant power to Congress. See id. 
amend. XIX, cl. 2 (giving Congress the power to enforce a prohibition on 
gender-based discriminatory voting practices); id. amend. XXIII, § 2 (giving 
Congress the power to enforce the District of Columbia's participation in the 
electoral college); id. amend. XXIV, § 2 (giving Congress the power to enforce 
a prohibition against poll taxes); id. amend. XXVI, § 2 (giving Congress the 
power to enforce a prohibition against denying eighteen-year-old people the 
vote on account of age).

n33 See New York v. United States, 112 S. Ct. 2408, 2428-29 (1992); National 
League of Cities v. Usery, 426 U.S. 833, 852 (1976).

n34 See Garcia v. San Antonio Metro. Transit Auth., 469 U.S. 528, 531 (1985) 
(overruling Usery).

n35 See, e.g., Perez v. United States, 402 U.S. 146, 156-57 (1971) (holding the 
Consumer Credit Protection Act to be within Congress's power to regulate 
interstate commerce); Wickard v. Filburn, 317 U.S. 111, 128-29 (1942) (holding 
regulation of the production of wheat grown for personal consumption to be 
within Congress's power to regulate interstate commerce). The lower federal 
courts have basically followed suit, though there has been a modest 
counterrevolution in the past two years. See Hoffman Homes, Inc. v. 
Administrator, United States EPA, 961 F.2d 1310, 1311 (7th Cir.) (stating that 
the EPA could not regulate, as "wetlands" subject to the Clean Water Act, a 
small depression that occasionally filled with rainwater), vacated, 975 F.2d 
1554 (7th Cir. 1992); United States v. Cortner, 834 F. Supp. 242, 244 (M.D. 
Tenn. 1993) (holding that Congress could not make carjacking a federal criminal 
offense, because the activity "lacks any rational nexus to interstate 
commerce"); cf. United States v. Lopez, 2 F.3d 1342, 1366-68 (5th Cir. 1993) 
(holding that Congress could not, in the absence of explicit legislative 
findings of an effect on interstate commerce, prohibit knowing possession of a 
firearm within one thousand feet of a school).

n36 See Frank H. Easterbrook, Presidential Review, 40 CASE W. RES. L. REV. 905 
(1989-1990).

n37 10 Op. Off. Legal Counsel 40, 40-42 (1986).

n38 U.S. CONST. art. I, § 1.

n39 Id. art. II, § 1.

n40 Id. art. III, § 1.

n41 The President, through the presentment and veto provisions, see id. art. I, 
§ 7, cls. 2-3, is given a sui generis role in the legislative process that 
defies classification along tripartite lines. See Gary Lawson, Territorial 
Governments and the Limits of Formalism, 78 CAL. L. REV. 853, 858 n.19 (1990). 
The Vice President is made an officer of the Senate and is given the power to 
break ties in that body. See U.S. CONST. art. I, § 3, cl. 4. The Senate is 
given the seemingly judicial power to try impeachments. See id. art. I, § 3, 
cl. 6. Certain other powers, such as the power to make treaties and to appoint 
national officers, are shared among the various departments. See id. art. II, § 
2, cl. 2.

n42 A number of state constitutions of the founding era did contain such 
express separation of powers provisions. The most famous example is the 
Massachusetts Constitution of 1780:

In the government of this Commonwealth, the legislative department shall never 
exercise the executive and judicial powers, or either of them: The executive 
shall never exercise the legislative and judicial powers, or either of them; 
The judicial shall never exercise the legislative and executive powers, or 
either of them: To the end it may be a government of laws and not of men.

MASS. CONST. of 1780, pt. I, art. 30; see also VA. CONST. of 1776 P2 ("The 
legislative, executive, and judiciary departments shall be separate and 
distinct, so that neither exercise the powers properly belonging to the 
other.").

n43 I am profoundly indebted to Marty Redish for this important insight.

n44 U.S. CONST. art I, § 8, cl. 18 (emphasis added). The word "proper" in the 
Sweeping Clause provides the textual vehicle for enforcement of the 
Constitution's nondelegation principle. See Lawson & Granger, supra note 22, at 
333-34.

n45 See Thomas W. Merrill, The Constitutional Principle of Separation of 
Powers, 1991 SUP. CT. REV. 225, 256 ("[T]he Constitution makes no effort to 
define the 'legislative,' 'executive,' and 'judicial' powers."). The framers 
harbored no illusions that these powers were self-defining. Madison, for 
example, observed in The Federalist:

Experience has instructed us that no skill in the science of government has yet 
been able to discriminate and define, with sufficient certainty, its three 
great provinces -- the legislative, executive, and judiciary. . . . Questions 
daily occur in the course of practice which prove the obscurity which reigns in 
these subjects, and which puzzle the greatest adepts in political science.

THE FEDERALIST NO. 37, at 228 (James Madison) (Clinton Rossiter ed., 1961). The 
problem of distinguishing the three functions of government has long been, and 
continues to be, one of the most intractable puzzles in constitutional law. See 
Wayman v. Southard, 23 U.S. (10 Wheat.) 1, 46 (1825) ("[T]he maker of the law 
may commit something to the discretion of the other departments, and the 
precise boundary of this power is a subject of delicate and difficult inquiry. 
. . ."). See generally William B. Gwyn, The Indeterminacy of the Separation of 
Powers and the Federal Courts, 57 GEO. WASH. L. REV. 474, 503 (1989).

n46 See Paul M. Bator, The Constitution as Architecture: Legislative and 
Administrative Courts Under Article III, 65 IND. L.J. 233, 264 (1990).

n47 Circularity of this kind is neither fatal nor unprecedented. For example, 
under relevant (and correct) case law, a federal employee is an officer subject 
to the Appointments Clause, U.S. CONST. art. II, § 2, cl. 2, if he or she is 
sufficiently important to be subject to the Appointments Clause. See Lawson, 
supra note 41, at 865 n.63.

n48 For a description and analysis of the case law on nondelegation, see 
Cynthia R. Farina, Statutory Interpretation and the Balance of Power in the 
Administrative State, 89 COLUM. L. REV. 452, 478-88 (1989).

n49 Marty Redish has independently formulated a very similar principle for 
distinguishing the legislative and executive powers, which he calls the 
"political commitment principle." See MARTIN H. REDISH, THE CONSTITUTION AS 
POLITICAL STRUCTURE (forthcoming 1994) (manuscript ch. 5, at 2-4, on file with 
author). This principle requires of valid legislation "some meaningful level of 
normative political commitment by the enacting legislators, thus enabling the 
electorate to judge its representatives." Id. ch. 5, at 4; see also David 
Schoenbrod, The Delegation Doctrine: Could the Court Give It Substance?, 83 
MICH. L. REV. 1223, 1252-58 (1985) (distinguishing between statutes that 
prescribe rules of conduct and invalid statutes that merely state legislative 
goals).

n50 See Schechter Poultry Corp. v. United States, 295 U.S. 495, 541-42 (1935); 
Panama Refining Co. v. Ryan, 293 U.S. 388, 430 (1935). The Court does 
occasionally invoke delegation concerns in the course of statutory 
interpretation. See, e.g., Industrial Union Dep't, AFL-CIO v. American 
Petroleum Inst. (Benzene), 448 U.S. 607, 646 (1980) (plurality opinion) 
(holding that an OSHA statute, if interpreted broadly, would be a sweeping and 
unconstitutional delegation of power).

n51 See 15 U.S.C. § 78j(b) (1988) (proscribing the use or employment, "in 
connection with the purchase or sale of any security . . ., [of] any 
manipulative or deceptive device or contrivance in contravention of such rules 
and regulations as the [Securities and Exchange] Commission may prescribe as 
necessary or appropriate in the public interest or for the protection of 
investors").

n52 See 47 U.S.C. § 307(a) (1988) (prescribing that the Federal Communications 
Commission shall grant broadcast licenses to applicants "if public convenience, 
interest, or necessity will be served thereby").

n53 See 21 U.S.C. § 41 (1988) (forbidding the importation of "any merchandise 
as tea which is inferior in purity, quality, and fitness for consumption to the 
standards" set by the Secretary of Health and Human Services).

n54 See, e.g., Touby v. United States, 111 S. Ct. 1752, 1756 (1991); Skinner v. 
Mid-America Pipeline Co., 490 U.S. 212, 218-24 (1989); Mistretta v. United 
States, 488 U.S. 361, 378-79 (1989).

n55 The problem with the Goodness and Niceness Commission under current law (if 
indeed there is a problem) would be that it had been delegated too much of 
Congress's power in one fell swoop. Modern law, in other words, will permit 
Congress to create a set of miniature Goodness and Niceness Commissions, no one 
of which has authority over all aspects of life, but would likely balk at a 
single agency exercising unconstrained legislative authority over too broad a 
range of subjects. See Industrial Union Dep't, AFL-CIO v. American Petroleum 
Inst. (Benzene), 448 U.S. 607, 646 (1980) (plurality opinion) (narrowly 
construing the Occupational Safety and Health Act because a broad construction 
would give the Secretary of Labor "unprecedented power over American industry" 
and would thus constitute "such a 'sweeping delegation of legislative power' 
that it might be unconstitutional") (quoting Schechter Poultry Corp. v. United 
States, 295 U.S. 495, 539 (1935)).

n56 488 U.S. 361 (1989).

n57 Id. at 372.

n58 David Schoenbrod has documented that President Reagan never vetoed a bill 
on nondelegation grounds nor did his Justice Department ever oppose such 
legislation. See David Schoenbrod, How the Reagan Administration Trivialized 
Separation of Powers (and Shot Itself in the Foot), 57 GEO. WASH. L. REV. 459, 
464-65 (1989). I have confirmed that this same fact is true of the Bush 
administration through my own recollections and those of several Bush 
administration officials and by consulting published opinions of the Office of 
Legal Counsel.

n59 U.S. CONST. art. II, § 1, cl. 1.

n60 See Steven G. Calabresi & Kevin H. Rhodes, The Structural Constitution: 
Unitary Executive, Plural Judiciary, 105 HARV. L. REV. 1153, 1177 n.119 (1992).

n61 U.S. CONST. art. II, § 3, cl. 3.

n62 See Calabresi & Rhodes, supra note 60, at 1198 n.221.

n63 See Steven G. Calabresi, The Vesting Clauses As Power Grants, 88 NW. U. L. 
REV. (forthcoming 1994) (manuscript at 2, on file with the Harvard Law School 
Library).

n64 U.S. CONST. art. I, § 1, cl. 1.

n65 The qualifier "discretionary" is important. If a statute requires a 
ministerial act, such that a writ of mandamus would properly lie to compel its 
performance, it does not matter in whom the statute vests power. See Kendall v. 
United States, 37 U.S. (12 Pet.) 524, 610-13 (1838).

n66 See, e.g., Thomas O. McGarity, Presidential Control of Regulatory Agency 
Decisionmaking, 36 AM. U. L. REV. 443, 465-72 (1987) (arguing that Congress 
"may provide that the President may not substitute his judgment . . . for that 
of the official to whom Congress has delegated decisionmaking power"); cf. 
Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 
COLUM. L. REV. 1, 55 (1994) (claiming that, under an originalist interpretation 
of the Constitution, "Congress has wide discretion to vest . . . 
[administrative powers] in officers operating under or beyond the plenary power 
of the President").

n67 Several legal scholars have compiled impressive lists of historical 
materials suggesting that many early legal actors and writers did not 
contemplate any wide-ranging presidential power of supervision. See Lessig & 
Sunstein, supra note 66, at 15-17; Morton Rosenberg, Presidential Control of 
Agency Rulemaking: An Analysis of Constitutional Issues That May Be Raised by 
Executive Order 12,291, 23 ARIZ. L. REV. 1199, 1205-10 (1981).

n68 See Lessig & Sunstein, supra note 66, at 46-52; McGarity, supra note 66, at 
466; Rosenberg, supra note 67, at 1209.

n69 Steve Calabresi has recently formulated and marshalled these arguments. See 
Calabresi, supra note 63, at 4-22; Steven G. Calabresi, The Trinity of Powers 
and the Lessig/Sunstein Heresy passim (March 6, 1994) (unpublished manuscript, 
on file with the Harvard Law School Library). I can here summarize only a few 
of Professor Calabresi's arguments. First, the Sweeping Clause gives Congress 
power to carry into execution "all other Powers vested by this Constitution in 
the Government of the United States, or in any Department or Officer thereof." 
U.S. CONST. art. I, § 8, cl. 18 (emphasis added). In view of this language, it 
is very hard to argue that the Article II Vesting Clause does not vest powers. 
Second, a close textual and structural comparison of Articles II and III 
demonstrates that the Vesting Clauses in each Article serve the same function. 
Inasmuch as the Article III Vesting Clause must be read as a grant of power to 
courts to decide cases rather than as merely a designation of office, the 
Article II Vesting Clause must also be a grant of power. Third, the Article II 
Vesting Clause is the only plausible source of a constitutional power to 
execute the laws. The only other conceivable source of such a power -- the Take 
Care Clause, U.S. CONST. art II, § 2, cl. 3 (declaring that the President 
"shall take Care that the Laws be faithfully executed") -- is worded as a duty 
of faithful execution rather than as a grant of power.

n70 See Calabresi & Rhodes, supra note 60, at 1166.

n71 See id.; Lee S. Liberman, Morrison v. Olson: A Formalist Perspective on Why 
the Court Was Wrong, 38 AM. U. L. REV. 313, 353-54 (1989).

n72 The executive power, unlike the legislative and judicial powers, has always 
been understood to be delegable by the President. See Mistretta v. United 
States, 488 U.S. 361, 424-25 (1989) (Scalia, J., dissenting); 2 ANNALS OF CONG. 
712 (1792) ("[I]t is of the nature of Executive power to be transferrable to 
subordinate officers; but Legislative authority is incommunicable, and cannot 
be transferred.") (statement of Representative Findley). Accordingly, if the 
President can directly exercise all powers vested by statute in executive 
officials, the President can presumably designate any subordinate official to 
exercise that power. Thus, if a statute vests authority to promulgate standards 
for workplace safety in the Secretary of Labor, the President could, on this 
understanding, personally assume that power and then delegate it to the 
Secretary of Defense. Perhaps this is the correct view of the President's 
power, but it seems more plausible to suppose that Congress can at least 
determine which subordinate officials, if any, are permitted to exercise 
delegated executive powers. See Geoffrey P. Miller, The Unitary Executive in a 
Unified Theory of Constitutional Law: The Problem of Interpretation, 15 CARDOZO 
L. REV. 201, 205 (1993). On this supposition, if a statute vests power to 
promulgate workplace standards in the Secretary of Labor, the President cannot 
personally promulgate safety standards nor designate anyone other than the 
Secretary of Labor to perform that task, although the President can issue 
instructions -- including instructions so detailed that they take the form of 
regulations -- with which the Secretary of Labor must comply if he or she is to 
act at all.

n73 See, e.g., Morrison v. Olson, 487 U.S. 654, 696-97 (1988) (upholding the 
constitutionality of the independent counsel provisions of the Ethics in 
Government Act); Bowsher v. Synar, 478 U.S. 714, 721-27 (1986) (striking down a 
provision of the Gramm-Rudman-Hollings Act that gave the Comptroller General a 
role in the appropriations process).

n74 It is therefore constitutionally nonexistent as well. The only mode of 
removal specifically mentioned in the Constitution is impeachment. See U.S. 
CONST. art. II, § 4. Accordingly, one could reasonably believe: that 
impeachment is the only permissible form of removal, that Congress's power to 
create offices carries with it the power to prescribe the form of removal, or 
that the power of removal follows the power of appointment, so that if the 
Senate must consent to an officer's appointment, it must also consent to that 
officer's removal. See Lawson, supra note 41, at 883 n.172. One can infer a 
presidential removal power only by assuming that such a power is necessary in 
order to ensure a unitary executive. See Myers v. United States, 272 U.S. 52, 
132-35 (1926) (making such an inference). Inasmuch as even the strongest 
removal power does not ensure compliance with the Article II Vesting Clause, 
any such inference of a constitutionally based presidential removal power seems 
hard to justify.

n75 See 1 ANNALS OF CONG. 384-412, 473-608, 614-31, 635-39 (1789). Should this 
fact give pause to advocates of the unitary executive? Probably, although the 
framers' silence is not decisive in the face of compelling textual and 
structural arguments for presidential control of execution. In order to 
establish that something is the original meaning of a constitutional provision, 
one needs to show that the general public would have acknowledged that meaning 
as correct if all relevant arguments and information had been brought to its 
attention. Actual instances of usage (or non-usage) are therefore probative but 
not dispositive.

n76 See, e.g., 18 Op. Att'y Gen. 31, 32 (1884); 11 Op. Att'y Gen. 14, 15-19 
(1864); 1 Op. Att'y Gen. 624, 625-29 (1823). But see 6 Op. Att'y Gen. 326, 
339-46 (1854) (disavowing the reasoning and conclusions of these opinions).

n77 See Merrill, supra note 45, at 253-54. Nor is it obvious that courts would 
have validated limitless delegations directly to the President rather than to 
"expert, non-political" agencies.

n78 See 15 Op. Off. Legal Counsel 8, 16-17 (1991) (construing a statute to 
permit the Secretary of Education to review decisions of administrative law 
judges on the ground, inter alia, that foreclosure of review would be 
unconstitutional); 13 Op. Off. Legal Counsel 299, 306-07 (1989) (objecting 
generally to concurrent reporting requirements that allow agencies to transmit 
budget requests or legislative proposals to Congress without presidential 
review); 12 Op. Off. Legal Counsel 58, 60-71 (1988) (asserting the 
unconstitutionality of a congressional resolution requiring the Centers for 
Disease Control to mail AIDS information free from executive supervision).

n79 U.S. CONST. art. III, § 1.

n80 See id.

n81 Joshua I. Schwartz, Nonacquiescence, Crowell v. Benson, and Administrative 
Adjudication, 77 GEO. L.J. 1815, 1835 (1989).

n82 See Freytag v. Commissioner, 111 S. Ct. 2631, 2654-56 (1991) (Scalia, J., 
concurring in part and concurring in the judgment); Murray's Lessee v. Hoboken 
Land & Improvement Co., 59 U.S. (18 How.) 272, 280 (1855); Frank H. 
Easterbrook, "Success" and the Judicial Power, 65 IND. L.J. 277, 280-81 (1990).

n83 See Freytag, 111 S. Ct. at 2655; Murray's Lessee, 59 U.S. at 284.

n84 See REDISH, supra note 49, ch. 5, at 9-11 (discussing Locke and 
Montesquieu).

n85 59 U.S. (18 How.) 272 (1855).

n86 U.S. CONST. amend. V.

n87 Legislation that does not require executive and judicial adherence to 
principles of due process is not "proper" under the Sweeping Clause and thus 
would have been unconstitutional even before ratification of the Fifth 
Amendment in 1791. See Lawson & Granger, supra note 22, at 329-30.

n88 Richard H. Fallon, Jr., Of Legislative Courts, Administrative Agencies, and 
Article III, 101 HARV. L. REV. 916, 918 (1988).

n89 Martin H. Redish, Legislative Courts, Administrative Agencies, and the 
Northern Pipeline Decision, 1983 DUKE L.J. 197, 227.

n90 See Chevron U.S.A. Inc. v. National Resources Defense Counsel, Inc., 467 
U.S. 837, 842-43 (1984).

n91 See, e.g., 5 U.S.C. § 706(2)(E) (1988) (specifying the substantial evidence 
test); 29 U.S.C. § 660(a) (1988) (same).

n92 See Calabresi & Rhodes, supra note 60, at 1155-56.

n93 THE FEDERALIST NO. 47, at 301 (James Madison) (Clinton Rossiter ed., 1961).

n94 See Sunstein, supra note 2, at 446-47.

n95 See OFFICE OF THE FED. REGISTER, NATIONAL ARCHIVES & RECORD ADMIN., UNITED 
STATES GOVERNMENT MANUAL 1993/1994, at 612-17 (1993).

n96 See FTC v. Standard Oil Co., 449 U.S. 232, 245 (1980) (refusing to permit 
judicial review of the filing of an administrative complaint on the ground that 
such agency action is nonfinal).

n97 See Rebecca L. Brown, Separated Powers and Ordered Liberty, 139 U. PA. L. 
REV. 1513, 1556 (1991).

n98 See supra p. 1234.

n99 Although, as Lawrence Lessig and Cass Sunstein point out, the Sweeping 
Clause gives Congress substantial power to control the manner in which the 
executive department executes the laws, see Lessig & Sunstein, supra note 66, 
at 66-69, that power is limited by the Sweeping Clause's terms. Congress is 
permitted to create a particular governmental structure if, but only if, other 
constitutional provisions or background understandings establish that such a 
structure conforms to a "proper" conception of separation of powers. See Lawson 
& Granger, supra note 22, at 333-34; see also Lessig & Sunstein, supra note 66, 
at 67 n.278, 69 (noting that there are constitutional limits on Congress's 
power under the Sweeping Clause). Thus, the scope of Congress's power to 
structure the national government depends largely on the extent to which the 
Vesting Clauses of Articles II and III do or do not grant power to the 
President and the federal courts, respectively -- and thus do not or do leave 
governmental powers unallocated by the constitutional text. Accordingly, 
Professors Lessig and Sunstein's conclusion that "the framers wanted to 
constitutionalize just some of the array of power a constitution-maker must 
allocate, and as for the rest, the framers intended Congress (and posterity) to 
control as it saw fit," id. at 41, ultimately rests, as a textual matter, on 
their argument that the Article II and Article III Vesting Clauses are not 
grants of power, see id. at 46-52 -- an argument that is very difficult to 
sustain either textually or structurally. See supra note 69.

(Patty Granger and I are grateful to Professors Lessig and Sunstein for their 
generous use of our article on the Sweeping Clause in their recent work on the 
presidency. See Lessig & Sunstein, supra note 66, at 41 n.178, 67 n.278. At the 
risk of appearing to quibble in the name of clarification, however: Professors 
Lessig and Sunstein cite our article, under a "see also" signal, in support of 
the conclusion that the framers left the allocation of some important 
governmental powers to "Congress (and posterity) to control as it saw fit." See 
id. at 41 n.178. Our article neither directly supports nor directly rebuts such 
a claim of congressional power. It demonstrates that Congress can structure the 
government only through laws that are objectively necessary and proper, see 
Lawson & Granger, supra note 22, at 276, but whether a particular governmental 
structure is "proper" depends on constitutional norms external to the Sweeping 
Clause. Thus, as noted above, the soundness of Professors Lessig and Sunstein's 
conclusion concerning congressional power depends largely on the soundness of 
their interpretation of the Article II and Article III Vesting Clauses. The 
phrase "necessary and proper" in the Sweeping Clause is a neutral player in 
that dispute -- although the Sweeping Clause's use of the phrase "powers 
vested" supports a power-granting construction of the Vesting Clauses. See 
supra note 69. By way of further clarification: Professors Lessig and Sunstein 
cite -- and endorse -- our conclusion that the word "proper" in the Sweeping 
Clause constrains Congress's power, but with the proviso that they "do not 
agree that the clause is a limitation on Congress's power (rather than a grant 
of power)." Lessig & Sunstein, supra note 66, at 67 n.278. In fact, on this 
point (as on many others), there is no disagreement among us. Ms. Granger and I 
emphatically maintain that the Sweeping Clause is a grant of power to Congress, 
see, e.g., Lawson & Granger, supra note 22, at 270, 276, 328, but insist that 
it is a grant of limited rather than unlimited power.)

n100 See supra note 1.

n101 I suspect that this claim is controversial only because of a failure to 
distinguish between theories of interpretation and theories of adjudication. 
Imagine, for example, that a second American revolution openly discards the 
Constitution, so that there is no chance that any conclusions about the 
Constitution's meaning could have any significant effects on the real world. In 
the absence of any plausible concern about the practical consequences of 
constitutional interpretation (and putting aside for the moment the 
interpretative significance of precedent), it seems inconceivable that one 
would even think to apply anything other than originalist interpretivism when 
interpreting the Constitution -- just as no one would today think of 
interpreting the Articles of Confederation by any other method. In other words, 
I suspect that originalist interpretivism is controversial only because its 
descriptive interpretative conclusions are widely thought to have prescriptive 
adjudicative consequences.

n102 See ACKERMAN, supra note 3, at 34-57; Bruce Ackerman, Constitutional 
Politics/Constitutional Law, 99 YALE L.J. 453, 510-15 (1989) [hereinafter 
Ackerman, Constitutional Politics]; Bruce Ackerman, The Storrs Lectures: 
Discovering the Constitution, 93 YALE L.J. 1013, 1051-57 (1984).

n103 The Constitution was ratified in a manner inconsistent both with the 
amendment process specified in the Articles of Confederation and with the 
ratification procedures of a number of state constitutions. The ratification of 
the Reconstruction Amendments involved something very close to vote fraud. See 
Ackerman, Constitutional Politics, supra note 102, at 500-07.

n104 Akhil Amar has argued that the ratification of the original Constitution 
was valid because it was consistent with an accepted background norm for the 
ratification of fundamental law: ratification by direct majority vote of "We 
the People." That norm, he argues, is carried forward in the existing 
Constitution as an unenumerated right of the people, so an amendment ratified 
by direct majority vote would be constitutionally valid. See Akhil Reed Amar, 
Philadelphia Revisited: Amending the Constitution Outside Article V, 55 U. CHI. 
L. REV. 1043, 1044 (1988). This analysis, however, cannot save the 
administrative state, because no such amendment has ever been so ratified.

n105 Ackerman, Constitutional Politics, supra note 102, at 508 (paraphrasing 
Coleman v. Miller, 307 U.S. 433, 449 (1939)).

n106 Henry P. Monaghan, Stare Decisis and Constitutional Adjudication, 88 
COLUM. L. REV. 723, 748 (1988) (emphasis added).

n107 See Gary Lawson, The Constitutional Case Against Precedent, 17 HARV. J.L. 
& PUB. POL'Y 23, 25-33 (1994). It is perhaps a bit arrogant to toss off a 
proposition of this magnitude so casually, but the prima facie case against 
precedent is undeceptively simple: if interpreters have the power and duty to 
prefer the Constitution to statutes or executive acts because the Constitution 
is supreme law, they a fortiori have the same power and duty to prefer the 
Constitution to prior judicial decisions.

n108 Peter B. McCutchen, Mistakes, Precedent, and the Rise of the 
Administrative State: Toward a Constitutional Theory of the Second Best 
(October 20, 1993) (unpublished manuscript, on file with the Harvard Law School 
Library).

n109 Id. at 3.

n110 See id. at 26-32.

n111 Id. at 3, 4.

n112 U.S. CONST. art. I, § 7, cls. 2-3.

n113 See McCutchen, supra note 108, at 62-65.

n114 See supra note 107.

n115 See Suzanna Sherry, An Originalist Understanding of Minimalism, 88 NW. U. 
L. REV. 175, 182 (1993). Of course, there may be tactical reasons for casting 
normative political arguments in the (often unaccommodating) language of 
constitutionalism. If official actors or the public believe, or act as though 
they believe, that the Constitution matters, effective rhetorical strategy 
requires that one couch arguments in constitutional language -- and perhaps 
even that one lie about one's goals and methods. But truth-seekers have no 
interest in such rhetorical games.

n116 See Gary S. Lawson, An Interpretivist Agenda, 15 HARV. J.L. & PUB. POL'Y 
157, 160-61 (1992); Larry Simon, The Authority of the Constitution and Its 
Meaning: A Preface to a Theory of Constitutional Interpretation, 58 S. CAL. L. 
REV. 603, 606-07, 613-19 (1985).

n117 Political candidates seeking office typically do not call for abolishing 
administrative government in the name of the Constitution, which suggests that 
such a platform probably would not garner a large percentage of the popular 
vote.

n118 See Gary S. Lawson, The Ethics of Insider Trading, 11 HARV. J.L. & PUB. 
POL'Y 727, 778 (1988) ("It is conceivable that the ethical, epistemological, 
and metaphysical problems of the ages will be solved by an article in a 
twentieth-century, English-language law journal. But I rather doubt it.").