+(,121/,1(
Citation: 90 Tex. L. Rev. 943 2011-2012
Content downloaded/printed from
HeinOnline (http://heinonline.org)
Tue Jul 16 11:37:58 2013
-- Your use of this HeinOnline PDF indicates your acceptance
of HeinOnline's Terms and Conditions of the license
agreement available at http://heinonline.org/HOL/License
-- The search text of this PDF is generated from
uncorrected OCR text.
-- To obtain permission to use this article beyond the scope
of your HeinOnline license, please use:
https://www.copyright.com/ccc/basicSearch.do?
&operation=go&searchType=0
&lastSearch=simple&all=on&titleOrStdNo=0040-4411
Book Reviews
The Madisonian Constitution: Rightly Understood
THE EXECUTIVE UNBOUND: AFTER THE MADISONIAN REPUBLIC. By Eric A.
Posner & Adrian Vermeule. New York, New York: Oxford University
Press, 2010. 249 pages. $29.95.
Reviewed by Benjamin Kleinerman*
Eric Posner and Adrian Vermeule begin The Executive Unbound: After
the Madisonian Republic by announcing the failure of what they call liberal
legalism.' They write: "[T]he simplest version of liberal legal theory holds
that representative legislatures govern and should govern, subject to constitutional constraints, while executive and judicial officials carry out the law." 2
Associating legal liberalism with James Madison, they draw on the thought
of the Weimar legal theorist Carl Schmitt in order to conclude, like Schmitt
had about seventy years ago, that the Madisonian constitutional republic has
been eclipsed by the political reality of the unconstrained Executive.3 Of
course, unlike Schmitt, they do not suggest that the Executive is utterly
Instead, they claim: "[T]he major constraints on the
unconstrained.4
executive, especially in crises, do not arise from law or from the separationof-powers framework defended by liberal legalists, but from politics and
public opinion."5 Legal liberalism, they argue, goes wrong when it equates
"a constrained executive with an executive constrained by law."6
One of the first major reviews of this book rightly noted that one of the
crucial mistakes made in this book comes in its simplistic equation of liberal
legalism with the thought of James Madison.7 It is almost too easy to show
* Garwood Visiting Fellow, James Madison Program in American Ideals and Institutions,
Princeton University; Assistant Professor of Constitutional Democracy, James Madison College,
Michigan State University.
1. ERIC A. POSNER & ADRIAN VERMEULE, THE EXECUTIVE UNBOUND: AFTER THE
MADISONIAN REPUBLIC 3-4 (2010).
2. Id. at 3.
3. See id. at 4 (positioning their book as drawing heavily on Carl Schmitt's critiques of liberal
legalism over more authoritarian forms of government); id. at 7-10 (proposing that liberal legalism
has been unable to generate meaningful constraints due to the "twin problems" of delegation and
emergencies that undermine legalistic constraints on the Executive).
4. Id at 4.
5. Id
6. Id at 5.
7. See Harvey Mansfield, Is the Imperial Presidency Inevitable?, N.Y. TIMES, Mar. 13,
2011, available at http://www.nytimes.com/2011/03/13/books/review/book-review-the-executive-
944
Texas Law Review
[Vol. 90:943
that Madison was never, not even in his later post-Federalist affiliation with
Thomas Jefferson's Republican Party, a liberal legalist of the sort described
by Posner and Vermeule. Part of my aim in this Review will also be to
demonstrate this same point. I will do this not so much, however, so as to
show the obvious-that Posner and Vermeule simply get Madison wrong on
this question-as to show that getting Madison right would have provided a
far more subtle and interesting portrait of our contemporary situation.
Similarly, it is ultimately insufficient to let the authors off the hook by
suggesting that they were simply using Madison as a stand-in for the liberal
legalism that tends to dominate the contemporary legal academy. It is surely
true that the type of liberal legalism that Posner and Vermeule describe can
be found in the legal academy.8 And it is also true that some of these same
scholars sometimes do look to Madison as their guiding light.9 But, in
claiming to dethrone James Madison, Posner and Vermeule have bigger
ambitions than just to criticize the legal academy and its inspiration. They
choose James Madison as their target because their argument is ultimately
aimed at the constitutional order itself. They want to criticize not just the
reigning manifestation of legal liberalism in the academy but the
fundamental assumptions of the constitutional order; as the "father of the
Constitution" that they aim to bring down, or at least make seem
fundamentally irrelevant, James Madison is their natural target.
If their real target is the constitutional order itself, however, one might
ask why they do not spend more time on the actual constitutional thought of
its principal founder, James Madison. I suspect the answer to this question
lies in their own cramped view of the meaning of a constitutional order-a
view that derives entirely from the very same liberal legalism that they
attempt to overcome. Most revealing in this context is an early footnote
discussing Nomi Lazar's book, States of Emergency in Liberal Democracy.o
Her work represents a number of recent books, all of which use the recent
questions about the range of executive power in a constitutional democracy
to show that the liberal constitutional tradition has a much more robust
accommodation for discretionary executive power than had become the
(remarking critically
unbound-by-eric-a-posner-and-adrian-vermeule.html
Vermeule's use of Madison as providing the basis for liberal legalism).
8. See, e.g., DAVID DYZENHAUS,
on
Posner
and
THE CONSTITUTION OF LAW: LEGALITY IN A TIME OF
EMERGENCY 2-5 (2006) (proposing that a response to emergencies should be governed by the rule
of law by introducing a "rule-of-law project" that sets out to sketch a relationship between the
Legislature, Executive, and Judiciary in which all of the powers are involved).
9. See, e.g., PETER M. SHANE, MADISON'S NIGHTMARE: How EXECUTIVE POWER THREATENS
AMERICAN DEMOCRACY 2-3 (2009) (hypothesizing that the recent concentration of power in the
hands of the federal Executive has battered the checks and balances of American government and
would have been Madison's worst constitutional nightmare had he ever envisioned such an
"ambitious and overreaching presidency").
10. POSNER & VERMEULE, supra note 1, at 213 n.6 (citing NOMI CLAIRE LAZAR, STATES OF
EMERGENCY IN LIBERAL DEMOCRACIES (2009)).
2012]
The Madisonian Constitution: Rightly Understood
945
conventional wisdom." In other words, taken seriously, her work should
have raised a profound challenge to Posner and Vermeule's claims about the
irreconcilability of a wide-ranging, discretionary executive with a constitutional order. After all, the upshot of Lazar's work, in addition to a number of
other recent books and articles, shows that Carl Schmitt has a much too
cramped legalistic understanding of a constitutional order. 12 In other words,
Schmitt refutes not the more robust constitutional visions represented by
John Locke and the American founders, but the much more limited, neoKantian vision of Hans Kelsen.13 To continue to take their bearings from
Schmitt, Posner and Vermeule would need to show why Schmitt's critique
remains relevant despite these criticisms. But rather than addressing this
scholarship, Posner and Vermeule mostly ignore it. In the previously mentioned footnote about Lazar, however, they make their one stab at addressing
it. I quote the passage at length because it is so revealing:
Lazar argues that legal constraints on the executive are merely of
instrumental value and that informal constraints can substitute for
them, and to that extent we agree. Lazar, however, remains
committed to liberalism, and in this respect we think she flinches from
the logic of her own conclusions. Once formal constraints on the
executive are abandoned and everything is remitted to politics, there is
11. See, e.g., JEREMY D. BAILEY, THOMAS JEFFERSON AND EXECUTIVE POWER 9-10 (2007)
(examining Thomas Jefferson's understanding of executive power as a strong and energetic power
that consists of the ability to act outside the law "on behalf of the public good" while also
maintaining limitations on the power by enabling the people to judge executive discretion); ROSS J.
CORBETT, THE LOCKEAN COMMONWEALTH 5-6 (2009) (arguing that Locke speaks to the problem
of unlawful executive action in his Two Treatises of Government and acknowledging that the
competing liberal principles of demanding that the government not proceed unlawfully while also
demanding that a government act for the good of the people in whatever mode necessary lead to an
inherent, but necessary, tension within constitutional liberalism); CLEMENT FATOVIC, OUTSIDE THE
LAW: EMERGENCY AND EXECUTIVE POWER 253-54 (2009) (concluding that "strict adherence to the
formal requirements of law was never considered [by liberal constitutionalists as] sufficient" to
fulfill the "principles underlying the rule of law" and that liberal constitutionalism "permits the
executive to enforce the spirit rather than the letter of the law in extraordinary circumstances");
BENJAMIN A. KLEINERMAN, THE DISCRETIONARY PRESIDENT: THE PROMISE AND PERIL OF
EXECUTIVE POWER 218 (2009) (arguing that the nature and extent of executive discretionary power
enables Congress and the Supreme Court to behave differently from how they would act if they
believed their decisions regarding executive power to be final).
12. See, e.g., LAZAR, supra note 10, at 52-53 (advocating an understanding of institutions as
flexible means to principled ends so as to provide for a liberal democracy that can confront the
problem of emergency without making exceptions to liberal democratic values); Douglas Casson,
Emergency Judgment: CarlSchmitt, John Locke, and the Paradox ofPrerogative,36 POL. & POL'Y
944, 947 (2008) (criticizing recent scholarly interest in the works of Schmitt because of the narrow
choice that Schmitt presents between embracing constitutional rationalism or Schmittian
decisionism).
13. See LAZAR, supra note 10, at 65-66, 83 (characterizing Schmitt's criticism of liberalism as
targeting the neo-Kantian liberalism shaped by Kelsen rather than the pragmatic post-Lockean
liberalism adopted by American founders like Hamilton and Jefferson).
946
Texas Law Review
[Vol. 90:943
little that is distinctively "liberal" about the resulting arrangements; it
is not obvious what is gained by defining liberalism so capaciously. 14
This passage, I would suggest, reveals that in their very rejection of
liberal legalism they remain oddly attached to a confined understanding of
the rule of law. They think Lazar "flinches from the logic of her own
conclusions" because, unlike them, her theory of a wide-ranging executive
constrained both informally and by formal constraints of only instrumental
value does not cause her to give up entirely on liberal constitutionalism. But
why should it? Why does liberal constitutionalism demand only what Posner
and Vermeule characterize as liberal legalism? In their characterization, the
law must approach something like automatic enforcement for it to be meaningful as a constraint. And for the constitutional order itself to be
meaningful, the Executive must be almost subservient to the robust "rule of
law" created by Congress. Because she rejects their conclusions, Posner and
Vermeule accuse Lazar of failing to follow her own logic. But the true failure lies in their inability to see a certain fundamental assumption underlying
their own approach. Their rejection of the power of both the law and the
constitutional order ultimately arises from their heroic assumptions about the
capability of the law and the constitutional order. It is as if they posited
Captain America as the only real human being and, failing to find him, gave
up on the possibility of real human beings. In a certain sense, Posner and
Vermeule exemplify an endemic feature of the school of legal realism from
which they emerge. Legal realism attempts to be hard-headed or toughminded in its claim that the law matters much less than politics.15 But its
rejection of the law is ultimately rooted in heroic assumptions about what the
law must be in order for it to be meaningful. Thus, its supposed realism was
rooted in certain unrealistic assumptions about the character of law. By
contrast, James Madison, their supposed liberal-legalism antagonist, was
actually much more realistic about the relation between politics and law.
Contrary to their portrait of him as committed to the law's almost automatic
enforcement, 16 Madison was profoundly aware of the problem that the law
often ends up being nothing except, in his words, "parchment barriers." 7 In
fact, so far from having an idealistic belief in law's power on its own,
properly understood, the whole Madisonian constitutional system is
14. POSNER & VERMEULE, supra note 1, at 213 n.6.
15. See id at 113-14 ("[Tlhe system of elections, the party system, and American political
culture constrain the executive far more than do legal rules created by Congress or the courts."); id.
at 62 (arguing that the "main process" of constitutional change is "the 'constitutional showdown'episodes of conflict between and among institutions over ... policymaking authority" and that
"[s]uch conflicts are settled by bargaining, brinksmanship, bluffing, and ultimately by the force of
public opinion, rather than by any distinctively legal form of decision making").
16. See id at 41 (arguing that according to the Madisonian view, "even in crisis situations the
executive may act only on the basis of clear congressional authorization that follows public
deliberation, and the executive's actions must presumptively be subject to judicial review").
17. THE FEDERALIST No. 48, at 308 (James Madison) (Clinton Rossiter ed., 1961).
2012]
The Madisonian Constitution: Rightly Understood
947
predicated on a realistic assessment of the relationship between law and
politics. And while Posner and Vermeule are correct to suggest that the
problems in the system lie at the feet of Congress,18 they diagnose the root of
those problems incorrectly. The problems derive not from a misjudgment
about the self-applicability of the law but from an overestimation of the
nature of congressional ambition. Or, it might be more precise to say that
members of Congress fail to exhibit the constitutional ambition that the
constitutional system expects from all of its political actors.
I.
Liberal Legalism
Throughout their book, Posner and Vermeule target what they call
"liberal legalism." According to the authors, one of the crucial failings of
liberal legalism is its expectation that the law will govern according to
clearly articulated standards that give the government unmistakable guidance
and give the people easy signs of the reach of the government's power.' 9 In
contrast to this idealized portrait of the rule of law, the authors suggest that
the modern administrative state places far more discretionary power in the
hands of agencies than the conventional picture should allow.20 Rather than
providing clear guidance to administrative agencies through a clearly articulated statutory framework, Congress passes legislation that mostly delegates
the decision about the nature of the framework to the agencies themselves.
Of course, articulating this fact would not be surprising to most of those who
might identify themselves as "liberal legalists" within the legal academyeven if Posner and Vermeule sometimes speak as though it would be. The
very forcefulness of their claims about what the law should look like stems
from the fact that they recognize that the law so often does not look as it
should. So, the real aim of Posner and Vermeule's argument is not to show
liberal legalists the true situation on the ground but to show them that the true
situation on the ground is not as dire as they make it out to be. Liberal
legalism is dead, and we should stop trying to exhume its corpse.
In order to evaluate whether Posner and Vermeule are correct to tell
liberal legalism to give up the game and leave the field, we must first assess
why liberal legalists think the fight is still worth having. Although Posner
and Vermeule are confident in the political constraints that remain on the
Executive Branch and its agencies despite the absence of any effective statutory constraints,2 1 the authors' odd decision to act as though they are
discovering the absence of these constraints for the first time precludes them
from engaging liberal legalism on the merits of the issue. Why, in contrast to
the legalists who think the law must be more robust, is politics enough? To
18.
19.
20.
21.
POSNER & VERMEULE, supra note 1, at 51.
Id. at 84-85.
Id at 85.
Id. at 14, 113.
948
Texas Law Review
[Vol. 90:943
answer this question, we must first evaluate why the legalists think politics is
not enough. The best analysis of Congress's penchant for delegation and the
problems that follow from it remains Theodore Lowi's The End of
Liberalism: The Second Republic of the United States. For Lowi, the
penchant for delegation enables, in the first place, what he calls "interestgroup liberalism," where organized groups can get what they want because
the absence of clear statutes means that there is no principle according to
which one can tell them "no." 2 2 Interest-group liberalism thrives in a situation where legislators are unwilling to make clear distinctions and clear
choices in their legislation. 2 3 Instead of making choices in the law, the statutory ambiguity creates a continuous process of bargaining that systematically
empowers the organized over the disorganized. Lowi criticizes the congressional penchant for delegation and calls for statutes that are more concrete in
the policy choices they make, not because he fails to realize that politics will
substitute for the law but because he thinks that this substitution is ultimately
wholly insufficient. 24
Lowi's analysis points to another problematic aspect in Posner and
Vermeule's criticism of liberal legalism. Part of the reason they argue that
we should stop worrying about the unconstrained status of the Executive
stems from the superior ability of the Executive Branch to navigate the pace
of the modem world.25 The Executive's unitary character endows it with
necessary speed and efficiency. These strengths of the Executive Branch are
accentuated in crises, both political and economic, but they are also
important in ordinary times. Yet, in Lowi's analysis, it is not the Executive
himself that is empowered by statutory ambiguity so much as the lower level
agency officials who must make the choices that Congress was not willing to
make in its legislation.26 As William Scheuerman suggests in his book about
the ways in which what he calls the "social acceleration of time" transforms
liberal democracy, "the contemporary executive is a complicated institutional
entity, made up of a rich variety of (oftentimes conflicting) bureaucratic
units."27 Thus, he continues: "[T]he emphasis in traditional reflections on the
unitary and even solitary nature of the executive badly obscures the empirical
realities of both the administrative state and present-day executive decision
22. See THEODORE J. Lowl, THE END OF LIBERALISM: THE SECOND REPUBLIC OF THE UNITED
STATES 92-93 (2d ed. 1979) (arguing that "[i]nterest-group liberalism has little place for law
because laws interfere with the political process" and that "[d]elegation of power provides the legal
basis for rendering a statute tentative enough to keep the political process in good working order all
the way down from Congress to the hearing examiner, the meat inspector, the community action
supervis[ojr, and the individual clients with which they deal").
23. Id. at 92.
24. Id. at 124-26.
25. POSNER & VERMEULE, supra note 1, at 41-44.
26. Lowi, supra note 22, at 124-25.
27. WILLIAM E. SCiHEUERMAN, LIBERAL DEMOCRACY AND THE SOCIAL ACCELERATION OF
TIME 101 (2004).
2012]
The Madisonian Constitution: Rightly Understood
949
making." 28 In other words, given the realities of the administrative state, the
question of who makes policy choices might not be so much between an
elected executive and an elected legislature as between a duly responsible
(because elected) legislature and an unelected bureaucracy. Although the
appeal of the rhetoric of speed (in a modem age obsessed with it) naturally
gravitates power and authority to the supposedly speedy Executive Branch,
the reality of the fragmented and bureaucratic administrative state means that
the Executive Branch is not actually all that fast.
The question concerning the relative governing abilities of Congress
and the presidency points to the crux of Posner and Vermeule's critique of
liberal legalism. Their argument about Congress's willingness to pass enabling statutes points ultimately to their claim about the ability and
willingness of Congress to govern more generally. Their deepest critique of
liberal legalism concerns its idealization of representative legislatures as the
only proper governing institutions.29 In articulating this critique, Posner and
Vermeule sometimes make it seem as though it is enough merely to show
that representative legislatures do not actually govern, thus refuting liberal
legalism by the mere fact of this empirical reality. As I have already said,
this argument does not go all that far because, at its best, liberal legalism
argues that legislatures should govern, even if they do not always actually
govern. In fact, so far from blithely assuming that legislatures actually
govern, contemporary liberal legalism often begins from the fact that they do
not. 3 0 At other times in the book, Posner and Vermeule meet this later claim
squarely, suggesting, by contrast to the claims of liberal legalism, that there
is no compelling reason why we should worry about the fact that Congress
does not govern. Our worries, they suggest, stem from a peculiar form of
what they call "tyrannophobia" that is inappropriate in a developed Western
state where we no longer need to fear the emergence of tyrants.3 1 In other
words, the preference for legislative government, they suggest, is ultimately
rooted in an irrational fear of executive power-a fear that we should learn to
abandon. Of course, the liberal legalist might respond by suggesting that it is
tyrannophobia itself that helps to preclude the emergence of tyrants: because
we worry about them, we are less likely to fall prey to them. To some
degree, Posner and Vermeule respond to this argument by citing empirical
evidence that shows that economic well-being is the biggest predictor of the
28. Id.
29. See POSNER & VERMEULE, supra note 1, at 15 ("Liberal legalism's essential failing is that it
overestimates the need for the separation of powers and even the rule of law.").
30. See SHANE, supra note 9, at 3 (arguing that a trend of concentrating power in the hands of
the federal Executive in combination with a campaign to direct the capacities of the national
government toward a conservative agenda has produced an approach to government that vests in the
President a "fixed and expansive category of executive authority largely immune to legislative
control or judicial review").
31. POSNER & VERMEULE, supra note 1, at 176-77.
950
Texas Law Review
[Vol. 90:943
rise of tyranny.32 Countries with a low per capita GDP are susceptible to this
threat, while countries with a high GDP are not. 3 3 From this empirical
evidence, Posner and Vermeule suggest that in the United States, it is our
high GDP, rather than our tyrannophobia, that explains the fact that we have
not had a tyrant arise. 34 In other words, they attempt to prove the irrationality
of tyrannophobia on the basis of this empirical evidence. Of course, the
empirical evidence they point to shows this relationship between economic
factors and tyranny at a macro level. Their empirical evidence shows that, in
general, economic factors are a primary explanatory variable in the rise of
tyranny. While this macro-level evidence might lead us to worry less about
tyranny in the United States because of its demographic factors, it does not
prove "that dictatorship in the strong sense is not a real possibility." Even
more problematic is the claim based on this macro-level evidence that
tyrannophobia "has little social utility in modem circumstances." 36 The fact
that, in general, empirical evidence shows that demographic factors have the
most explanatory power for predicting the rise of tyrannies cannot possibly
tell us anything about the utility of our fear of tyrants. Even if we ignore the
fact that this macro-level pattern cannot really be used to make arguments
about the specific variables at work, or not at work, in the micro level of the
United States, the nature of the macro-level argument itself precludes them
from making the claims they make about the utility of the fear of tyranny.
The empirical evidence about economic factors does not show us what
intervening variables are causing this connection. It might be that the fear of
tyranny is much more active within a society with a high GDP. After all,
economic factors cannot on their own preclude tyranny. Although some of
the empirical studies cited by Posner and Vermeule 37 do establish causality,38
the studies do not establish the nature of this causality and do not rule out
intervening variables like the fear of tyranny. So, to conclude based on this
empirical evidence that there is no utility to tyrannophobia is simply
incorrect. Posner and Vermeule do not provide the evidence they would
need to in order to come to this conclusion.
That Posner and Vermeule choose to refute liberal legalism's concern
with a strong executive in this manner is unfortunate because there are more
interesting directions they could have gone. They are right to find a naivete
32. See id. at 189 (observing that no democracy has ever collapsed in a nation with an average
per capita income greater than slightly over $6,000 in 1995 dollars).
33. See id. (concluding that the best safeguard against the threat of tyranny is probably wealth).
34. See id. at 189-91 (noting both the United States' enormous wealth and the correlation
between wealth and the lack of tyranny).
35. Id at 204.
36. Id.
37. Id at 239 n.53.
38. See, e.g., Adam Przeworski et al., What Makes Democracies Endure?, J. DEMOCRACY, Jan.
1996, at 39, 40-41 ("[E]conomic development has a very strong effect on the probability that
democracy will survive.").
2012]
The Madisonian Constitution: Rightly Understood
951
and idealism in liberal legalism's belief that Congress can and should govern
on its own, with the Judiciary and the Executive merely carrying out
Congress's will. Because Posner & Vermeule owe their conception of the
Constitution entirely to liberal legalism itself, they fail to see that in giving so
much preeminence to Congress, liberal legalism gets the constitutional order
itself wrong. The American constitutional order posits three independent
institutions with separate functions and corresponding-but overlappingspheres of power. 39 And while those functions are sufficiently distinct that
we can tell them apart, they are sufficiently overlapping that no institution is
simply preeminent in any domain. For instance, although Congress is
responsible for legislation, the President is independently tasked with
executing the law, and the courts are tasked with interpreting the law as it
applies to any given case. Constitutionally, even Congress's preeminence
over the domain of legislation contains a variety of potential points of
conflict for the President and the courts, because they are independent, to
engage Congress. For instance, citing his understanding of the Defense of
Marriage Act as an unconstitutional law, President Obama declared that he
would no longer defend it.40 Given that he is constitutionally independent of
Congress-he takes an oath to the Constitution rather than to Congress-it is
fully within the President's power to choose to not execute a law that he
thinks violates his oath.4'
In its predominant conception of the relation between Congress and the
other two institutions, liberal legalism tends to place too much emphasis on
the relative power of Congress under the Constitution. In this sense, Posner
and Vermeule's argument could offer a useful corrective to this tendency.
Their intent, however, is not to correct a misinterpretation of the
Constitution, but to lead us to the abandonment of constitutional thinking
altogether. Because they want us to abandon the Constitution, they insist on,
even to the point of overstating, liberal legalism's claim for the preeminence
of Congress.42 Given Congress's very profound dysfunction, to insist on its
superiority seems, at best, quaint and idealistic and, at worst, downright
dangerous. In critiquing Posner and Vermeule's argument, my intent, then,
39. See JEFFREY K. TULIs, THE RHETORICAL PRESIDENCY 41 (1987) (arguing that while each
branch of government has separate functions, each is merely superior in its own sphere and not the
sole power).
40. Press Release, Dep't of Justice, Statement of the Attorney General on Litigation Involving
the Defense of Marriage Act (Feb. 23, 2011), available at http://www.justice.gov/opalpr/2011/
February/I -ag-222.html (announcing that the Department of Justice would no longer defend § 3 of
the Defense of Marriage Act in two cases pending in the Second Circuit).
41. See Michael Stokes Paulsen, The Most Dangerous Branch: Executive Power to Say What
the Law Is, 83 GEO. L.J. 217, 221-22, 262 (1994) (asserting that the President may decline to
execute laws he believes to be constitutionally invalid and that the President's oath requires him to
exercise full, independent review of Congress's actions).
42. See POSNER & VERMEULE, supra note 1, at 3 (describing liberal legal theory as holding that
"representative legislatures govern and should govern" and that the "law does and should constrain
the executive").
952
Texas Law Review
[Vol. 90:943
is not to defend liberal legalism to the extent that it insists on the preeminence of Congress. Instead, I would suggest that an assessment of our
current situation that aims to rescue the Constitution but that does not fall
prey to liberal legalism's insistence on a preeminent Congress, would help us
the most. In other words, I would propose that we rescue the Constitution
from the clutches of both liberal legalism and legal realism. The Constitution
is a far more subtle and modem document than either of these approaches
realizes or appreciates. It can accommodate, and has accommodated, a wide
variety of institutional arrangements and power balances depending on the
time period. Thus, there is much more room for a powerful presidency
within the constitutional order than Posner and Vermeule accept. And, by
situating a powerful presidency within the constitutional order rather than
against it, we can retain and even encourage the points of conflict that are
encouraged by our system of separation of powers. This is good because
these points of conflict are much more useful than we typically realize.
Moreover, insofar as Posner and Vermeule rest their confidence in a legally
unbounded executive on the fact that political constraints remain,43 they
unintentionally point to the benefits of the very same separation-of-powers
system that they claim has none. They distinguish the legal system of separation of powers from the politics that occurs independently of it,44 but this
distinction becomes highly problematic once we see the extent to which
separation of powers should be understood politically rather than legally.
The constitutional separation of powers establishes the political system
within which the Executive can be constrained. Neither Madison nor any of
the serious Founders would have thought that the Executive could be constrained purely by legal checks. Instead, legal checks are a crucial
supplement to the political checks that are ultimately rooted in the constitutional system itself. Absent this constitutional system of duly independent
branches of power that do not owe their power to the Executive, the politics
of executive constraint becomes much more difficult to achieve. Posner and
Vermeule miss the political constraints arising from the separation of powers
because they have blinders that allow them to see only its legal character.
But in counting on politics to continue to restrain the Executive, they go seriously wrong in doing so much to undercut the basis of those politics. Thus,
many of the political checks that Posner and Vermeule count on would disappear if we abandoned the constitutional separation of powers they
disparage.
43. See id. at 12-14 (explaining that although the President is unbound by legal constraints,
significant political restraints still exist).
44. See id. at 14 (stating that the demise of the separation of powers does not remove political
constraints on the President).
2012]
II.
The Madisonian Constitution: Rightly Understood
953
A Madisonian Constitution
To see the extent to which the politics of executive restraint depends on
the constitutional system of separation of powers, it is necessary to revisit the
system as it was understood by James Madison and others who played a role
in creating it.4 5 I revisit Madison's understanding of this system not to arrive
at the correct "originalist" interpretation of the Constitution but with the hope
that interrogating the thought of one of the Constitution's leading Framers
will help us understand its underlying assumptions and functions. Doing so
helps us clear away some of the misconceptions about the way the
Constitution works best-misconceptions that are problematic not because
they do violence to the "original" Constitution but because they just do not
comport well with the Constitution as it actually is and as it actually frames
power.
Liberal legalism's belief that the Constitution demands and requires
strict legal boundaries on power is one of the most important current miscon46
ceptions of the Constitution. An investigation of Madison's arguments in
The Federalist Papers reveals the extent to which this was simply not his
understanding of the Constitution that the Founders had created. Further, not
only is this not the Constitution they had created, but the Constitution they
did create does not make sense understood through the lens of liberal
legalism. Why endow both the presidency and the courts with independent
power and authority if the ultimate aim is simply to have them carry out the
legal will of Congress? The legalistic and legislative Constitution envisioned
by legal liberalism comports poorly with the fractious and conflictual
Constitution that we actually have. Thus, as I have argued elsewhere, a
proper understanding of the Constitution we actually have would lead
Congress to pass laws that restrict the legal range of the President's power,
understanding as they do so that this does not necessarily foreclose the possibility of the President taking extralegal action.4 7 By contrast, an improper
legalistic understanding of the Constitution leads Congress in exactly the
opposite direction. For all the attempts liberal legalism has made to keep the
presidency within the law, its constitutional understanding causes Congress
to pass exactly the sorts of enabling legislation that Posner and Vermeule use
to show that the presidency is not really within the law at all. In other words,
misunderstanding the constitutional system of independent but overlapping
spheres of power is not problematic simply because it departs from the
Founders' system. I would suggest that there is nothing problematic about
departing from the Founders' system if it results in better government. It is
problematic because the departure from the Founders' system results in an
45. One of the best treatments of Madison's constitutional understanding is GEORGE THOMAS,
THE MADISONIAN CONSTITUTION (2008).
46. See POSNER & VERMEULE, supra note 1, at 3-5 (criticizing liberal legalists' belief that
major constraints on the Executive arise from law or the framework of separated powers).
47. KLEINERMAN, supranote 11, at 221-28.
954
Texas Law Review
[Vol. 90:943
executive that is ostensibly controlled by the law even as the Executive is, in
reality, almost wholly unconstrained.
Of course, the true Madisonian conception of the Constitution and of the
separation of powers does, in a certain sense, resemble liberal legalism
insofar as it recognizes that the real power that Congress possesses vis-a-vis
the presidency lies in the law and its power to make the law. But unlike liberal legalism, which expects Congress to think of itself as the only fount of
legitimate authority, the true Madisonian conception would have Congress
recognize that the President is also institutionally empowered to act, if the
situation demands it, independently of the laws passed by Congress. Instead,
in the Madisonian conception, Congress would be most vigorous about the
law because the law represents the "constitutional rights" of Congress.
Madison wrote: "Ambition must be made to counteract ambition. The interest of the man must be connected with the constitutional rights of the
place." 4 8 This passage, which, I would suggest, stands as one of the most
important guides to the Founders' understanding, emphasizes not only the
extent to which the Constitution is intentionally conflictual-a point oddly
ignored by the supposed realists, Posner and Vermeule-but also the extent
to which that conflict is framed by the Constitution itself. And it is framed
by the Constitution not in the legalistic manner that legal liberalism
conceives, but in terms of each branch asserting its interests by connecting
them to their constitutional rights. Where legal liberalism, to the extent that
it integrates the Madisonian conception of contending ambition, tends to
think of ambition as occurring only within the confines of a legally imposed
Constitution, Madison's actual conception puts the Constitution itself into
ambition's orbit. As I will suggest later, the crucial problem in the present
revolves around the extent to which Congress does not participate in the politics of constitutional ambition. Instead of asserting its constitutional rights
by attempting to insist on its intrinsic legal authority to limit and control the
presidency and, in a different sense, the courts, Congress tends to abdicate its
constitutional authority and defer to the other two institutions. Were it functioning properly, Congress would derive its political authority to act from
In this constitutional
arguments about its constitutional authority.
conception, the Constitution, broadly understood, acts as a fount of political
authority. So, for instance, in recent disputes between Congress and the
President about the authority to take the nation to war, Congress should insist
on its constitutional right, as the most deliberative branch, to make the deliberative decision to take the country to war. By contrast, the Executive should
insist, as it does, that-as the branch in charge of the military, representing
the nation's foreign interests, and most capable of "energetic" action-it
sometimes has to make immediate decisions to take the country into battles.
Each branch defends its political action by the nature of its constitutional
48. THE FEDERALIST No. 51 (James Madison), supra note 17, at 322.
2012]
The Madisonian Constitution: Rightly Understood
955
authority. Congress might supplement its broad constitutional argument by a
legalistic claim about the text of the Constitution that gives it, rather than the
President, the power to declare war. 49 But then this legalistic argument
would be supplementing, rather than supplanting, the broader constitutional
argument about the nature of congressional power and authority. The legalistic argument occurs within the broader confines of an argument about
political authority; the argument about political authority is fueled by and
framed by the constitutional separation of powers. By separating the law and
the Constitution from the realm of politics, Posner and Vermeule missed the
extent to which both are ultimately political.
In conceiving of the Madisonian Constitution as essentially apolitical or
even antipolitical, Posner and Vermeule's position is no different than the
dominant view within the legal academy. Even those "departmentalists"
within the legal academy who want to understand the Constitution in terms
of the separation of powers still tend to think of the Constitution entirely in
legalistic rather than political terms. Their conception of separation of powers resembles more the arguments of the Anti-Federalists5 0 than the
Federalists. 1 Instead of thinking of the boundaries between branches as
emerging politically through conflict between the branches over the authority
to exercise power, the legal departmentalists tend to conceive of each
branch's power discreetly, emphasizing the extent to which the Constitution
grants each branch a full array of powers that fall within its sphere. 52 In
49. Compare U.S. CONST. art. I, § 8, cl. 11 (giving Congress the power to declare war), with id
art. II § 2, cl. I (declaring the President the Commander in Chief of the armed forces).
50. See, e.g., Aristocrotis, The Government of Nature Delineated or an Exact Picture of the
New Federal Constitution, in 3 THE COMPLETE ANTI-FEDERALIST 196, 207-08 (Herbert J. Storing
ed., 1981) ("Perhaps some people may think that power, which the house of representatives
possesses, of impeaching the officers of government will be a restraint upon them; but this entirely
vanishes, when it is considered that the senate bath the principal say in appointing these officers,
and that they are the sole judges of all impeachments. Now it would be absurd to suppose that they
would remove their own servants for performing their secret orders perhaps.").
51. For example, in discussing the different methods of separation of powers used by the states,
Madison stated:
It is but too obvious that in some instances the fundamental principle under
consideration has been violated by too great a mixture, and even an actual
consolidation of the different powers; and that in no instance has a competent provision
been made for maintaining in practice the separation delineated on paper. What I have
wished to evince is that the charge brought against the proposed Constitution of
violating the sacred maxim of free government is warranted neither by the real meaning
annexed to that maxim by its author, nor by the sense in which it has hitherto been
understood in America.
THE FEDERALIST No. 47 (James Madison), supra note 17, at 307-08.
52. See, e.g., Steven G. Calabresi & Saikrishna B. Prakash, The President's Power to Execute
the Laws, 104 YALE L.J. 541, 547-49 (1994) (arguing against the claim that "the administration of
all federal laws is not [an] exclusively executive task[]" (emphasis omitted)); Steven G. Calabresi &
Gary Lawson, The Unitary Executive, Jurisdiction Stripping, and the Hamdan Opinions: A
Textualist Response to Justice Scalia, 107 COLUM. L. REv. 1002, 1005 (2007) (dismissing Justice
Scalia's "claims about Congress's power to strip jurisdiction from the Supreme Court" as "textually
wrong"). See generally Steven G. Calabresi & Kevin H. Rhodes, The Structural Constitution:
956
Texas Law Review
[Vol. 90:943
emphasizing distinct powers that correspond to distinct functions, the legal
departmentalists resemble the Federalists. They depart from the Federalists,
however, in their claim that these separations can and should emerge
They want to parcel both the Constitution's powers and the
legally.
interpretation of the Constitution's powers into discrete units that correspond
only to the distinct branches of the Constitution to which they correspond.54
This is closer to the manner in which the Anti-Federalists understood
the separation of powers than the Federalists. As Madison wrote, "One of
the principal objections inculcated by the more respectable adversaries to the
Constitution is its supposed violation of the political maxim that the
legislative, executive, and judiciary departments ought to be separate and
distinct."5 5 That is, it was the Anti-Federalists, rather than the Federalists,
who understood separation of powers in terms of strictly distinct powers that
would be constitutionally separated. By contrast, Madison suggested that
"unless these departments be so far connected and blended as to give to each
a constitutional control over the others, the degree of separation which the
maxim requires, as essential to a free government, can never in practice be
duly maintained." 6
Madison's argument here is often misunderstood too far in the other
direction. He is often misinterpreted as disagreeing with the principle of
separation itself. In this misinterpretation, Madison aimed only to create, in
the words of Richard Neustadt, "separated institutions sharing powers." 57
One might say that this misinterpretation views the system too politically,
failing to see the constitutional principle of separation underlying the system
of "connected and blended" powers. The powers are overlapping and
checking in order to achieve "the degree of separation" essential to a free
government. Madison's argument against the Anti-Federalists was not that
the separation of powers is a bad idea but that it can, paradoxically, only be
achieved by creating a system of overlapping and connected powers. Unlike
Unitary Executive, Plural Judiciary, 105 HARV. L. REV. 1153 (1992) (contrasting constitutional
interpretations that focus on the powers granted to each branch with those that focus on overlapping
grants of power).
53. See, e.g., Calabresi & Rhodes, supra note 52, at 1157-58 (describing the debates about
Congress's power to restrict or expand the jurisdiction of federal courts through passing laws and
about Congress's power to "vest executive power in subordinate officers while simutaneously
insulating these officers from the President's control").
54. See, e.g., A. Michael Froomkin, Note, In Defense of Administrative Agency Autonomy, 96
YALE L.J. 787, 800 (1987) ("If it were possible to give content to the President's power without
recourse to the text of the constitutional article that purports to define that office, then presumably
the first three articles of the Constitution could each have been expressed in a sentence. Those three
articles do list the areas in which a branch can go beyond its paradigmatic function, but their
primary purpose remains to define the contours of each branch and of offices within each branch."
(footnote omitted)).
55. THE FEDERALIST NO. 47 (James Madison), supra note 17, at 301.
56. THE FEDERALIST No. 48 (James Madison), supra note 17, at 308.
57. RICHARD E. NEUSTADT, PRESIDENTIAL POWER: THE POLITICS OF LEADERSHIP FROM FDR
TO CARTER 26 (rev. ed. 1980).
2012]
The Madisonian Constitution: Rightly Understood
957
Neustadt's characterization, however, Madison's separation-of-powers
system still maintains a separation of powers, not just a separation of
institutions.58 The overlapping constitutional controls are created in order to
maintain the separation of powers rather than to destroy them. Where the
legal departmentalists overemphasize the apolitical nature of the Constitution
and fail to see the extent to which Madison intended to create political disputes about constitutional authority, those like Neustadt overemphasize its
political character and fail to see the underlying constitutional principle of
separation that provides fuel for the political disputes.
So this brings us to the question: Why were overlapping constitutional
controls thought essential to the separation of powers itself? In the first
place, Madison's response to this question shows decisively that he is not a
"liberal legalist" of the sort described by Posner and Vermeule. Madison
wrote: "Will it be sufficient to mark, with precision, the boundaries of these
departments in the constitution of the government, and to trust to these
parchment barriers against the encroaching spirit of power?" 59 That is,
"parchment barriers," or mere legal limitations, will not suffice in the face of
encroaching power. Of course, for Madison at that point, the decisive danger
stemmed not from the Executive but from the Legislature: "The legislative
department is everywhere extending the sphere of its activity and drawing all
power into its impetuous vortex." 60 As I will discuss later, Madison here
both overestimated and misunderstood the nature of legislative ambition.
And, as I have argued elsewhere, he realized in the 1790s that the primary
danger in a republic arose not from the Legislature but from the Executive.61
But the basic insight remains, even if its application should be different.
Madison repeated the basic insight at the end of that same essay: "[A] mere
demarcation on parchment of the constitutional limits of the several departments is not a sufficient guard against those encroachments which lead to a
tyrannical concentration of all the powers of government in the same
hands."62
In the essay itself, Madison went to great lengths to illustrate with the
various state governments the ways in which mere demarcations on parchment have been insufficient against the concentration of power in the
Legislature. 6 3 In this sense, it is worth noting that part of Madison's constitu-
58. THE FEDERALIST No. 51 (James Madison), supra note 17, at 323 ("In a single republic, all
the power surrendered by the people is submitted to the administration of a single government ....
In the compound republic of America, the power surrendered by the people is first divided between
two distinct governments, and then ... subdivided among distinct and separate departments.").
59. THE FEDERALIST No. 48 (James Madison), supra note 17, at 308.
60. Id. at 309.
61. KLEFNERMAN, supranote 11, at 122.
62. THE FEDERALIST No. 48 (James Madison), supra note 17, at 313.
63. See id. at 310-12 (explaining that the legislatures in both Virginia and Pennsylvania
exceeded constitutional limitations on their power by usurping prescribed judicial and executive
powers).
958
Texas Law Review
[Vol. 90:943
tionalism is based on an empirical sense of the actual situation on the ground.
Thus, if Posner and Vermeule are right that the Executive now dominates the
field constitutionally but are wrong that we need not worry about it, a truly
Madisonian constitutional response would have to take its bearings by this
empirical fact. After all, if it is correct to view Madison's constitutionalism
as much more profoundly political than either liberal legalism or Posner and
Vermeule realize, then it does not make sense to "recover" Madison by
insisting on an apolitical, "originalist" understanding of the Constitution.
The Constitution should instead always be interpreted in terms of its political
results, given its underlying aims. Thus, I would say that the Founders' misplaced worries about the Legislature at the founding of the Constitution may
have more constitutional import than I can fully develop in this Review. It
might not be enough simply to apply the same insight regarding the
encroaching nature of power to the presidency and the courts-two
institutions that have fulfilled the Founders' expectations regarding ambition
and its relation to the constitutional order. Instead, because much of the constitutional structure itself is premised on the need to buttress the power of the
courts and the presidency-and to restrain the power of Congress-we may
need to rethink the nature of that structure itself. What should the constitutional structure look like, given that Congress needs to be encouraged to act
rather than discouraged? The answer to this question may not require a fundamental change to the Constitution's structure. I would suggest that
Congress possesses all the power it needs in the current structure. The
problem, instead, is that because of their fears of congressional overreach,
the Founders couched those powers in constitutional language that would
make Congress's constitutional authority to exercise them problematic."
Given their mistaken fears, I would suggest that we should rethink the constitutional language so as to encourage congressional ambition. For instance,
the Founders couched the impeachment power in legalistic language in order
to discourage Congress from using it. Given Congress's unwillingness to use
the impeachment power, I think we should become more open and accepting
about its political use.65
64. See THE FEDERALIST No. 48 (James Madison), supra note 17, at 310 (positing that the
extensive powers of the Legislative Branch, which are difficult to define with precision, permit it to
indirectly encroach on the other branches); THE FEDERALIST No. 71 (Alexander Hamilton), supra
note 17, at 431-32 (lamenting the tendency of the Legislative Branch to absorb the authority of the
other branches and cautioning against a separation of powers that is merely nominal); see also
Buckley v. Valeo, 424 U.S. 1, 129 (1976) (per curiam) ("[T]he debates of the Constitutional
Convention, and the Federalist Papers, are replete with expressions of fear that the Legislative
Branch of the National Government will aggrandize itself at the expense of the other two
branches.").
65. Jeffrey Tulis has also made this argument. See Jeffrey K. Tulis, Impeachment in the
Constitutional Order, in THE CONSTITUTIONAL PRESIDENCY 229, 238-41 (Joseph M. Bessette &
Jeffrey K. Tulis eds., 2009) (purporting that for the impeachment power to be effective, Congress
must think of this power as a political tool to wield against breaches of the Constitution or
"maladministration," even if they are not legal violations).
2012]
The Madisonian Constitution: Rightly Understood
959
To return to Madison's own constitutional conception, the inadequacy
of parchment barriers requires a political solution. And having first investigated Jefferson's solution (i.e., the possibility of seeking a political solution
in frequent constitutional conventions of the people themselves) Madison
66
concludes for a variety of reasons that this solution will be inadequate.
Instead, the solution to the problem lies in the manner in which one
structures the different departments of government. To structure the departments well, Madison argues that "each department should have a will of its
own." 67 Constitutional independence is possible only if it is supported by
political independence. Again, one of the mistakes often made about the
Madisonian system of checks and balances is that while it is true that the
"several constituent parts may, by their mutual relations, be the means of
keeping each other in their proper places," 68 this does not mean that the
sources of power are blended. That is, the checks occur on the exercise of
power, not on its source. In fact, Madison was confident that the branches
would check each other in the exercise of power only because they do not
depend on one another for support. 69 If the President simply depended on
Congress for his continuation in office, he would not have the political
independence necessary to exercise his constitutional independence. The
most illustrative example of the importance of political independence comes
in the seeming exception to the rule: the Judiciary. Whereas members of
Congress and the President derive their power from sources wholly independent of one another, members of the Court are chosen by the President
and approved by Congress-thus the worry would be that they would be
politically dependent on those who chose them. Madison responded to this
worry by emphasizing that "the permanent tenure by which the appointments
are held in that department must soon destroy all sense of dependence on the
authority conferring them." 70 That is, the fact that members of the courts no
longer immediately depend on Congress and the presidency for their continuation in office provides them with the political independence that they
require in order to exercise their crucial constitutional functions.
Liberal legalism often also attributes to the Madisonian Constitution the
view that it is the particular province of the courts to enforce the constitutional limits on the other branches of power. For instance, Keith Whittington
articulates and defends what he characterizes as Chief Justice Rehnquist's
66. See THOMAS, supra note 45, at 160 (describing Madison's denial that the majority is the
standard of right and wrong despite affirming that the people are the best protectors of their own
rights).
67. THE FEDERALIST NO. 51 (James Madison), supranote 17, at 321.
68. Id. at 320.
69. See id at 321-22 (asserting that the "separate and distinct exercise of the different powers
of government" requires that the members of each department of government have "as little agency
as possible in the appointment of the members of the [other departments]" and that the departments
be given "the necessary constitutional means . . . to resist encroachments of the others").
70. Id. at 321.
960
Texas Law Review
[Vol. 90:943
view of the courts as a referee or "umpire" of the separation of powers.' In
fact, in some articulations of the constitutional system of separation of
powers, the separation itself depends entirely on courts to enforce it.72 If
Madison is properly understood as a liberal legalist of the sort characterized
by Posner and Vermeule, then it would be expected that he would have held
something like this view. After all, if the separation of powers is enforced
legally, then it would seem the natural responsibility of the branch most representing the interpretation and legal enforcement of the Constitution. But in
this same essay articulating a political conception of the Constitution,
Madison also implicitly rejected the notion of the courts as the enforcers of
constitutional boundaries. Speaking of his worry about unjust majorities
uniting "by a common interest" and making "the rights of the minority ...
insecure," Madison wrote, "There are but two methods of providing against
this evil: the one by creating a will in the community independent of the
Of this first method, Madison
majority-that is, of the society itself... .
claimed this is what prevails "in all governments possessing [a] hereditary or
self-appointed authority."74 He continued: "This, at best, is but a precarious
security; because a power independent of the society may as well espouse the
unjust views of the major as the rightful interests of the minor party, and may
possibly be turned against both parties."75 Much of the notion of the courts
as "prudent umpires" of the separation of powers and of constitutional
limitationS76 depends on the possibility of some group being independent of
the politics of society. Madison here rejected the very possibility of such
political independence. There can be no group sufficiently independent to
serve as an umpire of the Constitution's limits. The Constitution's limits
must emerge within politics; they cannot be imposed from above.
Of course, Madison here, in defending the Constitution, cannot be said
to have anticipated the claims about the courts as the umpire of constitutional
boundaries. But both this argument and the related argument for judicial
supremacy emerged early in our constitutional politics. As early as 1789,
Madison already had begun articulating a critique of such claims. In a debate
in the House of Representatives concerning who possesses the constitutional
power to remove executive officers, Madison raised the question of whether
71. Keith E. Whittington, The Casey Five Versus the Federalism Five: Supreme Legislator or
Prudent Umpire?, in THAT EMINENT TRIBUNAL: JUDICIAL SUPREMACY AND THE CONSTITUTION
181, 186-87 (Christopher Wolfe ed., 2004).
72. See, e.g., David Gray Adler, The Judiciaryand PresidentialPower in Foreign Affairs: A
Critique, 1 PERSP. ON L. & PUB. INT. 1, 9-11 (1996) (discussing how the Court's prior decisions
contributed to an unjustifiable growth of executive power over foreign affairs). This is also, in a
certain sense, the view even of those who call for a more restrained judiciary but one that
nonetheless remains the "umpire" of the separation of powers. E.g., Whittington, supra note 71, at
186-87.
73. THE FEDERALIST No. 51 (James Madison), supra note 17, at 323.
74. Id. at 324.
75. Id
76. See supranote 71 and accompanying text.
2012]
The Madisonian Constitution: Rightly Understood
961
"the Legislature itself has [the] right to expound the Constitution"" and "that
wherever [the Constitution's] meaning is doubtful, you must leave it to take
its course, until the Judiciary is called upon to declare its meaning." After
acknowledging that "in the ordinary course of government, . . . the exposition
of the laws and Constitution devolves upon the Judiciary," Madison begged
to know "upon what principle it can be contended, that any one department
draws from the Constitution greater powers than another, in marking out the
limits of the powers of the several departments?" 79 The three independent
departments of the national government each have an independent right to
interpret the Constitution for themselves in attempting to claim and justify
their political authority.
Madison's rejection of the Judiciary's supreme authority to interpret the
Constitution points back to the deepest problem with Posner and Vermeule's
characterization of what they continually call "Madisonian" liberalism. The
deepest problem lies in their profoundly antipolitical conception of the
Madisonian constitutional order-a conception they unfortunately share with
contemporary liberalism. The concentration on the law in opposition to politics and on the Supreme Court as the final arbiter of the law's meaning and
application stems in part from a fear of the messiness and unkempt nature of
politics. 80 This essentially antipolitical conception is unwittingly revealed by
Posner and Vermeule themselves in their discussion of the politics that would
constrain the Executive since the law does not. In the first place, their rejection of the possibility that the law is, and should be, political 8' stems from a
lingering attachment to the pristine nature of the law. The law must simply
be the law; if it is more or less than that, it is entirely political. Moreover,
their conception of politics,82 as opposed to law, is strikingly restricted. It
seems to consist only of a relationship between a powerful presidency and
the essentially passive people whose "political" function is just to judge the
president's activity through reelection and presidential approval polls.
By contrast, properly understood, Madisonian liberalism attempts to
preserve republican politics while, at the same time, refining it. In
Madisonian liberalism, the intermediate powers, or what Madison called
"auxiliary precautions," are essential. 83 Thus, while it is true that Congress
77. 1 ANNALS OF CONG. 500 (1789) (Joseph Gales ed., 1834).
78. Id.
79. Id.
80. See Larry Alexander & Frederick Schauer, On Extrajudicial ConstitutionalInterpretation,
110 HARV. L. REv. 1359, 1376 (1997) (describing how authoritative interpretation of the
Constitution stabilizes against "shifting political fortunes" and "divergent substantive views and
personal agendas").
81. POSNER & VERMEULE, supra note 1, at 56-57 (suggesting that even independent agencies,
like independent judiciaries, are not insulated from politics due to the President's ability to appoint
agency heads with similar political preferences and practices that require independent agencies to
consult or collaborate with the President through his agents).
82. Id. at 88.
83. THE FEDERALIST NO. 51 (James Madison), supra note 17, at 322.
962
Texas Law Review
[Vol. 90:943
was, on the whole, rather passive during at least the early years of the
George W. Bush Administration, and while it is also true that the Supreme
Court's "war-on-terror" rulings were not as far-reaching as some claimed,
these institutions still contributed to a reining in of presidential power.84
Even if the Supreme Court was either unwilling or unable to enforce the law
on the Bush Administration insofar as it was not able to make the President
close Guantanamo, and even if Congress was either unable or unwilling fully
to rein in the President's exercise of war powers related to the confrontation
with terrorism, what they did do contributed significantly to the political
debate about the war on terror. To dismiss their efforts because the effects of
their actions were legally ambiguous 85 is to attribute too much power to the
legal aspect of separation of powers and insufficient power to the politics that
arise from the separation of powers. Madison saw the spheres of power as
variable and dynamic, depending on the politics arising from the constitutional authority asserted by the members of each branch. 86 The control of the
Executive would emerge from that dynamic and variable political situation
that, while it includes the law, is not exhausted by the law. By contrast,
Posner and Vermeule see the control of the Executive entirely in terms of the
legal boundaries established by the system of separation of powers.87
Although this is one component of the system, it simply does not exhaust it.
Failing to see this broader separation of powers, Posner and Vermeule see the
failure of this component of the system as the failure of the system itself.
III. Madisonian Realism vs. Legal Realism
Posner and Vermeule's argument emerges from legal realism. They
contrast their tough-minded view of the true applicability of the law with the
more idealistic views emerging from liberal legalism.88 And, where liberal
legalism laments the failure of the law, Posner and Vermeule embrace it. As
legal realists, they do not think the law could ever be as effective as its proponents would hope. Thus, rather than lament the impossible, they attempt
to move us beyond such false hopes. Instead of continuing to hold out the
hope that we could get the Executive within the law, we should accept and
even embrace the extent to which the Executive remains bound by politics
even if not bound by the law. 89 Legal realism begins with a realistic perspective on the true inefficaciousness of the law. And given this fact, it places
84. For a fuller case supporting this claim, see generally JACK GOLDSMITH, POWER AND
CONSTRAINT: THE ACCOUNTABLE PRESIDENCY AFTER 9/11 (2012).
85. POSNER & VERMEULE, supra note 1, at 36 (explaining how the numbers of releases from
Guantanamo resulting from a judicial order of release may understate or overstate the in terrorem
effect ofjudicial oversight).
86. THE FEDERALIST No. 51 (James Madison), supra note 17, at 320-23.
87. POSNER & VERMEULE, supra note 1, at 118-22 (analyzing constraints on the Executive
based on arguments for and against the effectiveness of the separation of powers).
88. Id. at 14-15.
89. Id.
2012]
The Madisonian Constitution: Rightly Understood
963
much more importance on politics and on political bargaining. Illustrating
this, Posner and Vermeule spend a considerable part of the book creating
rational-choice models of political bargaining to explain executive behavior.
Although I think such models can be exceedingly useful either to explain
what has happened90 or to predict what will happen, they are limited in the
extent to which they can create a truly robust account of the actual political
situation. They can capture well and help cast light on the complexities of
political bargaining, but they systematically fail to capture the much more
complicated matters that are involved with claims to political authority. In
this deeper sense, although legal realism is correct that the law cannot possibly provide for its own application, the law does help to shape the context in
which we will understand and judge claims to political authority.
The limits of legal realism can be seen particularly well by examining
its explanation of what happened during the Bush presidency. In both this
book and even more in their last book, Terror in the Balance: Security,
Liberty, and the Courts, Posner and Vermeule claim that the public
receptiveness to the claims of a robust executive power advanced by the
Bush Administration waned as we got further from September 11 and thus
further from palpable danger.9 1 Although this is surely part of the story,
Posner and Vermeule attempt to make it the whole story. And it is this
attempt that shows the limits of legal realism. Given the presuppositions of
their rational-choice political model, Posner and Vermeule simply cannot
contemplate or integrate the possibility that the problems for the Bush
Administration lay in the nature of their claims to political authority itself.
The starting assumption of legal realism is that the law is wholly colored by
the political situation from which it emerges. This means that the Bush
Administration's claim to a constitutional guarantee of a wholly unfettered
executive is compelling so long as there is a political situation to justify it.
When the public felt most threatened, it was willing to support this argument.
When that threat receded, the argument became much more problematic.
Although this explains part of what occurred, the odd thing about this sort of
explanation is that it is ultimately antipolitical even as it claims to be more
so. It is antipolitical because it rejects the essence of politics by rejecting the
political claims that undergird and surround political bargaining. In a constitutional order based on a system of separation of powers, this rejection is
especially unfortunate because the Founders thought the separation would
emerge from the necessarily differentiated nature of the claims to political
authority.
90. See, e.g., Jamie L. Carson & Benjamin A. Kleinerman, A Switch in Time Saves Nine:
Institutions, Strategic Actors, and FDR's Court-PackingPlan, 113 PUB. CHOICE 301, 306 (2002)
(analyzing FDR's court-packing plan under a game-theory model).
91. ERIC A. POSNER & ADRIAN VERMEULE, TERROR IN THE BALANCE: SECURITY, LIBERTY,
AND THE COURTS 42-43 (2007).
Texas Law Review
964
[Vol. 90:943
By contrast, a more robust explanation of the politics surrounding the
Bush Administration should integrate the critiques that emerged-critiques
that were ultimately rooted in the separation of powers itself. The disintegration of the Bush Administration's political standing in its claims to
unfettered authority in the war on terror partially related to these critiques. In
the first place, these critiques emerged most forcefully in the context of the
Supreme Court's first significant war-on-terror decisions, Hamdi v.
Rumsfeld 92 and Rasul v. Bush.93 Both of these decisions revolved around the
right of individuals to seek some sort of judicial redress for their detainment.
The essential question was whether the conditions of war allowed the Bush
Administration to hold both American citizens (Hamdi) and non-American
citizens (Rasul) for extended periods of time with absolutely no access to the
courts. 94 Posner and Vermeule would explain these rulings-both of which,
though not wholly destroying the Bush Administration's detention programs,
did significantly alter their course-as the result of the passage of time since
the September 11 attacks. But I would suggest that there is more going on
here than that. While I have been critical of one of these rulings-Hamdielsewhere, 9 5 the decisions do illustrate well something about the nature of
constitutional politics. Whereas Posner and Vermeule's monochromatic
opinion of politics views these sorts of decisions entirely in terms of a
political situation in which the Bush Administration was somewhat
unpopular but not sufficiently unpopular as to allow (or urge) the Supreme
Court entirely to strike down the Administration's detention program, 9 6 1
would suggest that the nature of the rulings can be explained much more
fully by reflecting on the nature of the Court's constitutional authority. To
the extent that the Supreme Court had political authority on this issue, it did
so because the individual rights of these detainees were at stake. In curbing
the claims to unfettered constitutional authority advanced by the Bush
Administration, it did so within its sphere of constitutional authority. If the
Supreme Court had attempted to jettison the Bush Administration's detention
program in toto, it would have been on considerably shakier political ground.
The Supreme Court went as far as it could go within its sphere of
constitutional authority. And although I would suggest that the judicialsupremacist claims implied in these cases, and made explicit in Boumediene
v. Bush, 97 are ultimately at odds with the Supreme Court's actual
constitutional and political authority, 98 the important thing is to see the
92. 542 U.S. 507 (2004).
93. 542 U.S. 466 (2004).
94.
95.
96.
97.
Hamdi, 542 U.S. at 516; Rasul, 542 U.S. at 475-76.
KLEINERMAN, supranote 11, at 15.
POSNER & VERMEULE, supra note 1, at 35-36.
128 S. Ct. 2229 (2008).
98. See id. at 2258-59 ("[Tjhe Government's view is that the Constitution had no effect [at
Guantanamo], at least as to noncitizens, because the United States disclaimed sovereignty in the
2012]
The Madisonian Constitution: Rightly Understood
965
political dynamics that arise within a system of separation of powers. Once
we see the politics of constitutional authority that occur within the
Madisonian Constitution, then we will be in a better position to understand
the problematic nature of the Supreme Court's claim to final constitutional
authority.
The Supreme Court's rulings were also then important in framing the
subsequent political debate about the Bush Administration's constitutional
authority. In Posner and Vermeule's account, the political dynamics go
entirely in one direction. The Supreme Court's rulings are shaped by the
political situation at that time; Posner and Vermeule do not make room for
the possibility that the rulings themselves will subsequently alter the political
situation on the ground. 99 But this is precisely what happened. The Supreme
Court's rulings in these cases brought even greater national attention to the
detention programs, and this attention, in conjunction with the passage of
time between these decisions and September 11, changed the political
situation for the Bush Administration. For all of its pretenses to toughminded realism, Posner and Vermeule's account becomes profoundly
unrealistic to the extent that it abstracts from and simply ignores the real
effect that cases like Rasul, Hamdi, Hamdan,0 0 and Boumediene had on the
political situation confronted by the Bush Administration.10 ' Of course, they
are right that these cases did not have quite the legal ramifications for which
some of their supporters might have hoped. 10 2 But these decisions as to what
the Supreme Court thought the law required as it pertained to the individual
rights of these detainees did have profound political implications for the
direction taken by the Bush Administration. 0 3 Both the Detainee Treatment
Act' 0 4 and the Military Commissions Act,105 neither of which gave the Bush
formal sense of the term. The necessary implication of the argument is that ... it would be possible
for the political branches to govern without legal constraint.").
99. See POSNER & VERMEULE, supra note 1, at 207 (calling the Court a "marginal player").
100. Hamdan v. Rumsfeld, 548 U.S. 557 (2006).
101. See Linda Greenhouse, Justices, 5-3, Broadly Reject Bush Plan to Try Detainees, N.Y.
TIMES, June 30, 2006, at Al ("The ruling marked the most significant setback yet for the
administration's broad expansions of presidential power.").
102. Cf Jonathon Mahler, Why This Court Keeps Rebuking This President, N.Y. TIMES,
June 15, 2008,
available at http://www.nytimes.com/2008/06/15/weekinreview/15mahler.html
(stressing how unusual it was for the Court to rule against the President during wartime but
concluding that "it's hard to know exactly what the ultimate effect of the court's enemy combatant
decisions is going to be").
103. See David A. Martin, Judicial Review and the Military Commissions Act: On Striking the
Right Balance, 101 AM. J. INT'L L. 344, 349 (2007) ("In response to both Rasul and Hamdiactually within two weeks after the Court's 2004 rulings-the executive branch significantly
changed its detention procedures, evidently to reduce exposure to negative rulings in the predictably
forthcoming Guantdnamo habeas challenges.").
104. Detainee Treatment Act of 2005, Pub. L. No. 109-148, 119 Stat. 2739 (codified at 42
U.S.C. §§ 2000dd to 2000dd-1 (2006)).
105. Military Commissions Act of 2009, Pub. L. No. 111-84, 123 Stat. 2574 (codified as
amended at 10 U.S.C. §§ 948a to 950t (Supp. III 2010)); Military Commissions Act of 2006, Pub.
L. No. 109-366, 120 Stat. 2600, invalidatedby Boumediene v. Bush, 128 S. Ct. 2229 (2008).
966
Texas Law Review
[Vol. 90:943
Administration everything it wanted in its confrontation with terrorism, arose
as a direct result of the Supreme Court's decisions. 10 6
Where legal realism creates an artificial and ultimately misleading
bifurcation between the law on the one hand and politics on the other,
Madisonian realism is much more realistic in its view of the true power of
law. Absent a strong political structure, the law is little more than
"parchment barriers" that collapse as soon as they are pushed.o 7 So, like
legal realism, Madison, along with many of the rest of the American
Founders, was profoundly aware of the fact that the law is intrinsically
connected to politics. But where legal realism concludes the utter powerlessness of law from its political connection, Madisonian realism comes to
the more sensible conclusion that while the law is more limited in its power
than some of its more fervent proponents (e.g., Anti-Federalists) hoped, it
still has profound political relevance.
To see Madison's "realistic" argument for the continued relevance of
the law, I will turn now to the famous debate he engaged in with Jefferson
about the wisdom of changing the Constitution on a regular basis. He first
recorded Jefferson's argument from Notes on the State of Virginia that
"whenever any two of the three branches of government shall concur in
opinion, each by the voices of two thirds of their whole number, that a
convention is necessary for altering the Constitution, or correcting breaches
of it, a convention shall be called for the purpose."' 0 8 Under the theory that
"the people are the only legitimate fountain of power," 09 which "seems
strictly consonant to the republican theory,""10 Jefferson's argument would
seem most compelling. Madison's response to it, however, shows the extent
to which he thought the law-or more precisely in this case, the
Constitution-had power over politics. The "frequent appeals" to the people
regarding the powers of the Constitution that would follow from Jefferson's
proposal "would, in great measure, deprive the government of that veneration
which time bestows on everything, and without which perhaps the wisest and
freest governments would not possess the requisite stability.""' The reverence for the Constitution provides a crucial supplement to politics insofar as
it provides stability and, to the extent that the constitutional structure itself is
rational, rationalizes politics. It is through the mediating effect of the
Constitution that Madison hopes "the reason, alone, of the public" will
106. See Jesse Choper & John Yoo, Wartime Process: A Dialogue on Congressional Power to
Remove Issues from the Federal Courts, 95 CALIF. L. REv. 1243, 1244-45 (2007) (describing how
Congress enacted both the Detainee Treatment Act and Military Commissions Act in response to
Supreme Court decisions).
107. See supra notes 17, 59 and accompanying text.
108. THE FEDERALIST No. 49 (James Madison), supra note 17, at 313.
109. Id.
110. Id. at 314.
111. Id.
2012]
The Madisonian Constitution: Rightly Understood
967
"control and regulate the government."I12 The Constitution, Madison argues,
will fortify the reason of man, which is "timid and cautious when left alone"
but "acquires firmness and confidence in proportion to the number with
which it is associated."' 13
Although an admittedly dicey proposition given the Federalists'
argument about the typical instability of law, we can, in a certain sense,
extrapolate this subtle and important argument about the Constitution to law
as such. The extrapolation follows from the last argument. The last argument actually points to something like a coordination game. The stability of
the Constitution changes politics because it provides a point around which
reasonable people can coordinate. Without the Constitution, reasonable people would have a harder time coordinating their opposition to unreasonable
policies. Posner and Vermeule's claim that neither the Constitution nor the
law has any real effects" l4 entirely ignores the coordinating component of
both. Opposition to governmental policies or actions consistently organizes
around the principle that they are unconstitutional. For instance, the opposition to the Bush Administration's detention and interrogation policies in the
war on terror unified politically around its claims that these actions were
Moreover, the Bush Administration's violation of the
unconstitutional."1
Foreign Intelligence Surveillance Act and the revelations about wiretapping
by the National Security Agency further galvanized opposition to President
Bush.116 The political opposition could coordinate and strengthen itself
through its claims about constitutional and legal violations. To dismiss the
law because it is political is to miss how powerful the law is precisely
112. Id. at 317.
113. Id. at 315.
114. See POSNER & VERMEULE, supra note 1, at 61 (discussing how constitutional procedures
are not the source of restraints on the Executive Branch).
115. See, e.g., Editorial, The Imperial Presidency 2.0, N.Y. TIMES, Jan. 7, 2007, at WK13
(criticizing "Mr. Bush's inhumane and unconstitutional treatment of prisoners"); Kenneth Roth, OpEd., Time to Stop "Stress and Duress," WASH. POST, May 13, 2004, available at http://
(arguing that the Bush
www.washingtonpost.com/wp-dyn/articles/A22623-2004Mayl2.html
Administration's interrogation techniques violated international treaties and that similar techniques
had been found unconstitutional when used by police in the United States); Farah Stockman &
Charlie Savage, Battles Brewing on Torture, Detainees, BOS. GLOBE, Oct. 18, 2006, available at
http://www.boston.com/news/nation/washington/articles/2006/10/18/battlesbrewing-on-torture
detainees/ (reporting that the ACLU called recently signed interrogation legislation "blatantly
unconstitutional").
116. See, e.g., Ronald Dworkin et al., On NSA Spying: A Letter to Congress, N.Y. REV. BOOKS,
Feb. 9, 2006, available at http://www.nybooks.com/articles/archives/2006/feb/09/on-nsa-spying-aletter-to-congress (arguing that the President lacks the authority to directly contravene a law passed
by Congress and that the wiretapping policies are unconstitutional under the Fourth Amendment);
Bob Herbert, Op-Ed., Who Will Stand Up for the Constitution?, N.Y. TIMES, Jan. 19, 2006, at A23
(calling the warrantless wiretapping program a "trampling of the Constitution"); Edward M.
Kennedy, Op-Ed., On Wiretapping, Bush Isn't Listening to the Constitution, BOS. GLOBE, Dec. 22,
2005, available at http://www.boston.com/news/globe/editorial opinion/oped/articles/2005/12/22/
onwiretappingbush isnt listening to the constitution/ (calling the warrantless wiretapping
policy an "arrogant usurpation of power and [a] refusal to follow well-established . .. laws").
968
Texas Law Review
[Vol. 90:943
because it is political. And this is why Madisonian realism is actually much
more profoundly realistic in its conception of the whole of politics than legal
realism. While realizing that the law does not enforce itself automaticallythe law requires the support of politics-Madisonian realism also realizes
that the law transforms politics.
IV. Madison's Mistake: Legislative Ambition
Once we realize that Madison had a realistic view of politics, then the
nature of our assessment of him must change. As it stands now, legal realists
like Posner and Vermeule chastise Madison for having had too idealistic a
view of the way politics should work." 7 By contrast, liberal legalists chastise us for failing to live up to the Madisonian ideals of action.' 18 But
Madison was not an idealist of this sort. Instead, he described politics as he
found it, helping to create and then defend an institutional arrangement that
channels human nature as he found it into good government. Then we can
turn from an evaluation of Madison's project on terms that he would not recognize to an evaluation on his own terms. Madison's constitutionalism
began with what might be called an empirical assessment of the way in
which politics actually works, and then it constructed institutional arrangements on that assessment.' 19
A Madisonian empirical assessment of the separation of powers begins
with the following insight: "In republican government, the legislative
authority necessarily predominates." 20 This means that while it is desirable
for each department to be equal and independent within its sphere, it is "not
possible to give to each department an equal power of self-defense."' 2 1 The
Legislature, Madison thought, will naturally have the greatest power of selfdefense because its roots are most in the people, so it will enjoy the greatest
political authority coming from the people.122 The "weight" of this natural
source of political authority "requires that [the Legislature] should be thus
divided." 23 The extent of the Legislature's authority requires both the intentional weakening of its power within the Constitution and the fortification of
the other two branches. Both the Executive and the courts must be constitutionally strengthened over and against the Legislature, or else the Legislature
will draw "all power into its impetuous vortex."' 24 In stark contrast to the
account of Madisonian liberalism offered by Posner and Vermeule, Madison
117.
E.g., POSNER & VERMEULE, supra note 1, at 18.
118. See infra text accompanying notes 128-32.
119. See THE FEDERALIST No. 51 (James Madison), supra note 17, at 322 (describing the
ambition at play in politics and arguing for a government structure that separates powers so that
"[a]mbition [can] be made to counteract ambition").
120. Id.
121. Id.
122. See supra notes 99-110 and accompanying text.
123. THE FEDERALIST NO. 51 (James Madison), supra note 17, at 322.
124. See supra note 60 and accompanying text.
2012]
The Madisonian Constitution: Rightly Understood
969
did not want the Legislature to dominate the rest of the government. 25In
fact, he was concerned about finding a way to prevent the Legislature from
dominating the rest of the government.126
As Madison indicates, the Constitution goes out of its way to fortify the
Executive and the courts at the expense of the Legislature.1 27 It does so not
because it wants a weak legislature but because it assumes that the
Legislature will be too strong without these constitutional adjustments.
Given our current political situation, we should reexamine this assumption,
based as it is on a realistic assessment of the political world as it actually is.
Insofar as Madison himself understood his political solutions as necessarily
following from the political situation as it is, Madison's own politics require
that we reassess his initial assumption about legislative ambition and
strength.
Posner and Venneule are on the most solid ground in their
characterization of the ineffectual and apathetic nature of Congress. Rather
than being the most ambitious and the most aggressive in its assertion of its
power, the Legislature is content to pass off responsibility either to the
presidency, as Posner and Vermeule show,1 2 8 or to the courts. 12 9 It may be
more precise to say that legislative ambition has been divorced from institutional ambition to a much greater degree than in the other institutions. While
the President's individual ambition is intrinsically connected to advancing
the power of the presidency, and a member of the Court's ambition is rooted
in the stature of the Court, a member of Congress can achieve this individual
ambition without advancing-and even at the expense of-the institutional
powers and standing of Congress. Members of Congress exist in what might
be roughly characterized as a collective-action dilemma. While it would be
better for Congress collectively to enjoy more power and standing, it is not
necessarily in the individual interest of any member of Congress to spend his
time advancing the power and standing of the institution as a whole. Instead,
as has been well documented by empirical research in political science,
125. See THE FEDERALIST No. 51 (James Madison), supra note 17, at 322 ("In republican
government, the legislative authority necessarily predominates. The remedy for this inconveniency
is to divide the legislature into different branches; and to render them, by different modes of
election and different principles of action, as little connected with each other as the nature of their
common functions and their common dependence on the society will admit. It may even be
necessary to guard against dangerous encroachments by still further precautions.").
126. See id. at 322-23 (advocating for both bicameralism and the presidential veto as checks on
the power of the national Legislature).
127. Id.
128. See POSNER & VERMEULE, supra note 1, at 50-52 (asserting that during crises, Congress
delegates its authority to the Executive).
129. Neal Devins, Congressas Culprit:How Lawmakers Spurred on the Court's Anti-Congress
Crusade, 51 DUKE L.J. 435, 441-42 (2001) ("Congress is increasingly abdicating its core powers,
including its power to interpret the Constitution independently. Sometimes, Congress treats the
Constitution as the exclusive province of the Supreme Court; on other occasions, Congress simply
seems indifferent to the constitutionality of its enactments . . . ."(footnote omitted)).
970
Texas Law Review
[Vol. 90:943
members of Congress can actually win reelection by running against
Congress itself.130 Their individual ambitions are not only divorced from
their ambitions for the institution; they actually come at the expense of the
institution as a whole.
This failure of Congress to assert itself in the institutional conflict
envisioned by the Founders has profoundly deleterious effects on the
functioning of the separation-of-powers system as a whole. Whereas the
Madisonian system would have expected the ambition and assertiveness of
members of Congress to control both the overinflated claims of the presidency and the judicial-supremacist claims of the Supreme Court, Congress
instead consents to and even encourages these sorts of claims.13 1 As Jeffrey
Tulis shows, this systematic deference by Congress to the other two branches
causes a profound imbalance in the conflictual system envisioned by James
Madison. 132
Moreover, the fact that Congress fails to be an active player in that
conflict becomes even more problematic in light of the Constitution's
assumption that Congress would be more involved in that fight. The
Constitution goes out of its way to insulate both the President and the courts
from the predations of an overly aggressive Congress. Additionally, the
Founders' expectation was not only that Congress would enjoy the most
republican authority-and thus be most dangerous-but also that the courts
and the President would suffer from a lack of popular authority.133 They both
had to be constitutionally empowered and buttressed because their authority
would always be suspect with the people. Instead, we have a scenario where
the courts enjoy more popular support than any other institution in our
system of government.134 And the President has become a political symbol
of the nation and of the people in a manner that was not entirely expected by
the Framers of the Constitution. Thus, these two branches enjoy both
popular support and the constitutional support that was given them because
the Framers thought they would not have popular support. At the same time,
130. See, e.g., RICHARD F. FENNO, JR., HOME STYLE: HOUSE MEMBERS IN THEIR DISTRICTS
168, 291 (1978) (providing evidence that running against Congress in an attempt to win reelection
to Congress is "ubiquitous" and "foolproof').
131. Jeffrey Tulis, Assoc. Professor of Gov't, Univ. of Tex. at Austin, Constitutional Decay and
the Politics of Deference, Speech at the 10th Annual Walter F. Murphy Lecture in American Ideals
and Institutions (Apr. 4, 2011), available at http://web.princeton.edu/sites/jmadison/calendar/flash/
Tulis.html.
132. Id.
133. See THE FEDERALIST No. 51 (James Madison), supra note 17, at 322-23 (describing the
Executive's position of weakness and need for fortification); THE FEDERALIST No. 78 (Alexander
Hamilton), supra note 17, 465-66 (describing the inherent weakness of the Judiciary and the
protections that would be required to protect its limited power).
134. See Jeffrey M. Jones, Supreme Court Approval Rating Dips to 46%, GALLUP (Oct. 3,
2011), http://www.gallup.com/poll/149906/supreme-court-approval-rating-dips.aspx (reporting that
63% of Americans approve of the judicial branch as a whole-tying the lowest mark in the history
of the poll-but noting that "Americans still have significantly more trust in the judicial branch than
in either the executive or the legislative branch").
2012]
The Madisonian Constitution: Rightly Understood
971
Congress, thought by the Framers to be most dangerous because of its excess
of popular support, maintains approval ratings that would make even the
worst presidents cringe. 135
V.
Conclusion
Posner and Vermeule are correct to the extent that they point to a
profound problem in our constitutional order (although they do not think it is
a problem): The Executive is increasingly "unbound" insofar as Congress has
continuously passed enabling legislation that promotes the Executive's complete freedom. So, legislation like the Authorization for Use of Military
Force or the legislation in the wake of the banking crisis exemplifies
Congress's behavior more generally. Crises only bring out further what has
always been characteristic of Congress's behavior over the last thirty years.
Posner and Vermeule want to suggest, over and against liberal legalism, that
this situation is not as dire as one might think. Although presidents are
unbound by the law, they remain bound by politics.
To some degree, I would agree with Posner and Vermeule in their
description of the current situation. I would also agree with them in their
critique of liberal legalism's hope to reestablish Congress's supposed constitutional preeminence. But where they implicitly cede to liberal legalism
the claim that, constitutionally, Congress should be preeminent, I very much
disagree. The American constitutional order consists of three independent
and equal branches of power. Liberal legalism's characterization of the necessary preeminence of Congress over the President is a mischaracterization
of our constitutional order. And in ceding the constitutional ground to liberal
legalism, Posner and Vermeule fail to see how much more the constitutional
order, properly understood, actually speaks to their political concerns.
Madisonian constitutionalism, far from being pie-in-the-sky idealism, takes
its bearings from the political situation as it actually is. The constitutional
order is constructed based on the expectations about people's actual political
behavior. The law is understood as being critical to that behavior, but also as
unable, on its own, to constitute that behavior. Institutions must structure
political incentives in such a way that what would otherwise be mere
"parchment barriers" become meaningful.
With that in mind, Posner and Vermeule's diagnosis of Congress should
trouble us. While the constitutional order depends upon three institutions
actively engaged in political conflict over the scope of their powers, the
135. See John T. Bennett, Poll: Obama Approval at 43%, Congress at 12%, BLOG BRIEFING
RooM (Sept. 17, 2011, 11:08 AM), http://thehill.com/blogs/blog-briefing-room/news/182143-pollobamas-approval-at-43-congress-at-12 (relaying that only 12% of people surveyed in a New York
Times-CBS News poll approve of Congress, while 43% approve of President Obama); see also
Presidential Approval Ratings: Gallup Historical Statistics and Trends, GALLUP, http://
www.gallup.com/poll 16677/presidential-approval-ratings-gallup-historical-statistics-trends.aspx
(showing that no presidential approval rating since 1945 has been lower than 22%).
972
Texas Law Review
[Vol. 90:943
current government includes a Congress that actively passes off its power to
both the presidency and the courts. Given that the constitutional order insulates those two institutions on the assumption that Congress will be too
aggressive, perhaps we should rethink the constitutional order itself. If the
Founders were constructing a Constitution knowing what we now know
about Congress's passivity, what would it look like? Instead of discouraging
Congress from using its powers as the Constitution does now, what would it
look like if we encouraged Congress into action? For instance, the
impeachment power belongs to Congress, but it was made to look more legal
than it actually is to discourage Congress from using it too often. Should we
rethink that legalism since we know Congress will not use it often enough?
The Goldilocks Executive
THE EXECUTIVE UNBOUND: AFTER THE MADISONIAN REPUBLIC. By Eric A.
Posner & Adrian Vermeule. New York, New York: Oxford University
Press, 2010. 249 pages. $29.95.
Reviewed by Saikrishna B. Prakash* & Michael D.
Ramsey
With all its defects, delays and inconveniences, men have discovered
no technique for long preservingfree government except that the
Executive be under the law, and that the law be made by
parliamentary deliberations. Such institutions may be destined to
pass away. But it is the duty of the Court to be last, not first, to give
them up.
In The Executive Unbound, Eric Posner and Adrian Vermeule claim that
the two institutions Justice Jackson described-the Executive acting "under
the law" and law made by assemblies-have passed away, at least at the federal level. 2 Yet we still have a free government-indeed, a far better
government than the antiquated model to which Jackson mawkishly clung. It
is better because our complex, fast-moving new world requires maximal
executive flexibility of the sort that outdated, standing laws cannot supply. 3
Though we have an executive unconstrained by law, our government is just
as free because the President is now under a better master: the people
themselves.4
We think The Executive Unbound is terrific-thought provoking and
refreshing. It is a challenging must-read for those who believe that the
Executive can or must be reined in by Congress or the courts and for those
who believe that the Executive has improperly usurped the powers of the
other branches, becoming the most roguish branch of government.
Yet we doubt the book's central claim that we live in a post-Madisonian
republic. First, the U.S. Executive is very much bound-by the Constitution,
* David Lurton Massee, Jr. Professor of Law and Sullivan and Cromwell Professor of Law,
University of Virginia School of Law. Thanks to Steve Walt for helpful conversations.
** Professor of Law, University of San Diego School of Law.
1. Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 655 (1952) (Jackson, J.,
concurring).
2. ERIC A. POSNER & ADRIAN VERMEULE, THE EXECUTIVE UNBOUND: AFTER THE
MADISONIAN REPUBLIC 3-5 (2010).
3. Id. at 32-33.
4. Id at 14-15.
974
Texas Law Review
[Vol. 90:973
Congress's laws, and the courts. Though we cannot peer into the many
minds populating the Executive Branch, we do not believe that executive
officials regard themselves as above the law and the courts, answerable only
to the people via elections and polls. The Executive Branch does not act this
way, and most of its actions are consistent with its own sense of what the law
requires and forbids (although, like most actors, it often reads the law to
maximize its discretion). To be sure, the Executive Branch takes advantage
of gaps and ambiguities in the law, as well as its speed, decisiveness, and
access to information, all as The Executive Unbound describes.5 But the
Executive does not systematically disregard orders from Congress or the
courts nor does it usually exercise core powers that the Constitution assigns
elsewhere; the Executive does not impose criminal punishments, spend
money without authorization, or rule by decree.
Second, while we agree with Posner and Vermeule that public opinion
colors Executive Branch decision making, we also believe that the public
favors an executive bound by the law. So long as the public expects the law
to constrain the Executive, the Executive will take into account this expectation and the public's sense of the law, even under Posner and Vermeule's
own light. In other words, the public has a taste for the rule of law, a taste
that the Executive Branch ignores at its peril.
We think the legal constraints on the modem Executive are so manifest
that we wonder whether Posner and Vermeule's real project is more aspirational than descriptive. Perhaps their ultimate objective is to persuade us that
we should have an unbound executive, not that we already have one. We
hedge here because the book seems of two minds. In keeping with the title,
most of the book forcefully argues that the Executive faces no material legal
constraints. For instance, Posner and Vermeule write that "the legally constrained executive is now a historical curiosity,,6 and that the Madisonian
separation of powers has "collapsed."7 There is no equivocation here. Yet
Chapter 6 argues that irrational fear of executive tyranny has prevented the
Executive from obtaining powers needed to handle modem emergencies.
Obviously this complaint assumes that there are constraints on the Executive.
And the conclusion in particular appears to admit that the courts and
Congress check the Executive-that the Executive is bound and that the
Madisonian republic lives on. 9
We are unsure what to make of the concession that the Executive is very
much bound, coming as it does on the heels of chapters explaining why the
5. See, e.g., id. at 34-37 (describing the Bush Administration's counterterrorism policy).
6. Id. at 4.
7. Id at 18.
8. Id. at 202-04.
9. See id. at 206 (noting that Congress "has subjected presidential lawmaking to complex
procedures and bureaucratic checks" and that "the Supreme Court retains certain quasi-lawmaking
powers, which it exercises by striking down statutes and blocking executive actions").
2012]
The Goldilocks Executive
975
Executive is unbound. In any event, if Posner and Vermeule mean to argue
for a new regime of legally unconstrained executive power, we doubt that
they have proved their case. The Executive Unbound persuasively argues
that a highly constrained executive is inconsistent with modem needs and
that implausible fears of executive tyranny improperly color modem policy
debates. But it does little to show that lifting the core constraints that remain
is either safe or necessary.
As developed below, we think that Posner and Vermeule have shown
that the modem Executive is much less bound by law than in the past and
that in general this may be for the good. But they have not shown that the
Executive is unbound by law, or that the Executive should be. Part I
describes the book's central arguments and situates them within the broader
literature on executive power. Part II discusses The Executive Unbound's
first several chapters as a description of the modem Executive and concludes
that Posner and Vermeule substantially overstate the lack of constraint the
Executive faces. Part III considers The Executive Unbound as a normative
argument for adopting a legally unbound executive and finds the case not
proven. We offer the tentative conclusion that separation of powers and
related constraints play an important role in creating something of a
"Goldilocks Executive": an executive neither much too strong nor much too
weak, but about right.
I.
The Executive Unbound and Its Place in the Literature
A. The Book andIts Central Thesis
The Executive Unbound's chief target is "liberal legalism"-the
common belief, in its words, "that representative legislatures govern and
should govern, subject to constitutional constraints, while executive and judicial officials carry out the law."' 0 As the book elaborates, "Liberal legalism
is intensely anxious about executive power, and sometimes goes so far as to
define tightly constrained executive power as an essential element of the rule
of law."" Against this supposedly antiquated and futile view, the authors set
their two central propositions: first, that given modem realities, the Executive
is no longer meaningfully constrained by law; and second, that public opinion rather than law sufficiently checks the Executive.12
The first proposition unfolds in the book's initial three chapters. The
first two chapters argue that the Founders' constitutional system of separation of powers is misconceived and irredeemably outdated. 3 The third
considers (and dismisses) modem attempts to fix or ameliorate the collapse
10. Id. at 3.
11. Id.
12. Id. at 15-16.
13. Id. at 15.
976
Texas Law Review
[Vol. 90:973
of constitutional checks through a series of feckless statutes purportedly limiting executive discretion.' 4
The second proposition is the focus of Chapters 4 and 6.15 Somewhat in
tension with their title, the authors maintain that public opinion strongly
binds the Executive.16 Thus, notwithstanding the breakdown of legal
constraint, the Executive remains bound by political constraints.17 And the
political constraints are effective enough, the authors believe, that fear of
executive overreaching is wildly excessive and has led to "tyrannophobia."18
Within this framework, the book advances four main arguments, two
factual and two normative. The first claim is factual: the modem Executive
makes policy subject to minimal oversight by Congress and the courts.' 9
Because Congress has limited capacity to monitor Executive Branch actions,
Congress has ceded policy-making authority to the Executive via delegation
and inaction.20 Similarly, courts lack the will and expertise to overturn
executive choices. 21 Thus, the "Madisonian" separation of powers no longer
describes how power is constitutionally allocated today, and a series of
statutes-such as the War Powers Resolution (WPR) and the Administrative
Procedure Act (APA)-designed to rein in the Executive have failed, either
because they have generally been ignored (the WPR) or contain exceptions
and ambiguities that allow executive discretion to flourish nonetheless (the
APA).2 2 As the authors summarize,
Legislatures and courts . . . are continually behind the pace of events
in the administrative state; they play an essentially reactive and
marginal role, modifying and occasionally blocking executive policy
initiatives, but rarely taking the lead. And in crises, the executive
governs nearly alone, at least so far as law is concerned.23
The book's second main claim, a normative one, is that releasing the
Executive from the bounds of law is a good and inevitable result.24 The
speed and complexity of modem events require executive discretion.
Congress and the courts are slow and deliberative and hence unable to cope
with our fast-paced, modem world, especially the increasingly frequent
14. Id.
15. Id. The somewhat oddly placed fifth chapter rejects the idea that, in the absence of
domestic limits, international law can provide a legal check on the Executive. Id. at 16. We do not
address that point, as it seems peripheral to the book's central claims.
16. Id. at 15-16.
17. Id. at 15.
18. Id.
at 176.
19. Id. at 25-29, 52-54.
20. Id. at 31-32.
21. Id. at 29-31.
22. See id. at 84-87 (explaining that statutes that are intended to constrain executive power
often fail because they are not enforced or because they are flexible and exclude certain government
actions from review).
23. Id. at 4.
24. Id. at 14-15.
2012]
The Goldilocks Executive
977
crises. 25 The "unbound" Executive is, the authors say, not only descriptively
accurate but normatively optimal: "Executive government is best in the thin
sense that there is no feasible way to improve upon it, under the conditions of
the administrative state."26 The natural conclusion is to reject liberal
legalism's misguided attempts to reform and constrain the modem
Executive.27
The third claim, again factual, is that "the major constraints on the
executive, especially in crises, . . . arise . . . from politics and public
opinion." 2 8 Thus, "[a] central fallacy of liberal legalism .. . is the equation of
a constrained executive with an executive constrained by law." 2 9 To the
contrary, "de facto political constraints ... have grown up and, to some
degree, substituted for legal constraints on the executive" such that "[t]he
executive, 'unbound' from the standpoint of liberal legalism, is in some ways
more constrained than ever before."30 In short, the (legally) unbound
Executive is (despite the book's title) not truly unbound, only bound in a
different way than the Madisonian system envisions.
Fourth, while the book is at times circumspect about the effectiveness of
political constraint-it denies that "political constraints necessarily cause the
executive to pursue the public interest" and says only that "politics and public opinion at least block the most lurid forms of executive abuse"3 1-it
contends that political constraints are sufficient to render fears of executive
tyranny irrational: hence, "tyrannophobia."3 2 Concerns about executive tyranny that permeate liberal legalism are counterproductive, for they prevent
useful delegations to the Executive and promote irrational fears.33 As a
result, we should move beyond reflexive fears of executive government, not
just because there is no alternative to it, but because irrational fears stoked by
liberal legalists may hamper executive action with no countervailing benefit.
In sum, Posner and Vermeule say, an executive unconstrained by law
but constrained by public opinion is the modern reality and the inevitable,
best available solution to modem challenges. Their subtitle could have been,
How We Learned to Stop Worrying and Love the Unbound Executive.34
25. Id. at 25-26, 52-53.
26. Id at 5.
27. Id at 14-15.
28. Id at 4.
29. Id at 5.
30. Id
31. Id.
32. Id at 176-77.
33. Id. at 202-04.
34. With apologies to DR. STRANGELOVE OR: How I LEARNED TO STOP WORRYING AND LOVE
THE BOMB (Columbia Pictures 1964).
978
Texas Law Review
[Vol. 90:973
B. The Executive Unbound's Place Within the Literature
The Executive Unbound occupies an unusual space within the broader
literature on executive power, whose scholarship tends to cluster around two
nodes. The first we might call the "Despotic Executive" thesis, exemplified
in recent books by Bruce Ackerman and by Peter Shane. 35 Ackerman and
Shane generally share Posner and Vermeule's descriptive conclusion that the
Executive is unbound, but they differ from Posner and Vermeule on the
Like The Executive Unbound, the Despotic
normative implications.
Executive thesis regards the modem Executive as unleashed from the bounds
of law as a result of excessive delegation, congressional and judicial
passivity, and relentless executive overreaching. 36 But these developments, it
is said, produce a bad-indeed, an appalling--outcome that threatens
democracy. The challenge for Despotic Executive theorists is to restore legal
constraint upon the Executive-precisely the project that The Executive
Unbound criticizes as hopeless, irrational, and counterproductive.
The Despotic Executive critique echoes Arthur Schlesinger's classic
38
The Imperial Presidency and an array of writings from the Nixon era.
35. See BRUCE ACKERMAN, THE DECLINE AND FALL OF THE AMERICAN REPUBLIC 32 (2010)
(describing the evolution of the President from "an eighteenth-century notable ... to a twenty-firstcentury demagogue, asserting extraconstitutional authority"); PETER M. SHANE, MADISON'S
NIGHTMARE: How EXECUTIVE POWER THREATENS AMERICAN DEMOCRACY 158 (2009) (arguing
that presidential control over "discretionary decision making by administrative agencies suggests a
transformation of the President not only from overseer to decider, but from chief executive to chief
lawmaker" and that such authority "would potentially give the President a single-handed role
dwarfing the role of Congress in prescribing rules that Americans are compelled to obey"). Daryl
Levinson and Richard Pildes offer a more modest view of executive overreach, arguing that the
dangers of executive overreaching arise principally in periods where the branches of government are
all controlled by the same political party and thus the checks of party rivalry-which they think
have largely superseded Madisonian checks-are relaxed. See generally Daryl J. Levinson &
Richard H. Pildes, Separation ofParties,Not Powers, 119 HARV. L. REV. 2311 (2006) (arguing that
the Madisonian system of checks and balances between institutions is largely effective only when
the branches of government are controlled by different and competing political parties and
suggesting reforms that might preserve checks and balances in eras of party-unified government).
36. See, e.g., SHANE, supra note 35, at vii (lamenting the abandonment of checks and balances
for "a virtually unchecked presidency, nurtured too often in its political aggressiveness by a feckless
Congress and obsequious courts").
37. The Executive Unbound says of Shane and Ackerman:
Curiously, these books ... go on to offer a series of prescriptions for reviving (some
version of) liberal legalism and the Madisonian separation of powers. We believe that
the diagnoses of decline are so convincing that the prescriptions for revival are futile;
the very motivations, beliefs, and opportunities that these authors ascribe to political
actors in their diagnoses, if true, rule out their prescriptions.
POSNER & VERMEULE, supra note 2, at 213 n. 1.
38. See ARTHUR M. SCHLESINGER, JR., THE IMPERIAL PRESIDENCY 377 (Mariner Books 2004)
(1973) (declaring that the Nixon Administration's belief "in its own mandate and in its own virtue
... had produced an unprecedented concentration of power in the White House" and arguing that
"[i]f this transformation were carried through, the President, instead of being accountable every day
to Congress and public opinion, would be accountable every four years to the electorate" and that
"[b]etween elections, the President would be accountable only through impeachment and would
govern, as much as he could, by decree").
2012]
The Goldilocks Executive
979
Similar charges were laid against Franklin Roosevelt, Lincoln, Jackson, and
even Washington in their days. Posner and Vermeule are surely right to say
40
that fear of an unconstrained Executive runs deep in the American psyche.
A strong countercurrent in modem discourse on executive power goes
largely unmentioned in The Executive Unbound. This "Fettered Executive"
thesis was associated most strongly in the George W. Bush Administration
with Vice President Dick Cheney and his aide David Addington,41 but it
dates back at least to the Reagan Administration. 42 Its counter-narrative is
that after Watergate, Congress and the courts have tied down the Executive,
moved by a misguided legalism and a mistaken view of the constitutional
meaning of executive power. In this account, Congress forced upon a weakened presidency a series of bad and unconstitutional laws-for example, the
WPR, the Ethics in Government Act, the Presidential Records Act, and many
others. At the same time, courts increasingly interfered in executive
operations (United States v. Nixon,4 3 and more recently, Boumediene v.
Bush 44 ) and allowed Congress to interfere (Morrison v. Olson4 5 ). Former
Vice President Cheney argued: "In the aftermath of Vietnam and Watergate
... there was a concerted effort to place limits and restrictions on presidential
authority.... [These efforts] were misguided ..... '
Like Posner and Vermeule, adherents of the Fettered Executive thesis
believe that "given the world that we live in, . . . the president needs to have
unimpaired executive authority" to meet the demands of modem
governance.47 Tying the Executive down with multitudes of laws, hearings,
39. POSNER & VERMEULE, supra note 2, at 184.
40. See id. at 176, 181-87 (declaring that "[t]yranny looms large in the American political
imagination" and describing how perceptions of tyranny and dictatorship have influenced the
development of American institutions and modem understandings of powerful presidencies).
41. See JACK GOLDSMITH, THE TERROR PRESIDENCY: LAW AND JUDGMENT INSIDE THE BUSH
ADMINISTRATION 84-98 (2007) (describing the Cheney-Addington view); see also JOHN YOO,
WAR BY OTHER MEANS: AN INSIDER'S ACCOUNT OF THE WAR ON TERROR 122 (2006) (claiming
that the War Powers Resolution was an overreach by Congress into the President's constitutional
authority over foreign affairs).
42. See THE FETTERED PRESIDENCY: LEGAL CONSTRAINTS ON THE EXECUTIVE BRANCH
(L. Gordon Crovitz & Jeremy A. Rabkin eds., 1989) (exploring the legal constraints and
institutional obstacles that hampered the Reagan Administration's implementation of its desired
policies); see also Saikrishna Bangalore Prakash, Imperial and Imperiled: The Curious State of the
Executive, 50 WM. & MARY L. REV. 1021, 1025-26 (2008) (describing "the sense amongst some
that Congress has subdued, even shackled the presidency" and citing additional sources).
43. 418 U.S. 683 (1974).
44. 553 U.S. 723 (2008); see A. Raymond Randolph, The Guantinamo Mess, in CONFRONTING
TERROR 241, 242-43 (Dean Reuter & John Yoo eds., 2011) (criticizing Boumediene as setting in
motion "an unprecedented wartime lawmaking exercise [by the courts] with broad implications for
the future").
45. 487 U.S. 654 (1988).
46. James Taranto, A Strong Executive, WALL ST. J., Jan. 28-29, 2006, at A8.
47. Id. (quoting Vice President Cheney). Fettered Executive partisans generally suppose that
fettering the Presidency is unconstitutional because such constraints runs counter to the
Constitution's original conception of a robust and vigorous executive. See generally GOLDSMITH,
980
Texas Law Review
[Vol. 90:973
independent prosecutors, and judicial proceedings robs it of the energy,
secrecy, and dispatch that Hamilton celebrated.48 As a result, the Executive
may be unable to offer decisive responses to complex, fast-moving modern
challenges, especially in emergencies.
As noted, Posner and Vermeule do not discuss the idea that the
Executive is unconstitutionally fettered. Presumably, they would dismiss it
as descriptively wrong. But it is noteworthy that prominent voices have a
strikingly different view of reality, even while agreeing with Posner and
Vermeule about the need for executive discretion.4 9
Thus, The Executive Unbound stakes out a position partially agreeing
with, yet in sharp contrast to, both sides of the conventional debate over
executive power. Like Despotic Executive proponents, Posner and Vermeule
think that, as a descriptive matter, the Executive is unbound by law. Like
Fettered Executive adherents, Posner and Vermeule think an unbound
executive is normatively preferable. Perhaps surprisingly, aside from their
book, this space is largely unoccupied, at least in U.S. scholarship.so With its
forceful argument, detailed-yet-accessible style, and carefully developed
conclusions, The Executive Unbound is destined to be a classic statement of
its position, one that cannot be ignored by future scholars of executive
power.
Each of the three positions-Despotic Executive, Fettered Executive,
and Unbound Executive-have valid points and insights. Yet each seems
overstated and not fully persuasive, either descriptively or normatively.
Further, their alignment suggests three quadrants of a four-quadrant box:
(1) executive power is unconstrained by law and that is bad (Despotic
Executive); (2) executive power is unconstrained by law and that is good
(Unbound Executive); and (3) executive power is constrained by law and that
is bad (Fettered Executive).
supra note 41 (discussing David Addington's views). Posner and Vermeule, in contrast, generally
do not make constitutional claims.
48. THE FEDERALIST No. 70, at 424 (Alexander Hamilton) (Clinton Rossiter ed., 1961).
49. It should be noted that Vice President Cheney believed that the Constitution constrained the
Executive. This, too, separates him from Posner and Vermeule. See Taranto, supra note 46
(quoting Cheney as saying that Cheney's view "doesn't mean, obviously, that there shouldn't be
restraints[;] ... [t]here clearly are with respect to the Constitution, and he's bound by those, as he
should be").
50. Outside of U.S. scholarship, this view is most closely associated with the Weimar and Naziera theorist Carl Schmitt. See generally CARL SCHMITT, CONSTITUTIONAL THEORY (Jeffrey Seitzer
ed. & trans., Duke Univ. Press 2008) (1928); CARL SCHMITT, THE CRISIS OF PARLIAMENTARY
DEMOCRACY (Ellen Kennedy trans., MIT Press 1985) (1923). For a response to Schmitt, see
WILLIAM E. SCHEUERMAN, CARL SCHMITT: THE END OF LAW 61-84 (1999). Posner and Vermeule
rely on Schmitt to a limited extent in developing their criticism of the Madisonian framework.
POSNER & VERMEULE, supra note 2, at 32-54. We do not address Schmitt's arguments, as we
regard them as neither necessary to nor especially supportive of Posner and Vermeule's position.
Schmitt wrote for his time and place, as Posner and Vermeule write for theirs. We think Posner and
Vermeule's claims stand or fall on their own merit and that Schmittian references are largely a
distraction.
The Goldilocks Executive
2012]
981
Missing from this grid is a fourth position-the one we find most
attractive. The modem Executive is constrained by law and that is good. We
call this the "Goldilocks Executive," where the Executive is neither too
strong nor too weak. We are not so confident to believe that the Executive is
"just right."" But it is about right, which is the best one can expect from the
government. Table 1 illustrates the four positions.
Table 1. Views of the Modem Executive
Good
Good
Bad
Executive Is Unbound
Executive Is Bound
Unbound
UbudEeuieExecutive
Goldilocks Executive
(Posner & Vermeule)
Despotic Executive
(Ackerman, Shane)
Fettered Executive
(Cheney, Addington)
We develop this theory of the Goldilocks Executive in subsequent
sections as a counterpoint to The Executive Unbound. As an opening sketch,
consider the following. As Part II discusses, we find the Despotic Executive
and Unbound Executive perspectives unpersuasive. Of course, the Executive
is in many ways much less bound than at the founding. Posner and Vermeule
are surely right that the growth of the military and the bureaucracy, the rise
of delegation and the administrative state, and the pace and complexity of
modem events have greatly altered and enhanced the Executive Branch. But
less bound is not unbound. The Constitution's core separation-of-powers
constraints remain in place, in practice as well as in theory: the Executive
cannot spend money, impose criminal punishments, or make law (outside of
delegated areas) without the cooperation of the other branches. Further,
other branches have gained power in other respects: for example, the postWorld War II constitutional rights revolution empowered judicial oversight
of executive operations beyond anything imagined in the eighteenth century.
Indeed, the very existence of the Fettered Executive thesis and its adherents
in the Executive Branch suggests the implausibility of the idea that the
Executive faces no legal constraints-apparently the Executive Branch itself
does not feel unbound. If Posner and Vermeule are descriptively accurate,
they must not only counter tyrannophobia but also merinthophobia-the fear
of irrational Executive Branch officials who imagine binding, even chafing,
constraints when there are none.
We also find the Fettered Executive thesis overstated. The Executive
retains substantial discretion even as it is subject to oversight, reporting
requirements, and inspectors general. Posner and Vermeule's claims are
implausible only to the extent they try to describe an executive with no
51. See AMANDA GRAHAM & ANNIE WHITE, GOLDILOCKS AND THE THREE BEARS 5, 7, 9
(2008) (describing Goldilocks as finding some porridge, a chair, and a bed as "just right").
982
Texas Law Review
[Vol. 90:973
meaningful legal constraints; as an account of the substantial power and discretion wielded by the modern President, The Executive Unbound seems
entirely persuasive. Again, the existence of two diametrically opposed
descriptive accounts of the modem Executive suggests that both greatly
exaggerate. As a descriptive matter, we would say that the law constrains the
President, in many areas quite a bit and in some areas hardly at all.
Turning to the normative side, we think that Posner and Vermeule
would have excellent arguments if their arguments were deployed to support
the moderately constrained Executive that actually exists as opposed to the
wholly unconstrained Executive they think exists (or wish for). It is true that
an effective executive needs flexibility and discretion to respond to fastpaced events-a point recognized in the founding era and (we believe) built
into our constitutional system to some extent. 52 The Founders had experience with extraordinarily weak executives-the feeble executive-byassembly of the Articles of Confederation and the highly dependent, legislatively dominated governors and executive councils of the early state
constitutions-and had judged them to be failures. A central point of the
Constitution's structure of executive power was to convey more "energy" to
As Posner and Vermeule
the Executive via unity and adequate powers.
recount, modern events have induced Congress to enhance executive energy
to an extent the Founders never envisioned. 54 Congress has delegated its
authority to the Executive, created a massive military and administrative
bureaucracy, and remained passive in the face of executive initiatives. And
the system has proved largely successful: the United States has weathered a
series of crises in significant part thanks to energetic executive leadership;
but despite recurrent refrains of impending executive tyranny-whether
under Washington, Jackson, Lincoln, Franklin Roosevelt, or George W.
Bush-tyranny never comes.
Finally, to say that the Executive requires flexibility and discretion
(even more than the Founders envisioned) is not to say that the Executive
requires flexibility and discretion wholly unlimited by law. The Founders
also recognized the dangers of executive power and appreciated that forming
52. See, e.g., THE FEDERALIST No. 70 (Alexander Hamilton), supra note 48, at 423-31 (arguing
that an energetic executive is necessary to respond to foreign attacks and administer the laws).
53. See MICHAEL D. RAMSEY, THE CONSTITUTION'S TEXT INFOREIGN AFFAIRS 118-19 (2007)
it is, therefore,
("[T]hese developments reflected desire for an executive branch with 'energy' ....
incomplete to say that Americans feared unified executive power. . . . [T]hey also embraced unified
executive power, as a means to get things done.. . . The Constitution's provisions on presidential
power . .. reflect an interplay between these two competing goals.").
54. See POSNER & VERMEULE, supra note 2, at 41-51 (analyzing Congress's role in relation to
the Executive's in the war-on-terror context and in the context of the 2008 financial crisis, and
concluding that "[w]e are left with a picture of congressional (and judicial) passivity in the wake of
the two major crises of the first decade of the twenty-first century[;] . . . [a]fter the crisis is under
way, the executive seeks a massive new delegation of authority and almost always obtains some or
most of what it seeks").
2012]
The Goldilocks Executive
983
an effective executive required balancing energy and constraint.55 Even with
the changes Posner and Vermeule point out, many of the Founders' central
constraints remain in place. We are unsure whether Posner and Vermeule
would prefer these constraints to be lifted-that is, that we should have an
executive who, among other things, can ignore enacted law, impose criminal
punishments, spend money without legislative authority, and rule by decree.
Despite the book's title and several chapters about the unconstrained nature
of the modem Executive, they make no arguments directed specifically
against these familiar constraints. In any event, we doubt such arguments
could be sustained. The Founders recognized an essential truth that modem
events do not undermine: the need for a Goldilocks Executive, where the
President wields significant power but is still constrained by law. Whether
the Founders struck exactly the right balance, either for their time or ours,
can be debated. But the idea that we would be better off with no legal constraints on the Executive strikes us as unproved.
II.
Exaggerating Executive Power
A. The Undeniable Growth ofExecutive Power
We start with areas of accord. We agree that executive power has, in
certain respects, swelled over the course of two centuries. The causes are
both foreign and domestic. In part, the growth is a byproduct of the role that
the United States has played in world events for the past seventy years.
When a nation plays a dominant role on the world stage, its executive necessarily reaps additional power. Of the three branches, only the Executive is
constantly on duty and built to react quickly to explosive foreign events, both
military and diplomatic. The growth of the U.S. military, from a small army
and no navy in 1789 to the massive force deployed worldwide today, reflects
Congress's acceptance of this fact. When Helvidius complained that war
was the "nurse of executive aggrandizement,", 5 6 he might have generalized
his complaint and directed it at the whole realm of external relations.
Increased executive clout is also traceable to the federal government's
unimagined domestic expansion across many fronts. As the federal government has taken on new tasks related to health care, poverty alleviation, and
intrusive regulation of the economy, the Executive has typically been the one
to implement these agendas. Generally speaking, the more federal rules and
programs there are, the more power the Executive wields.
The expansion of federal power has not merely increased the size of the
law-enforcement Executive. As Posner and Vermeule point out, delegation
has increased the Executive Branch's role in crafting federal rules that serve
55. See, e.g., THE FEDERALIST No. 51 (James Madison), supra note 48, at 320-25 (explaining
the balance of powers and checks provided in the new system of government).
56. JAMES MADISON, HELVIDIUS NUMBER IV (1793), reprinted in THE PACIFICUS-HELVIDIUS
DEBATES OF 1793-1794, at 84, 87 (Morton J. Frisch ed., 2007).
984
Texas Law Review
[Vol. 90:973
as substitutes for or complements to congressional legislation.57 In an era of
vast federal lawmaking, many people suppose that Congress cannot legislate
with the specificity necessary, for it lacks the time and expertise. Thus, the
need to delegate legislative authority to others arises. The Executive has
been (and will continue to be) a beneficiary of these increasingly frequent
and broad legislative delegations.
Posner and Vermeule also point to the "Chevronization" of parts of
federal law as a reason for increased executive power.58 The Supreme Court
has concluded that it should defer to executive agencies' reasonable constructions of federal law, either because Congress has required as much or
because the Executive is more democratically legitimate than the courts that
might second-guess its constructions of statutes.59 Whatever the reason,
courts permit the Executive to adopt any one of a number of reasonable
readings of a federal statute, a stance that effectively cedes to the Executive
broad authority over the statute's meaning. Multiply this discretion across
the thousands of statutes and programs, combine it with the express
delegation of rulemaking authority in many areas, and the Executive may
seem unbound, even unhinged.
Posner and Vermeule's account of the rise of executive power seems
generally persuasive. We agree that the President has gained power in ways
that the Founders could not have imagined. But Posner and Vermeule claim
not just that the President is less bound than was imagined at the founding,
but that the President is unbound-that "the legally constrained executive is
now a historical curiosity" 6 0 and that the Madisonian separation of powers
has "collapsed." 6 1 That is a rather different claim, one that is inconsistent
with the realities of modem practice.
As if to recognize the difficulties of their descriptive claim, Posner and
Vermeule belatedly conclude their book with a peculiar admission. In their
conclusion, they seem to concede that the Executive is legally bound:
Congress retains the formal power to make law. It has subjected
presidential lawmaking to complex procedures and bureaucratic
57. See POSNER & VERMEULE, supra note 2, at 32-34 ("[L]egislatures are incapable of
supplying the necessary policy adjustments at the necessary pace.... The result is that in the
administrative state, broad delegations to executive organs will combine lawmaking powers with
administrative powers . . . .").
58. See id. at 52-54 (explaining that courts tend to defer heavily to the Executive for reasons
both pragmatic and political, such as a comparative lack of political legitimacy in times of crisis).
Posner and Vermeule do not specifically use the term Chevronization, but their discussion of the
ways in which federal courts defer to executive power because of perceived issues of democratic
legitimacy aligns with the conclusions set forth in Chevron U.S.A. Inc. v. NRDC, 467 U.S. 837
(1984).
59. See Chevron, 467 U.S. at 866 (explaining that federal judges have a duty to respect the
legitimate policy choices made by an agency when that choice is a reasonable interpretation of a
statutory gap left by Congress).
60. POSNER & VERMEULE, supra note 2, at 4.
61. Id. at 18.
2012]
The Goldilocks Executive
985
checks, and it has created independent agencies over which the
president in theory has limited control. The federal courts can expect
the executive to submit to their orders, and the Supreme Court retains
certain quasi-lawmaking powers, which it exercises by striking down
statutes and blocking executive actions.62
This description comports with reality. Yet it is buried in the
conclusion and stands at odds with the sweeping claims of the introduction,
the first five chapters, and the book's bold title. Perhaps they ultimately
flinched from their descriptive claims. Because we are uncertain what to
make of this concession, if concession it is, we first consider their bold
descriptive claim. Part III then discusses the possibility that they are
primarily making a normative claim rather than a descriptive claim.
B. The Ties that Bind the Executive
In this subpart we turn to constraints on the Executive-constraints that
Posner and Vermeule understate or substantially ignore. The Executive faces
at least five core constraints: the Constitution, the Congress, the courts, the
public, and party politics. Sometimes these constraints mutually reinforce
each other, and other times they act at cross-purposes. The first three of
these constraints are not foreign to the Madisonian system, but are part and
parcel of it. The fourth was not built into the Constitution; rather, the
Constitution presupposed that the public would be a check on all three
branches, including the Executive. The fifth constraint is unanticipated
because the Founders seem not to have envisioned the rise of political
parties. Finally, we note a further phenomenon, although we are unsure
whether it counts as a separate constraint or as good evidence of the existence of the first five: the Executive believes it is constrained by law.
1. The Constitution.-The Constitution explicitly and implicitly limits
the President's authority. Most importantly, the President cannot make law
using his constitutional authority alone. Article I, Section One says that
"[a]ll legislative Powers herein granted shall be vested in a Congress of the
United States." 63 It is true that broad and frequent delegations of lawmaking
power by Congress to executive agencies have undermined the implicit principle that the President is no lawmaker. But there is no universal delegation
of lawmaking authority, and large areas remain nondelegated. Presidents
understand this. They do not believe that they can, in the manner of Draco,
create crimes or alter private rights by diktat. If presidents want to change
law in nondelegated areas, they must go to Congress, and in so doing, they
often surrender substantial initiative. President Barack Obama's 2010 health
care reform is a striking example: it being a nondelegated area, the President
62. Id. at 206 (endnote omitted).
63. U.S. CONST. art. I, § 1.
986
Texas Law Review
[Vol. 90:973
needed Congress to pass a framework law to effect the reforms because he
could not unilaterally impose the individual mandate, cut Medicare spending,
or provide health insurance to those lacking it. Because congressional
approval was necessary, the reform evolved in directions (such as the
elimination of the so-called public option) that the President disfavored.
Similarly, in the fall of 2011, President Obama developed measures
meant to lift the economy and relieve high unemployment. Yet he did not
implement them himself; they took the form of proposed legislation recommended to Congress (as the Constitution envisions in Article II, Section
Three). Though a majority of Senators wanted to bring the bill to a vote,
they did not have enough votes to break a filibuster. Newspaper accounts
concluded that the filibuster "kill[ed]" the President's jobs bill, at least when
considered as a package.64 Although at least one congressman called on
the President to take 'extra-constitutional' action" to respond to the
"emergency," 65 few regarded that as a serious suggestion, and of course
the President did not act on it. Again, in nondelegated areas, Congress is the
lawmaker; this is an important check on the President, one that serious
people do not question.
The Constitution further says that money may only be drawn from the
Treasury through appropriations made by law.66 Recognizing that they lack
power to make law, presidents do not suppose that they can unilaterally
withdraw money from the Treasury and spend it. That is why they submit a
67
And although Congress
proposed budget to Congress, not an actual one.
requires the President to submit a budget, Congress often takes little guidance from the President on budgetary matters (reflected, for example, in the
Senate's unanimous rejection of President Obama's 2011 budget).6 8 The
massive appropriations bills that Congress enacts every year are mostly a
product of congressional priorities. There is no doubt that were he solely
responsible, the President would have different spending priorities.
Another recent example vividly demonstrates that the President's
constitutional authority (or lack of it) matters: the semi-comical fight over the
debt limit. In July 2011, President Obama and the congressional leadership
conducted tough negotiations about the conditions under which Congress
64. E.g., Andrew Taylor & Ben Feller, Senate Republicans Vote to Kill Obama's Jobs Bill,
BLOOMBERG BusINESSWEEK (Oct. 11, 2011), http://www.businessweek.com/ap/financialnews/
D9QACO500.htm.
65. Nicholas Ballasy, Rep. Jackson: Obama Should 'Declarea National Emergency,' Add Jobs
with 'Extra-constitutional'Action, DAtLY CALLER (Oct. 12, 2011), http://dailycaller.com/2011/10/
12/jackson-obama-should-declare-a-national-emergency-add-jobs-with-extra-constitutional-action/.
66. U.S. CONST. art. I, §9.
67. By law, Congress has imposed upon the President the obligation to submit a budget. 31
U.S.C. § 1105(a) (2006). The law makes clear that the President is to make proposed appropriations
for the upcoming fiscal year. Id. § 1105(a)(5).
68. Alexander Bolton, President'sBudget Sinks, 97-0, HILL (May 25, 2011), http://thehill.com/
homenews/senate/163347-senate-votes-unanimously-against-obama-budget.
2012]1
The Goldilocks Executive
987
would raise the debt ceiling.69 This bargaining was meaningful on the widely
held assumption that the President otherwise lacked authority to issue more
debt. Some people argued that the President had independent authority, but
that was because they read the Constitution (Section Four of the Fourteenth
Amendment in particular) to grant such authority to the President. 70 Notably,
Posner and Vermeule, in newspaper commentary, urged the President to raise
We admire
the debt ceiling unilaterally regardless of his legal authority.
their consistency, for having argued for an unbound executive in their book,
they continued the argument in the public square. But their argument fell on
deaf ears, likely because everyone else believed the President was bound by
law. And rather than raising the ceiling unilaterally, the President accepted
what he clearly regarded as a distasteful compromise with Congress.
The recent intervention in Libya might suggest that the President is
unbound by the Constitution in one of the most consequential of areas: war.
This intuition will prove most powerful amongst those who believe (as we
do 72 ) that the Constitution gives the power to initiate conflict to Congress.
Yet here too the President's legal advisors made legal claims about Libya,
asserting that the President had constitutional authority to act.73 Though we
regard such claims as mistaken, what is notable for present purposes is the
attempted legal justification. No one in the Executive Branch seems to have
argued that the Executive could launch attacks on Libya at his discretion,
irrespective of law. On the Posner and Vermeule account, this legal
justification was unnecessary because the President was unbound: the
justification was as gratuitous as a defense that the President's actions
74
satisfied the Prime Directive, a frequent theme of the Star Trek series.
Further, the Executive Branch's justification acknowledged limits,
suggesting that a substantial conflict involving U.S. ground forces might
require congressional authorization. 75 Notably, President Bush did seek
69. Carl Hulse & Jackie Calmes, Boehner and Obama Nearing Deal on Cuts and Taxes, N.Y.
TIMES, July 22, 2011, available at http://www.nytimes.com/2011/07/22/us/politics/22fiscal.html.
70. E.g., Adam Liptak, The 14th Amendment, the Debt Ceiling and a Way Out, N.Y. TIMES,
July 25, 2011, availableat http://www.nytimes.com/2011/07/25/us/politics/25legal.html.
71. Eric A. Posner & Adrian Vermeule, Op-Ed., Obama Should Raise the Debt Ceiling on His
Own, N.Y. TIMES (July 22, 2011), http://www.nytimes.com/2011/07/22/opinion/22posner.html
(arguing that the President should raise the debt ceiling based on his "role as the ultimate guardian
of the constitutional order").
72. See generally Saikrishna Prakash, Unleashing the Dogs of War: What the Constitution
Means by "Declare War," 93 CORNELL L. REv. 45 (2007); Michael D. Ramsey, Textualism and
War Powers, 69 U. CHI. L. REV. 1543 (2002).
73. Authority to Use Military Force in Libya, 35 Op. O.L.C. (2011), available at
www.justice.gov/olc/201 1/authority-rnilitary-use-in-libya.pdf.
74. E.g., Star Trek: Breadand Circuses (NBC television broadcast Mar. 15, 1968).
75. See Authority to Use Military Force in Libya, supra note 73 (suggesting that conflicts
constituting "war" in the constitutional sense need congressional authorization and applying factors
to conclude that Libyan conflict was not such a war).
In a recent commentary on the Libya intervention, Professor Posner argues that because the
Executive must interpret the law in order to enforce it, the Executive is also a lawmaker. See Eric
988
Texas Law Review
[Vol. 90:973
congressional authority for the 2003 Iraq invasion despite some advice that
such authorization was unnecessary.76
In sum, the Constitution establishes a basic framework that limits
executive initiative. As a general matter, the President cannot make law or
otherwise alter private rights, spend money, or impose criminal punishments
without the cooperation of one or more independent branches." This is the
Some modem arrangements, including
core of the Madisonian design.
delegations and assertions of independent presidential power in wartime,
may weaken this framework in particular areas. The scope and force of such
arrangements may be debated. But the basic structure remains operative.
2. Federal Statutes.-We do not believe that the Executive is unbound
from the laws of Congress. Evidence for this proposition comes from
In Congress, the drawn-out legislative
Congress and the President.
passing
bills-presupposes that the entire
process--crafting, debating, and
enterprise leads to meaningful results. To be sure, not every resolve matters;
unfortunately, many do not. But by the same token, no one thinks that
Congress is an expensive debating society. New laws and new delegations
matter because they often cede new discretion to the Executive, discretion
that it would otherwise lack. New laws are also meaningful because what
Posner, Stop Complaining About Harold Koh's Interpretation of the War Powers Act, NEW
REPUBLIC (July 1, 2011), http://www.tnr.com/article/politics/91166/harold-koh-war-powers-johnyoo-libya ("Once again, in the area of foreign relations, the president's authority prevails."). We
see his point, even if we might quibble with his terms. But for our purposes what is interesting is
that Posner does not adopt the view that what happens in Congress is irrelevant because laws do not
matter: "The question now is whether Congress, other U.S. institutions or officers, or the public will
try to force President Obama to back down .... There are just not the votes to compel the president
to withdraw from Libya. . . ." Id. He seems to concede here that Congress might have forced the
President to back down, via appropriations or otherwise.
76. BOB WOODWARD, PLAN OF ATTACK 167 (2004).
77. Theodore Roosevelt claimed the President had authority to do anything not prohibited by
law. See THEODORE ROOSEVELT, AN AUTOBIOGRAPHY 504 (1913) (arguing that "occasionally
great national crises arise which call for immediate and vigorous executive action," that in such
crises the President "is the steward of the people," and that "the proper attitude for him to take is
that he is bound to assume that he has the legal right to do whatever the needs of the people
demand, unless the Constitution or the laws explicitly forbid him to do it"). Importantly, however,
Roosevelt admitted that legal prohibitions constrained the President. He did not claim that the
President was beyond the law.
78. The President is also constrained because the states are independent power centers. It is
true that the states' independent authority has been diluted by the expansion of federal constitutional
power. But many things are still left to the states. Under modem doctrine, Congress decides what
residual authority is left with the states. The President cannot unilaterally command the states, even
in an area of clear federal interest. See Medellin v. Texas, 552 U.S. 491, 528-30 (2008) (holding
that President Bush's determination that the United States would follow an International Court of
Justice decision did not compel Texas to do so); Barclays Bank PLC v. Franchise Tax Bd. of Cal.,
512 U.S. 298, 328-29 (1994) (dismissing arguments that executive actions proscribing a certain
method of state income taxation were relevant because it is Congress, not the President, that is given
the relevant powers under the Constitution).
2012]
The Goldilocks Executive
989
Congress gives, it can take, leaving less authority and discretion for the
Executive.
The President cares about what transpires in Congress because he
recognizes that its laws matter. The Constitution presumes as much because
it arms him with a veto. The veto is not a relic of a bygone era, for it affects
legislative dynamics. Even when presidents do not routinely veto, the possibility influences the kinds of legislation Congress takes up and passes.7 9
That is so because the President signals his intentions early in the legislative
process as a means of avoiding having to veto presented bills.
Administrations routinely issue Statements of Administration Policy
regarding bills that sometimes say that the cabinet heads would recommend a
veto or that the President would exercise a veto.so There would be no reason
for this signaling if bills were necessarily inconsequential because the
President could do whatever he wished anyway. More generally, the veto
itself would be irrelevant in a world where the underlying legislation
imposed no constraint. Debating societies never grant someone a veto to
nullify one side's victory, precisely because little turns on who wins.
Further, consider presidential signing statements, much criticized during
the Bush Administration. Why do the words in those statements cause heartbum in some quarters? Because they purport to define words and phrases
found in statutes in ways that some find objectionable and because the
President claims power to decline to enforce statutes that he believes are
unconstitutional. Such statements only matter against a backdrop where
everyone supposes that the Executive has to follow the law. If the law is
irrelevant to what presidents do, both signing statements and opposition to
them are rather pointless.
Once laws are enacted, presidents routinely act as if Congress produces
meaningful products. For example, the Administrative Procedure Act
imposes considerable procedural hoops that the Executive must jump through
79. For example, even though the Republican "cut, cap and balance" plan had almost no chance
of passing the Democrat-controlled Senate, President Obama still threatened a veto. Obama
Threatens To Veto Republican Budget Plan, REUTERS (July 18, 2011), http://www.reuters.com/
article/2011/07/18/usa-debt-veto-idUSWNA390520110718.
80. See, e.g., OFFICE OF MGMT. & BUDGET, EXEC. OFFICE OF THE PRESIDENT, STATEMENT OF
ADMINISTRATION POLICY: S. 1867-NATIONAL DEFENSE AUTHORIZATION ACT FOR FY 2012
http://www.whitehouse.gov/sites/default/files/omb/legislative/sap/l 12/
available at
(2011),
saps 1867s_20111117.pdf (listing at least eleven separate parts of the bill to which the
Administration objected, stating for example that "[t]he administration strongly objects to the
military custody provision of section 1032, which would appear to mandate military custody for a
certain class of terrorism suspects" and declaring that "[a]ny bill that challenges or constrains the
President's critical authorities to collect intelligence, incapacitate dangerous terrorists, and protect
the Nation would prompt the President's senior advisers to recommend a veto"). Notably, the
provisions of the bill against which the statement aimed the most forceful objections included
proposed limits on the President's detention authority. Id. at 1. These are aspects of the war on
terror in which one might think the President is the most unbound-yet the President appeared
concerned that limits contained in the proposed bill would impose substantial and unacceptable
constraints upon him. Id.
990
Texas Law Review
[Vol. 90:973
before exercising rulemaking discretion. Posner and Vermeule argue that
these statutes have enough flexibility built into them that the Executive can
ultimately achieve its goals. Perhaps so, but the Executive nonetheless
expends time and expense to comply with them, as if they (and the entire
system of laws) mattered.
Similarly, laws create independent agencies and limit the President's
removal authority.
While some scholars believe these laws to be
unconstitutional,' presidents rarely dismiss commissioners within these
agencies and never purport to exercise the authority granted to these
agencies. Indeed, we might say that with respect to law execution, the
President's power has diminished over time, at least in the sense that there
were no independent agency fiefdoms at the founding. 82
This is not to say that presidents never take matters into their own
hands. If they feel there is legal authority for acting on their preferences,
they will take the actions they desire and avoid asking permission from
Congress where they believe none is necessary.83 But when they sense they
lack authority, they do not just argue that the public is the ultimate arbiter of
their actions and that existing laws are irrelevant. We think there are countless episodes of a President refraining from implementing his policies
because he concluded that existing laws did not permit him to do so.
What of those instances where the President seems to act contrary to
law, provoking a hue and cry against his lawlessness? Again, our point is not
that the President never does anything beyond the law. In a world full of
law, where people have reasonable disagreements about what the law means,
the Executive inevitably will take measures that seem to violate the law. Our
point is that when the Executive acts, it almost always believes that its
actions are consistent with its understanding of the law. Whatever others
might think about the Executive, the Executive almost always believes that it
has acted lawfully. The Executive never claims that the laws passed by
Congress are irrelevant in the way that the laws of Mexico or Japan are
irrelevant to the actions of the U.S. Executive.
3. The Courts.-The courts constrain the Executive, both because
courts are necessary to the Executive imposing punishments and because
courts can enforce the Constitution and laws against the Executive. It is true,
as Posner and Vermeule say, that courts often operate ex post and that they
81. See Steven G. Calabresi & Saikrishna B. Prakash, The President's Power to Execute the
Laws, 104 YALE L.J. 541, 581 (1994) (arguing that the Constitution grants executive power to the
President alone and that Congress is given no power to create subordinate entities that exercise
executive power).
82. See Prakash, supra note 42, at 1051 ("Unlike modem statutes, which expressly proclaim
that various entities are 'independent' of the executive branch..., no early statute ever declared
that an officer or department would be 'independent' of the President.").
83. William G. Howell, Unilateral Powers: A Brief Overview, 35 PRESIDENTIAL STUD. Q. 417,
417-18 (2005).
2012]
The Goldilocks Executive
991
may defer to executive determinations, especially in sensitive areas such as
national security. But these qualifications do not render the courts meaningless as a Madisonian constraint.
First, to impose punishment, the Executive must bring a criminal case
before a court. If the court, either via jury or by judge, finds for the
defendant, the Executive does not suppose that it can nonetheless impose
punishment (or even, except in the most extraordinary cases, continue
detention). This is so even if the Executive is certain that the court is
mistaken and that failure to punish will lead to bad results. As a result, the
Executive's ability to impose its policies upon unwilling actors is sharply
limited by the need to secure the cooperation of a constitutionally independent branch, one that many suppose has a built-in dedication to the rule of
law. 84 And one can hardly say, in the ordinary course, that trials and convictions in court are a mere rubber stamp of Executive Branch conclusions.
Second, courts issue injunctions that bar executive action. Although it
is not clear whether the President can be enjoined, the rest of his branch
surely can and thus can be forced to cease actions that judges conclude vioAs a practical matter, while courts
late federal law or the Constitution.
issue such injunctions infrequently, injunctions would be issued more often if
an administration repeatedly ignored the law.
Third, courts' judgments sometimes force the Executive to take action,
such as adhering to a court's reading of a statute in areas related to benefits,
administrative process, and even commission delivery. Though the claim in
Marbury v. Madison87 that courts could issue writs of mandamus to executive
officers was dicta,88 it was subsequently confirmed in Kendall v. United
States ex rel. Stokes, 89 a case where a court ordered one executive officer to
pay another. 90
Finally, there is the extraordinary practice of the Executive enforcing
essentially all judgments. The occasions in which the Executive has refused
to enforce judgments are so few and far between that they are the stuff of
legend. To this day, we do not know whether Andrew Jackson said, "John
84. A limited exception to this generalization is the ability of military commissions to try
alleged enemies in wartime. This process, though somewhat judicial in nature, takes place within
the Executive Branch and thus without interbranch checks. But military commissions are extremely
rare and extremely controversial, highlighting the extent to which, in the ordinary course, the
President relies upon an independent branch for conviction and punishment.
85. Compare Mississippi v. Johnson, 71 U.S. (4 Wall.) 475 (1867) (holding that the President
cannot be enjoined), with United States v. Nixon, 418 U.S. 683 (1974) (concluding that the
President can be subpoenaed).
86. See, e.g., Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 582-89 (1952)
(affirming a preliminary injunction against the Secretary of Commerce for an unconstitutional
seizure of property).
87. 5 U.S. (1 Cranch) 137 (1803).
88. Id. at 169-72.
89. 37 U.S. (12 Pet.) 524 (1838).
90. Id. at 608-09.
992
Texas Law Review
[Vol. 90:973
Marshall has made his decision, now let him enforce it." 9 1 Lincoln's
disobedience of Chief Justice Taney's writ of habeas corpus is so familiar
because it was so singular.
Yet to focus on actual court cases and judgments is to miss the broader
influence of the courts. Judicial review of executive action matters because
the knowledge of such review affects what the Executive will do. Executives
typically do not wish to be sued, meaning that they often will take measures
designed to stave off such suits and avoid actions that raise the risk of
litigation. The ever-present threat that someone will take a case to court and
defeat the Executive acts as a powerful check on executive decision making.
The Executive must take account of law, including law defined as what a
court will likely order.
4. Public Opinion.-As noted, we agree with Posner and Vermeule that
the public checks the Executive. We agree with them so much that we wonder if their title oversells their claim. Even under their view, the Executive is
checked in a meaningful way, just not by law and lawyers. A more appropriate title would have been "The Executive Unbound by Law but Tied Down
by Public Opinion." Although their title is punchier, it comes with a risk of
confusion for those unfamiliar with the book's contents.
Though we agree that the public checks the Executive, we find their
description of public preferences a bit truncated. They write as if the public
only has preferences about substance (guns, butter, civil liberties) and that
the Executive acts inconsistently with these preferences at his peril. This
ignores another important public preference. The public has not only substantive preferences but also a commitment to a system where the Executive
is not at liberty to do exactly what any particular majority coalition wants at
any given time. Moreover, the public believes in the separation of powers
at some basic level and expects that the Executive will act consistent with
standing laws, the Constitution, and judicial orders. In short, the public
expects compliance with the Madisonian Constitution.
If we are right about the public's preferences for an executive under the
law, it supplies a solution and a dilemma for Posner and Vermeule. The
solution comes in the form of rehabilitating their descriptive claim that the
President is not bound by law. They could combine our assertion that the
President acts consistent with the law and our claim that the public has a
preference for the Executive following the law. By doing so, they could
maintain that the President follows the law not because he is bound to do so
but only because he feels constrained to satisfy his actual master, the public.
So, perhaps their best response would be to say that the President acts as if
the law binds him because that is what the public brainwashed by a
Madisonian ethos expects of him. In a sense, the Executive would be playing
91. See JEAN EDWARD SMITH, JOHN MARSHALL: DEFINER OF A NATION 518 (1996) (noting
that although Jackson was reported to have made that statement, he probably did not).
2012]
The Goldilocks Executive
993
the part of a Madisonian executive for public consumption, when in fact he is
an elective Schmittian dictator.
The public's preference for an executive under the law also solves
another problem. Though Posner and Vermeule say the public is the only
check on the Executive, they never say why the President does not merely
take the Gallup Poll and implement the preferences it reveals, at least where
those preferences cohere with his own. Why does the President not just raise
marginal tax rates on millionaires if that is what he and the people want? It
seems that the President clearly should raise those marginal rates if the people had only substantive preferences and they were his only masters. But if
the public also has a preference for an executive under law and that preference generally takes precedence over the public's substantive policy
commitments (more guns, less butter, etc.), we know why the President does
not always act unilaterally whenever he and the public share policy
preferences.
Unfortunately, this theory creates its own dilemma. Should Posner and
Vermeule concede that the Executive follows the law and that the public has
an appetite for an executive constrained by law, it becomes rather difficult
for them to say with certainty why the Executive follows the law. We admit
that we cannot prove that the Executive follows the law because it feels
obliged to adhere to the rule of law, an obligation that does not merely arise
because the public expects it. Yet we would say that if the Executive looks
as if it is bound by law, acts as if it is bound by law, and speaks as if it is
bound by law, we think it safe to presume that it believes itself to be bound
by law and is bound by law, even as it also faces other pressures and
constraints. Those who might espouse the view that the Executive is
insincere when pledging fealty to the rule of law face the onus of establishing
that insincerity.
5. The President'sParty.-Duringthe George W. Bush Administration,
those critical of the President's wartime policies lamented Congress's
unwillingness to check the President. Some went so far as to proclaim that
what really mattered was separation of parties across branches and not the
separation of powers itself.92
The tendency to find flaws in a system increases when the system
produces results that one deems flawed. But if one succumbed to the
temptation to find flaws in this manner, one would find systemic flaws
whenever one disliked the products of the system. Libertarians would find
the process that generated social welfare programs like Social Security and
the Great Society fatally flawed because Congress did not say "no" enough
to Franklin Roosevelt and Lyndon Johnson. And liberals would find the
92. Levinson & Pildes, supra note 35, at 2329-30.
994
Texas Law Review
[Vol. 90:973
system of checks and balances flawed because they abhorred Congress's
failure to check George W. Bush.
We think that the Executive's party is not just an enabler or a silent
partner in the Executive's agenda, but can sometimes thwart presidential
initiatives. The separation of powers reinforces that tendency because the
President, despite his office, is not the sole leader of his party. Party leaders
in Congress (and in the states) have positions and powers independent of the
President and, regardless of party affiliation, will not always go where he
tries to lead.
For example, consider President George W. Bush's initiative to revamp
Social Security through the use of private accounts. Despite securing reelection with a sizable margin and despite a Republican Congress, Bush's
proposal went nowhere. Republicans rushed to the microphones to reject it.
Bush never even got a vote on it in either chamber. 93 The Republican Party
had checked the Republican President, showing that separation of parties is
not necessary across the branches. The Republicans essentially decided they
were unwilling to touch the third rail of American politics, notwithstanding
the fact that it was a signature issue for Bush, one that he ran on in 2004.94
Something similar happened in 1993, when Democrats, who controlled both
chambers of Congress, balked at President Clinton's health care plan. 95 To
paraphrase Madison, ambition had countered ambition. 96
Our point is not that parties are a reliable check on the Executive, only
that the President cannot get too far out in front of his party. When he does,
he finds himself isolated, with little chance of a legislative success. Though
politicians typically feel an allegiance to a President of the same party, they
rarely will fall on their sword to further his agenda. And their refusal to do
so frustrates the President's policies, because the President (for separationof-powers reasons) often cannot achieve his goals unilaterally.
6. The Executive's Perception of Legal Constraint.-A final feature of
modem practice that is inconsistent with Posner and Vermeule's description
is that the Executive Branch feels constrained by law. In part this can be
seen from the way it behaves. As discussed above, the Executive Branch
asks Congress to enact legislation and make appropriations rather than doing
93. See Richard W. Stevenson, Many Hurdlesfor Bush Plan, N.Y. TIMES, Mar. 2, 2005, at Al
(discussing political resistance to the proposal).
94. See STEVEN E. SCHIER, PANORAMA OF A PRESIDENCY: How GEORGE W. BUSH ACQUIRED
AND SPENT His POLITICAL CAPITAL 44 (2009) (describing how Bush spoke of creating private
accounts at the Republican nominating convention).
95. See, e.g., Michael Kramer & Janice Castro, The Political Interest: Pat Moynihan's
Healthy Gripe, TIME, Jan. 31, 1994, available at http://www.time.com/time/magazine/article/
0,9171,980052,00.html (quoting influential Democratic Senator Daniel Patrick Moynihan as saying
"anyone who thinks [the Clinton plan] can work in the real world as presently written isn't living in
it").
96. See THE FEDERALIST No. 51 (James Madison), supra note 48, at 322 ("Ambition must be
made to counteract ambition.").
2012]
The Goldilocks Executive
995
so independently. The Executive Branch brings alleged wrongdoers before
courts for punishment rather than punishing independently. The Executive
Branch obeys court orders to act or refrain from acting, as implicitly required
by the Constitution.
But also of significance is the Executive Branch's internal recognition
of legal constraints. The President employs an enormous and growing staff
of lawyers spread among all executive offices and agencies. Anecdotal evidence suggests that legal determinations made within the Executive Branch
have the effect of constraining Executive Branch action. In one particular
episode in the Bush Administration, the Office of Legal Counsel (OLC)
reportedly refused to approve the legality of a surveillance program strongly
favored by the White House, culminating in a showdown in the Attorney
General's hospital room. 9 7 Apparently, when the Attorney General backed
the OLC conclusions, the President acquiesced. More generally, Trevor
Morrison argues that OLC legal conclusions are reached with a sense of
independence from presidential policy preferences and have substantially
influenced presidential decision making. 98 To be sure, Executive Branch
lawyers often may seek to justify presidential actions under law. They may
identify with the Executive and hope to expand its legal discretion. But that
role in itself undermines Posner and Vermeule's claims, for if the President is
truly unbound by law, why expend resources dealing with the law's nonexistent bounds?
We accept that the President's lawyers search for legal arguments to
justify presidential action, that they find the President's policy preferences
legal more often than they do not, and that the President sometimes disregards their conclusions. But the close attention the Executive pays to legal
constraints suggests that the President (who, after all, is in a good position to
know) believes himself constrained by law. Perhaps Posner and Vermeule
believe that the President is mistaken. But we think, to the contrary, it represents the President's recognition of the various constraints we have listed and
his appreciation that attempting to operate outside the bounds of law would
trigger censure from Congress, courts, and the public.
C. The Executive Unbound's Discussionof a Bound Executive
The idea that the President is generally unbound by law is mistaken. A
more plausible view is that the President is relatively unbound in particular
97. See Andrew Zajac, Ashcroft Ex-aide Details Wiretap Infighting, CHI. TRIB. (May 16, 2007),
http://articles.chicagotribune.com/2007-05-16/news/0705151141lgonzales-and-then-white-genjohn-ashcroft-alberto-gonzales (detailing the confrontation between Attorney General Ashcroft and
top Justice Department officials as Ashcroft lay ill in a Washington hospital in March 2004).
98. Trevor W. Morrison, Constitutional Alarmism, 124 HARV. L. REV. 1688, 1708 (2011)
(book review); see also GOLDSMITH, supra note 41, at 90 (describing the "daily clash inside the
Bush administration between fear of another attack, which drives officials into doing whatever they
can to prevent it, and the countervailing fear of violating the law, which checks their urge toward
prevention").
996
Texas Law Review
[Vol. 90:973
areas, such as military and national security matters. Presidents have made
something like this claim, and in arguing for the Unbound Executive, Posner
and Vermeule draw examples not from the ordinary times but from
emergencies, specifically the war on terror and the 2008 financial crisis.
Even if these episodes are valid examples of an unbound executive, they
do not prove the more general collapse of the Madisonian system. At most
they are aberrations from a pattern. 99 But in any event, we think Posner and
Vermeule's examples actually prove the opposite, illustrating constraints
upon the Executive.
Posner and Vermeule argue, for example, that the Supreme Court's waron-terror cases have not meaningfully checked the Executive because the
courts have never finally ordered anyone to be released. 0 0 Yet that hardly
makes their case, for the question is what the Executive would have done in a
world with no judicial review of the claims of alleged enemy combatants.
The Executive's goal has been to thwart attacks in the United States and to
blunt attacks on U.S. forces overseas. Given this goal, the Executive's
incentive is to detain anyone (especially foreigners) who might plausibly
appear to be involved in terrorism against the United States. Members of
al Qaeda and the Taliban generally cannot vote in the United States, and the
costs visited upon them will not enter the Executive's calculus. Despite this
incentive to detain the enemy indefinitely, the Executive released many suspected enemy combatants once held in Guantanamo, some of whom have
taken up arms against the United States.' 0 ' Why did the Bush Administration
release detainees knowing that it was likely that some would take up arms
against the United States? We surmise it was done to stave off judicial
release orders. There is little doubt that the prospect of judicial review and
release had an in terrorem effect on the Executive Branch.10 2
Posner and Vermeule also invoke Congress's 2001 Authorization to Use
Military Force (AUMF) against al Qaeda as an example of a broad delegation proving that the Executive is unbound. 0 3 Yet they admit that the
Executive did not get the sweeping authorization it sought.104 Congress
modified the Administration's proposal in various respects meant to reduce
the discretion ceded to the Executive.
99. To be clear, we do not deny the existence of aberrational contexts in which the Executive
operates to some extent beyond the bounds of law.
100. POSNER & VERMEJLE, supra note 2, at 34-37.
101. See John Mintz, Released Detainees Rejoining The Fight, WASH. POST, Oct. 22, 2004,
available at http://www.washingtonpost.com/wp-dyn/articles/A52670-20040ct21.html ("At least
10 detainees released from Guantanamo Bay prison after U.S. officials concluded they posed little
threat have been recaptured or killed fighting U.S. or coalition forces in Pakistan and Afghanistan,
according to Pentagon officials.").
102. See GOLDSMITH, supra note 41, at 123-25 (discussing the implications of Supreme Court
threats to review one of the Bush Administration's terrorism policies).
103. POSNER & VERMEULE, supra note 2, at 34-35.
104. Id. at 35.
2012]
The Goldilocks Executive
997
The Troubled Asset Relief Program (TARP) bill sent by Treasury
Secretary Henry Paulson to Congress in 2008 similarly received a
congressional rewrite designed to put checks in place.105 What started off as
a bill slightly bigger than a blank check ended up a law hundreds of pages
long, full of limits and deadlines. 06 In the interim, the bill was defeated,
only to be resurrected in a different form after the stock markets dived. 07
Posner and Vermeule might respond that even with the added restraints
the Executive still got blank checks from Congress. We disagree. But even
if one assumed that these statutes gave the President carte blanche, it begs the
question why the Executive went to Congress in the first place, for on their
descriptive theory, he did not need to go to Congressat all, either for the war
on terror or for TARP. While some in the Administration thought congressional authorization for the wars on al Qaeda and Saddam Hussein was
unnecessary, no one argued that the President could bail out financial institutions without first obtaining appropriations.
More generally, the existence of multiple delegations across dozens of
substantive areas implies something left nondelegated. If the Executive had
all the authority it needed and was unconstrained by law, Congress would
have had nothing to do in its various sessions but pass resolutions creating
National Asbestos Awareness Week. Yet every year Congress passes lawsoften quite meaningful ones-related to, among other things, the
environment, employment discrimination, and the military.
Put another way, Posner and Vermeule are right that delegations may
unbind the Executive in particular areas and in particular ways. Yet because
there is no universal delegation of all lawmaking authority, these delegations
hardly show that the Executive is unbound. Rather, they demonstrate that the
Executive is at least legally bound in areas that lack a delegation in the sense
that when neither the Constitution nor statutes delegate authority, the
Executive has none. And there are many nondelegated areas.
The authors' claim about delegations is not the only instance in which
arguments meant to show unboundedness inadvertently prove the opposite.
For instance, they fret about tyrannophobia and what Congress might do in
response. 08 But if law does not matter and cannot constrain the Executive,
105. Id. at 39.
106. Compare Treasury's Bailout Proposal, CNNMONEY.COM
(Sept. 20, 2008), http://
(a 3-page bill), with
money.cnn.com/2008/09/20/news/economy/treasuryproposallindex.htm
Emergency Economic Stabilization Act of 2008, Pub. L. No. 110-343, 122 Stat. 3765 (a 169-page
act).
107. See David M. Herszenhom, Bush Signs Bill, N.Y. TIMES, Oct. 4, 2008, availableat http://
(reporting that the initially
www.nytimes.com/2008/10/04/business/economy/04bailout.html
rejected TARP bill was passed by a higher-than-expected majority in the House of Representatives
due to "fears of a global economic meltdown" and "old-fashioned political inducements added by
the Senate").
108. See POSNER & VERMEULE, supra note 2, at 196-98 (considering the effects of
tyrannophobia and the possibility that it might affect the risk of dictatorship).
998
Texas Law Review
[Vol. 90:973
Congress really cannot do anything of consequence whether or not tyrannophobia exists. If Posner and Vermeule truly believe that law does not
constrain the Executive, maybe tyrannophobia should matter insofar as it
affects the public's tolerance for rule by the Executive and thereby leads the
Executive to make decisions that are suboptimal in order to avoid stoking
tyrannophobic fears. But even here, it is not obvious how tyrannophobia
would matter in a President's second term. Not having to face the electorate
again, a President in his second term should not care about the public's
tyrannophobia, at least under their argument. We think the President cares
because he still has a policy agenda that requires the cooperation and assistance of Congress, an agenda that will be crippled by the sense that he is (or
might become) a tyrant.
Posner and Vermeule's claim that law does not bind the Executive is
inconsistent with how our system actually works. The Executive rarely acts
as if it is above the law or has power to do whatever it thinks best. At most,
the Executive only claims to be unbound in particular areas (such as military
operations) and, even so, employs an army of lawyers to comply with the
laws of war and the laws of Congress. To be sure, the Executive is routinely
criticized for violating the law. But to say that someone has violated the law
hardly establishes that they have no regard for it. Whenever someone claims
that the Executive has transgressed the law, the Executive replies that it is
acting consistent with the true sense of the law, not that it imagines that it
stands above the law. And when courts conclude the Executive has violated
the law, the Executive invariably complies with their judgments. This compliance itself is a recognition that the President operates under the
Madisonian framework where Congress creates ex ante statutory constraints
and the courts enforce those restraints (and constitutional ones) ex post.
While we agree with Posner and Vermeule that the public checks the
Executive, we also believe that the public has a preference for an executive
under law. When one combines this preference with a tradition and a constitutional system that presupposes an executive under the law and presidents
generally desirous of complying with that tradition and system, one has an
executive very much bound by law.
III. The Continuing Relevance of Separation of Powers
A. The Executive Unbound as Descriptive or Aspirational?
The Executive Unbound purports to be a description of the modem
Executive and an antidote for antediluvian law professors who mistakenly
believe that reality mirrors the Madisonian and Montesquieuian framework
that animated the Constitution's framers. Yet as discussed in the prior part,
the book's portrait of reality seems mistaken. The Executive is less bound
than it was 200 years ago; it is not unbound. The Madisonian system has
been diluted from its initial design by delegations and emergencies. But its
2012]
The Goldilocks Executive
999
basic structure remains-a framework that checks the Executive in numerous
ways. Posner and Vermeule's purported description of reality seems so
mistaken that it forces us to consider the possibility that the book is at least in
part meant to be aspirational rather than descriptive.
At times, we think the authors mean to say that the world would be a far
better place if the Executive faced only the considerable constraints that
come from public opinion and not the checks emanating from the
Constitution or federal laws. And we think they might prefer a world in
which the public ignores the legality of executive actions and instead focuses
on whether those actions generate good results.
This perspective is especially notable in Chapter 6, which seems
difficult to reconcile with the rest of the book. Chapter 6's central theme is
"tyrannophobia"-the irrational fear of executive dictatorship. 09 The main
problem with tyrannophobia, according to the authors, is that it prevents the
Executive from being given important powers necessary to deal with modem
emergencies.o10 Indeed, Chapter 6 should be read in conjunction with Posner
and Vermeule's prior work, Terror in the Balance, which, among other
things, objects to statutory and constitutional restrictions on the President's
conduct of the war on terror.' 11 As Posner and Vermeule explain,
"Americans . . . overestimate the risk of executive power and hence recoil
against even reasonable moves toward greater executive authority. . . .
Americans . . . are reluctant to support legislative and constitutional changes
112 Or as they say even more
that could increase executive power . ....
directly (and with even more evident frustration) later, "Elsewhere, we have
described a range of institutions and policy initiatives that would increase
welfare by increasing executive power, . . . but that are blocked by
'libertarian panics' and tyrannophobia."ll 3 And as noted earlier, the book's
conclusion also identifies constraints on the Executive, often in a lamenting
tone.
But these discussions are sharply inconsistent with the claim advanced
in early chapters that the Executive is already legally unbound. We do not
know which argument the authors really favor. We suspect that, given the
implausibility of the legally unbound Executive, the authors' most heartfelt
argument rests on the normative or aspirational claim. Perhaps what is most
important to them is not what is, but what should be.
109. Id. at 176.
110. Id. at 196-98.
111. ERIC A. POSNER & ADRIAN VERMEULE, TERROR IN THE BALANCE: SECURITY, LIBERTY,
AND THE COURTS (2007); see also Adrian Vermeule, LibertarianPanics, 36 RUTGERS L.J. 871, 872
(2005) (criticizing the "widespread and thoroughly irrational, even hysterical, reaction to small legal
changes adopted after 9/11").
112. POSNER & VERMEULE, supra note 2, at 195.
113. Id. at 203; see id at 242 n.93 (citing Terror in the Balance "[fjor examples of 'tangible
security harms' resulting from civil-libertarian rules that constrain executive power").
1000
Texas Law Review
[Vol. 90:973
What they seem to desire is an elected dictator who may do as he sees
fit for as long as his term lasts-what they call the "plebiscitary
presidency." 1 l 4 The dictator is benign because his ability to go rogue is
constrained by his need to keep the public on his side. Public support is necessary for the Executive's reelection, for his party's electoral prospects, and
for the success of his policy agenda.
Undoubtedly, Posner and Vermeule's preferred system has features that
some would find attractive. An executive unbound by law might be more
responsive to the public because there would not be multiple institutions that
collectively thwart or impede majority sentiment. The system maximizes
flexibility, minimizes response time, and concentrates responsibilityfeatures often needed in a rapidly changing world. Their regime is
unconcerned with constitutional checks, many of which may have outlived
their usefulness. Moreover, an elective dictatorship is not without precedent.
England has been described as an elective dictatorship"'5 because Parliament
can (or could until recently) act without limit. And of course, ancient Rome
had elected dictators. Even the United States has had elective dictators,
when some state legislatures made temporary dictators of their executives
during the Revolutionary War.116
In sum, the most significant contribution of The Executive Unbound
may not be its description of the modern Executive (which we think is
overstated, perhaps for effect) but its invitation to abandon existing checks in
the interest of more efficient government. Hence, even if we are right
descriptively, if Posner and Vermeule are right about the prescription, then
the book carries an important message: we should dismantle the checks that
exist by (for example) letting the President appropriate money, punish
suspects, dispense with enacted law, and rule by decree.
Posner and Vermeule's normative account has two components. First,
checks on the Executive are counterproductive because they cause the
Executive to pursue suboptimal policies. Second, checks on the Executive
yield no benefits because, under modern circumstances, executive tyranny is
chimerical. As a result, under any rational cost-benefit analysis, the
Unbound Executive is superior. Below we analyze these claims separately.
Our conclusion is that there is little evidence of the first proposition, so their
argument largely stands or falls on the second. And here, although Posner
and Vermeule make important claims, we find them unpersuasive.
114. Id. at 206.
115. Lord Hailsham, Elective Dictatorship, LISTENER, Oct. 21, 1976, at 496, reprinted in
THREE POLITICAL SYSTEMS: A READER IN BRITISH, SOVIET AND AMERICAN POLITICS 70 (Martin
Burch et al. eds., 1985).
116. See MARGARET BURNHAM MACMILLAN, THE WAR GOVERNORS IN THE AMERICAN
REVOLUTION 75-76, 91 (1943) (noting that South Carolina and Pennsylvania granted extremely
broad powers to their executives).
2012]
The Goldilocks Executive
1001
B. Separation ofPowers and the Goldilocks Executive
We do not claim to know the optimal level of legal constraints on the
Executive, nor do we fully defend the precise set of legal constraints that
make up the modem system. But we think the Goldilocks Executive functions reasonably well within the Madisonian framework and that Posner and
Vermeule do not show otherwise.
To restate, by Madisonian framework we mean generally (1) a
Constitution with checks on all three branches, including the Executive; (2) a
legislature with responsibility for lawmaking, taxing, and appropriations;
(3) a judiciary with power to check executive actions to ensure the Executive
honors the constitutional and statutory bargains related to criminal
punishment, appropriations, and individual rights; and (4) an executive
responsive to enacted law and court orders and who generally lacks powers
allocated to the courts and legislature. From the outside, the public checks
all three branches. Despite the adjective, we do not mean a system that looks
exactly as Madison imagined, just a system that approximates what he (and
others) had in mind.
We think our federal government has been, over the course of two-plus
centuries, generally Madisonian, including periods of relatively weak
presidents, such as the mid-nineteenth century, and periods of relatively
strong executive power during the Civil War and the modem era. Under this
Madisonian system, the President enjoys substantial independent power,
often receives open-ended and wide-ranging delegations from Congress, and
benefits from considerable judicial deference on some matters. 17 For our
purposes, the key characteristic is a relatively robust executive materially
constrained by law.
The Executive so constructed and bounded reflects the enduring tension
noted by the Framers: the need for a strong (but not too strong) executive.
An executive too strong might lead to a Caesar. An executive too weak
might lead to disintegration and a failed state, as almost occurred because of
the weak plural executive (the Continental Congress) established by the
Articles of the Confederation. Leading Framers knew the history of previous
republican experiments and saw recurring patterns. Thus, their creation of a
robust executive nested within a system of separated powers was not an accident or a matter of pure theory.
Given this preference, the principal aim of the Constitution was not
(contrary to what is often said) to tie down the Executive. The design was
more complicated: to empower the Executive-certainly as compared to the
ineffective state executives and the plural, deliberative, and often absent
executive authority of the Articles-while retaining familiar and tested
117. We do not mean to suggest that Madison (or the other Founders) would have approved of
every feature of this system, especially the modem trend. Moreover, even as we believe that the
modem Goldilocks Executive is about right, we do not endorse all of its features.
1002
Texas Law Review
[Vol. 90:973
limits. It was to create, as it were, a Goldilocks Executive. We agree with
Posner and Vermeule that it is hard to say whether the Framers chose the
optimal balance for their own time, much less for ours. The key point,
however, is that the Framers' design reflected an intermediate ground, one
reflecting two competing objectives. While it is impossible to identify precisely the ideal balance between these objectives, an intermediate position of
some sort (which is what we have) seems preferable to either extreme.
A second key point is that, in spite of complaints, the Framers' system
of separated powers has withstood the test of time, allowing strong-willed
presidents to respond to great emergencies without a slide into permanent
dictatorship or widespread loss of freedom (despite repeated alarmism on the
latter point). It is true, as Posner and Vermeule note, that our greatest presidents have been those who used the power of their offices to respond to
crises-Washington, Jackson, Lincoln, Roosevelt, etc. But for the most part,
these presidents achieved greatness by managing the crises within the rule of
law, not by overthrowing it. With some well-known exceptions (well-known
precisely because they were unusual and later condemned), these presidents
did not become plebiscitary dictators, and when the crises receded, executive
power receded as well.' 18 Despite complaints from both the Ackermans and
the Cheneys, this pattern remains true today. George W. Bush successfully
managed the war on terror (despite concerns that he might have been too
bound) without lasting and atypical executive overreaching (despite concerns
that he might have been too unfettered). Much of what President Obama has
done-continuing Bush policies across a range of issues-is what one would
expect of any wartime President operating under the Madisonian framework.
Against this argument from history, Posner and Vermeule counter with
changed circumstances. Modern conditions, they contend, have softened or
eliminated the tension the Framers perceived.1 19 Fettered executive power is
now more of a threat, and legally unbound executive power now less of a
problem, than in times past.
As noted, this claim has two components. The first is that legal limits
on executive power are much more problematic in the modern age. Because
of the speed and complexity of events, only the Executive has the capacity to
respond to them. Anything that materially slows the Executive or limits its
range of action threatens a suboptimal national response. Neither Congress
nor courts have the institutional capacity to understand the exigencies, and
hence, their interference would necessarily be counterproductive.' 20
118. Indeed, Professor Vermeule has spoken of "libertarian panics" and the fear that once the
government acquires powers or invades rights in times of crisis, it never cedes them back. See
Vermeule, supra note 111, at 871.
119. See POSNER & VERMEULE, supra note 2, at 62 (arguing that in today's administrative
state, "the rate of change in the policy environment is high and plausibly accelerating over time,
making constitutional rules set down in past generations particularly suspect" (endnote omitted)).
120. Id. at 25-29, 52-54.
2012]
The Goldilocks Executive
1003
Yet on this point The Executive Unbound provides surprisingly little
support. The authors do not offer examples of limited executive power
leading to bad results in the modem age. They do not point to episodes in
which the Executive refrained from some useful action because of the
Madisonian system. Of course, they are in a bit of a bind here, because they
claim that as a matter of fact the Executive is not really bound; thus, offering
examples of the bad effects of limited executive power would advance their
prescriptive claims only at the cost of their descriptive ones.
Because they do not support their normative claim about the desirability
of an executive unbound and because we think well of our Madisonian
system, we offer some comments on its recent functioning. As discussed, we
think President Bush, during the war on terror, operated within a system of
bounded discretion, facing material but not crushing constraints. He sought
approval from Congress for the attacks against al Qaeda and the Taliban in
the AUMF and in repeated spending authorizations. He also sought approval
for military action in Iraq, though initially he hesitated to do so. He had to
defend his policies in court; he lost some cases and would have lost others
had he pursued his power to its practical limits. He acted knowing that the
courts might second-guess his decisions, a knowledge that affected his decision making. He faced the possibility that Congress would limit his authority
on various dimensions, and on occasion Congress did. Yet he made a robust,
we would say largely effective, response to the challenge of terrorism, at
least if we use the substantial disruption of terrorist networks and the absence
of a subsequent substantial terror attack on the United States as a yardstick.
Facing similar constraints and incentives, President Obama also has acquitted
himself well in the war against al Qaeda.
To be clear, we concede that substantial, even colossal, mistakes were
made in the war on terror. But the leading ones do not seem to have resulted
from legal constraints upon executive power. For example, after initial success in Afghanistan in overthrowing Taliban rule, the President failed to
consolidate the position of the new Afghan government, allowing the Taliban
to regroup as a powerful opposition force. Instead, the President opened a
new front in Iraq. Again, after initial success in displacing Saddam Hussein,
the Administration failed to deploy enough troops to secure the country (at
least until the "surge" in 2007). By some accounts, the distraction of Iraq
harmed the fight against al Qaeda, including by delaying the targeting of
Osama bin Laden.121 Whether these are fairly counted as presidential
121. See RICHARD A. CLARKE, AGAINST ALL ENEMIES: INSIDE AMERICA'S WAR ON TERROR,
at ix (2004) ("[T]he administration .. . squandered the opportunity to eliminate al Qaeda and instead
strengthened our enemies by going off on a completely unnecessary tangent, the invasion of Iraq.").
1004
Texas Law Review
[Vol. 90:973
failures, they are not ones attributable to legal constraints on the Executive,
meaning that they could have occurred even under an unbound executive.122
Perhaps President Bush would have accomplished more had he not
faced a Madisonian system. And perhaps he could have acted with even
greater secrecy, dispatch, and success had not Congress and the courts been
looking over his shoulder. 12 3 We are not in a position to evaluate such
counterfactuals at a detailed level. But we think there is little case to be
made for very substantial costs attributable to executive constraint, even in
the context of a crisis such as the war on terror. Of course, one may speculate that in future crises, or future aspects of the continuing war on terror,
executive constraint may prove highly detrimental. But we see this as no
more than speculation.
Because Posner and Vermeule's arguments about the advantages of an
unbound executive are speculative, we spend more time on their second
proposition-that there is little danger of executive tyranny. Indeed, this
point--developed in Chapter 6-becomes, we believe, the book's animating
and most interesting claim. According to Chapter 6, there is no possibility
under current circumstances that the Executive will become a tyrant (hence,
tyrannophobia, or irrational fear of tyrants). 124 If that is true, then it is silly to
worry about imposing checks on the Executive to stave off a phantom
menace. The checks are worthless, at best. Moreover, it is unnecessary to
develop strong evidence that checks on the Executive yield bad results. If
the checks serve no purpose, even plausible speculation that they could have
bad effects should cause us to abandon them.
Here is the core of their argument:
The best explanation for the lack of dictatorship in America-at
least in America today, as opposed to the nineteenth century-is
neither psychological nor institutional, but demographic. Comparative
evidence suggests that wealth is the best safeguard for democracy.
Equality, homogeneity, and education matter as well. How does the
United States, circa 2011, fare on these dimensions? Ethnic, religious,
and linguistic homogeneity have declined, and inequality has risen,
but because of its high performance on other margins, there is little
cause for concern about American democracy. The United States has
an enormously rich, relatively well-educated population and multiple
overlapping cleavages of class, race, religion, and geography. Simply
by virtue of its high per capita income, the likelihood of dictatorship in
the United States is very low, at least if the historical pattern reflects
causation.
122. One possible counterexample is the President's decision not to bomb Iranian nuclear
facilities in 2007, which may be attributable, at least in part, to perceptions that the legal basis for
such an attack was subject to substantial debate.
123. See supra note 48 and accompanying text.
124. POSNER & VERMEULE, supra note 2, at 176-77.
2012]
1005
The Goldilocks Executive
The modem economy, whose complexity creates the demand for
administrative governance, also creates wealth, leisure, education, and
broad political information, all of which strengthen democracy and
make a collapse into authoritarian rule nearly impossible.
. .
. The
modem presidency is a fishbowl, in large part because the costs of
acquiring political information have fallen steadily in the modern
economy, and because a wealthy, educated, and leisured population
has the time to monitor presidential action and takes an interest in
doing so.
In support, The Executive Unbound marshals statistics to show,
principally, that national wealth correlates with support for democracy. As
the authors explain, "Probably the most robust result of cross-country
empirical work on dictatorship is that the best safeguard for democracy is
wealth. No democracy has collapsed in a nation whose average per capita
income was greater than a little over $6,000 in 1995 dollars."1 26 Since the
United States is well above this amount, they conclude that all is well. 127
We take less comfort in such statistics. The cross-cultural study
covered a relatively short period of history and a small number of
countries. 128 Moreover, we presume that most of the countries with per
capita GDPs over $6,000 have some form of a separated-powers system or at
least are not plebiscitary dictatorships. To make their point, Posner and
Vermeule would need to show that truly unbound executives rarely become
tyrants, not that legally constrained executives rarely do.
Relatedly, Posner and Vermeule argue, very plausibly, that there is little
chance of executive tyranny in the United States: "The high-water mark of
the modern presidency's approach to domestic dictatorship-Nixon's 'thirdrate burglary' of the offices of his political opponents-was pathetic stuff in
historical and comparative perspective, and immediately put Nixon on the
path to disgrace." 2 9 This is all true, of course, but Nixon operated in a
separated-powers system that was in part responsible for his downfall: an
independent Court that ordered release of the tapes; an independent Congress
that demanded an investigation and wielded the threat of impeachment; and
ultimately a public that flatly refused to accept Nixon's view (and it seems
also Posner and Vermeule's preference) that when the President does
something, that means it is not illegal-that is, a public preference for an
executive constrained by law.
125. Id. at 200-02.
126. Id. at 189.
127. See id at 189-91 (concluding that if the relationship between national wealth and support
for democracy is causal, the United States is "unlikely to become a dictatorship in the foreseeable
future").
128. See id. at 190 tbl.6.1 (reflecting data covering twenty-two countries over a period spanning
from 1950 to 2007).
129. Id. at 200-01.
1006
Texas Law Review
[Vol. 90:973
We agree that the chances of slipping into tyranny are low, but that does
not prove Posner and Vermeule's point because (despite their efforts to argue
otherwise) the United States retains the core of a Madisonian system that
imposes legal limits on the President. We have less confidence that, were
those legal limits to be lifted, the chances would not appreciably increase. It
is true that Posner and Vermeule offer various plausible and specific ways
that public opinion constrains the President, but while the particular methods
of checking the Executive through public opinion seem important when
linked with a Madisonian system (as Part II says), they do not seem all that
persuasive without it. At minimum, what would happen if the constraints
were lifted seems open to speculation.
Perhaps most importantly, we do not think the separated-powers
constraints on the President are solely (or even mainly) designed to avoid
full-scale executive tyranny. There can be executive abuses and material
losses of freedom without all-encompassing abuses and total loss of freedom.
A "soft" dictatorship is also to be feared. By putting the focus on "tyranny,"
the book's Chapter 6 sets up a bit of a straw man. Of course, a bloodthirsty
dictator of the Hitler model is to be feared. But perhaps of more practical
concern is a Hugo Chavez-type figure who operates to some extent within
the constitutional framework but also abuses his authority and frequently
overreaches to punish opponents and achieve his ends.
Posner and Vermeule admit that the soft dictatorship is a more plausible
fear,130 but they do little to assuage it. They believe that "an educated and
leisured population, and the regular cycle of elections, will themselves check
executive abuses."'31 Yet elsewhere they say only that "politics and public
opinion at least block the most lurid forms of executive abuse." 32 In
particular, one may doubt the effectiveness of majority opinion to check
abuses against political minorities and unpopular views, and a legally unconstrained executive may be able to reward supporters and punish dissent in
ways that make organizing popular opposition (even by a majority) difficult.
Of course, the Madisonian system is not a sure check on such abuses either.
But legal constraints that offer some hope of protection to dissenting views,
and at least force a potentially abusive or self-dealing executive to work in
conjunction with independent branches, seem to offer materially better assurance than the purely unconstrained Executive.
Of course, it is impossible to prove that the core Madisonian constraints
have protected against executive tyranny (of the soft or hard kind) in the past.
Perhaps, as Posner and Vermeule suggest,'13 the United States has merely
been lucky, or its great wealth and other advantages have allowed it to
130. See id. at 203-04 (acknowledging that tyrannophobia can more plausibly be justified "as a
deterrent to low-level executive abuses").
131. Id. at 204.
132. Id at 5.
133. Id. at 196-202.
2012]
The Goldilocks Executive
1007
overcome structural disadvantages in its political system. And it is equally
speculative to say what might happen if the Madisonian constraints were
removed. But, to harness a different clich6, we suggest that "if it's not
broken, don't fix it." We do not see evidence that legal constraint renders the
U.S. Executive fatally ineffective. Without opining about incremental
changes to executive power, we doubt that a substantial case can be made for
the legally unbound Executive that Posner and Vermeule appear to favor.
Conclusion
The Executive Unbound rests on four claims, two descriptive and two
normative. The descriptive claims are the book's most prominent ones-that
the modern Executive is unbound by law but bound by public opinion. The
normative claims are more challenging and provocative-that a legally
unbound executive is desirable because legal constraints generate suboptimal
executive policies and that legal constraints are unnecessary because the
restraints of public opinion are sufficient.
As a description of the modern Executive, we think the book greatly
overstates. It is true that the Executive wields greater power today than 200
years ago. The reasons for executive expansion are well described in the
book. But enhanced executive power does not equal unbound executive
power. The basic Madisonian framework remains intact: the Constitution
limits the powers the President can exercise; Congress and the courts wield
powers that potentially or actually check the President. We agree that public
opinion constrains the President, but even here, the book may overstate by
failing to appreciate the public's preference for an executive bound by law.
Because of this preference, public opinion reinforces legal constraints on the
President while also providing additional political constraints. Thus, if the
authors mean to say, as they appear to say in the conclusion, merely that
legal constraints have weakened, we agree (but we do not think that claim is
especially novel). If they mean to say-as they appear to say in the
introduction-that legal constraints have vanished, we think their claim is
manifestly implausible.
Stripped of its main descriptive point, the book becomes a normative
argument for dismantling existing legal constraints on the Executive. This, it
is said, will lead to better outcomes, especially in crises. But the book provides few, if any, concrete examples of constrained executive power leading
to bad results. More significantly, the book argues forcefully that legal constraints on the Executive are worthless because there is no danger of
executive tyranny. We remain skeptical. The book employs social science
data to suggest that tyranny does not occur in wealthy democracies. But we
find the data inconclusive and the historical record thin. And in any event,
the concern is not just tyranny but executive overreaching and abuse.
In contrast, the current system of modest constraints on an energetic
President seems the safer course. We are not persuaded that the trade-off
1008
Texas Law Review
[Vol. 90:973
between executive energy and executive constraint that the Framers identified has been rendered obsolete by modem developments. Although there
will be various views of how the balance should be struck, some attempt to
balance the two seems preferable to adopting one extreme. A system with a
series of checks on a robust executive has stood the test of time, giving us
something of a Goldilocks Executive. It should, we think, take powerful evidence for us to abandon it. Posner and Vermeule simply do not provide that
powerful evidence. So while we might be better off with a stronger
executive, and public opinion might be enough to prevent tyranny, we, at
least, are unwilling to make the leap.