1791
VOLUME 122 MAY 2009 NUMBER 7
© 2009 by The Harvard Law Review Association
ARTICLE
LAW FOR STATES: INTERNATIONAL LAW,
CONSTITUTIONAL LAW, PUBLIC LAW
Jack Goldsmith and Daryl Levinson
TABLE OF CONTENTS
I. I
NTRODUCTION
……………………………………………………………………………………………………1792
II. T
HE
P
ROBLEM
OF
U
NCERTAINTY
……………………………………………………………………..1801
A. International Law ……………………………………………………………………………………………..1803
B. Constitutional Law ……………………………………………………………………………………………1808
C. Public Law ………………………………………………………………………………………………………..1817
III. T
HE
P
ROBLEM
OF
E
NFORCEMENT
……………………………………………………………………1822
A. International Law ……………………………………………………………………………………………..1824
B. Constitutional Law ……………………………………………………………………………………………1830
C. Public Law ………………………………………………………………………………………………………..1840
IV. T
HE
P
ROBLEM
OF
S
OVEREIGNTY
………………………………………………………………………1843
A. International Law ……………………………………………………………………………………………..1845
B. Constitutional Law ……………………………………………………………………………………………1852
C. Public Law ………………………………………………………………………………………………………..1858
V. C
ONCLUSION
………………………………………………………………………………………………………..1862
1792
L
AW
FOR
S
TATES
: I
NTERNATIONAL
L
AW
,
C
ONSTITUTIONAL
L
AW
, P
UBLIC
L
AW
Jack Goldsmith
∗
& Daryl Levinson
∗∗
International law has long been viewed with suspicion in Anglo-American legal thought.
Compared to the paradigm of domestic law, the international legal system seems different
and deficient along a number of important dimensions. This Article questions the
distinctiveness of international law by pointing out that constitutional law in fact shares
all of the features that are supposed to make international law so dubious. In mapping
out these commonalities, the Article suggests that the traditional international/domestic
distinction may obscure what is, for many purposes, a more important and generative
conceptual divide. That divide is between “public law” regimes like international and
constitutional law that constitute and govern the behavior of states and governments and
“ordinary domestic law” that is administered by and through the governmental
institutions of the state.
I. I
NTRODUCTION
he divide between international and domestic law runs deep in
Anglo-American legal thought. Domestic law is taken to be the
paradigm of how a legal system should work. Legal rules are promul-
gated and updated by a legislature or by common law courts subject to
legislative revision. Courts authoritatively resolve ambiguities and un-
certainties about the application of law in particular cases. The indi-
viduals to whom laws are addressed have an obligation to obey legiti-
mate lawmaking authorities, even when legal rules stand in the way of
their interests or are imposed without their consent. And in cases of
disobedience, an executive enforcement authority, possessing a monop-
oly over the use of legitimate force, stands ready to coerce compliance.
Measured against the benchmark of domestic law, international law
seems different and deficient along each of these dimensions. Interna-
tional law has no centralized legislature or hierarchical court system
authorized to create, revise, or specify the application of legal norms,
and as a result is said to suffer from irremediable uncertainty and po-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
∗
Henry L. Shattuck Professor of Law, Harvard Law School.
∗∗
Fessenden Professor of Law, Harvard Law School.
For helpful comments we thank Matthew Adler, Akhil Amar, David Bederman, Eyal
Benvenisti, Nehal Bhuta, Gabriella Blum, Lorenzo Casini, Richard Fallon, David Golove, Ryan
Goodman, Duncan Hollis, Benedict Kingsbury, Mattias Kumm, Maximo Langer, Frank
Michelman, Eric Posner, Kal Raustiala, Jim Ryan, Scott Shapiro, Seana Shiffrin, Paul Stephan,
Cass Sunstein, Adrian Vermeule, and participants at workshops at the Emory, NYU, Rutgers-
Camden, Stanford, Tel Aviv, Temple, UCLA, University of Chicago, University of Pennsylvania,
and Yale law schools. We also thank Steven Horowitz, Will Levi, Joshua Minix, Hagan Scotten,
and David Zimmer for very helpful research.
T
2009] LAW FOR STATES 1793
litical contestation. Out of deference to state sovereignty, international
law is a “voluntary” system that obligates only states that have con-
sented to be bound, and thus generally lacks the power to impose obli-
gations on states against their interests. As a result, the content of in-
ternational law often reflects the interests of powerful states. And to
the extent that international law diverges from those interests, power-
ful states often interpret it away or ignore it. They are able to do so
because the international legal system lacks a super-state enforcement
authority capable of coercing recalcitrant states to comply. These
characteristics of the international legal system have led realists and
other skeptics to conclude that, in both form and function, interna-
tional law is a qualitatively different and lesser species of law — if it
qualifies as law at all.
Constitutional law, in contrast, has been subject to few such
doubts. Conceived as the overarching framework for, and thus insepa-
rable from, the statutes, regulations, and common law rules that com-
prise the familiar domestic legal system, constitutional law sits securely
opposite international law on the domestic side of the divide. Unlike
the decentralized and institutionally incomplete international legal sys-
tem, moreover, constitutional law in the United States and other coun-
tries appears closer in form to ordinary, paradigmatically “real” domes-
tic law because it typically features a proto-legislative enactment and
amendment process, as well as an authoritative judiciary to resolve
ambiguities about meaning and to enforce obligations against govern-
ment officials. In contrast to the dubious efficacy of international law,
constitutional law is generally assumed to serve as an important and
effective constraint on government behavior, a meaningful check on
the interests of the powerful.
The perceived differences between international and constitutional
law have taken on a normative cast as well. For centuries, theorists
have worried about how to reconcile the legal constraints of interna-
tional law with the idea, or ideal, of state sovereignty. Sovereignty is
supposed to mean that states cannot be subject to any higher author-
ity; international law and the institutions it creates seem to represent
just such authorities. As applied to democratic states like the United
States, assertions of sovereignty often blur into defenses of democratic
self-determination. A deep strain of U.S. political thought portrays in-
ternational law as an illegitimate attempt by democratically unac-
countable foreigners to interfere with the legitimate self-governance of
democratic majorities at home.
1
Constitutional law could, and some-
times does, provoke similar objections, since it too purports to interfere
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
1 For background, see generally D
UANE
T
ANANBAUM
, T
HE
B
RICKER
A
MENDMENT
C
ONTROVERSY
(1988).
1794 HARVARD LAW REVIEW [Vol. 122:1791
with the ability of the “sovereign” people to govern themselves as they
see fit. Yet the most insistent proponents of U.S. sovereignty in the
face of international law do not see constitutional law as a comparable
threat. To the contrary, they hold up constitutional law as the ultimate
expression of American sovereignty and self-government.
2
On this
view, “[t]o support international law is to support fundamental con-
straints on democracy,”
3
but constitutional law “represent[s] the na-
tion’s self-given law.”
4
This Article questions whether these apparent differences between
international and constitutional law really run as deep as is commonly
supposed. Despite superficial appearances to the contrary, consti-
tutional law, like international law, lacks a centralized legislature to
specify and update legal norms, and although constitutional courts
possess some ability to resolve the existence and meaning of constitu-
tional norms, they are limited in special ways that prevent them from
providing authoritative settlement. As a result, constitutional law suf-
fers from the same kinds of foundational uncertainty and contestation
over meaning that are viewed as characteristic of international law.
Constitutional law also shares with international law the absence of an
enforcement authority capable of coercing powerful political actors to
comply with unpopular decisions. This lack of an enforcement author-
ity raises doubts about legal compliance and, more generally, the abil-
ity of legal norms to constrain and not just reflect political interests.
And in much the same way as international law, constitutional law
strains to legitimate the limits it purports to impose on popular self-
government by invoking various forms (or fictions) of prior sovereign
consent.
There are many complexities here, which we discuss in the pages
that follow. But the general point is that the basic features of interna-
tional law that lead lawyers and theorists to question its efficacy and
legitimacy are shared by constitutional law. Whatever one makes of
the descriptive and normative doubts to which international law is
perpetually subject, we argue that constitutional law should be subject
to the same doubts.
We are less interested in assessing these doubts on the merits, how-
ever, than in understanding their common origins and consequences.
In mapping out these commonalities, we hope to show that the tradi-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
2 See, e.g., J
EREMY
R
ABKIN
, W
HY
S
OVEREIGNTY
M
ATTERS
9 (1998) (“Because the
United States is fully sovereign, it can determine for itself what its Constitution will require. And
the Constitution necessarily requires that sovereignty be safeguarded so that the Constitution it-
self can be secure.”).
3 Jed Rubenfeld, Commentary, Unilateralism and Constitutionalism, 79 N.Y.U. L. R
EV
. 1971,
2020 (2004).
4 Id. at 1994.
2009] LAW FOR STATES 1795
tional divide between domestic and international law obscures what is,
for many purposes, a more important and generative conceptual divide
between public law and ordinary domestic law.
5
By “public law” we
mean constitutional and international law — legal regimes that both
constitute and govern the behavior of states and state actors. By “or-
dinary domestic law” we mean the usual assortment of statutes and
common law that apply to private actors within a state and are admin-
istered by and through the governmental institutions of that state.
6
The respects in which both international and constitutional law differ
from ordinary domestic law follow from the distinctive aspiration of
public law regimes to constitute and constrain the behavior of state in-
stitutions and the distinctive difficulty these regimes face of not being
able to rely fully on these same state institutions for implementation
and enforcement.
The difficulty is indeed distinctive. We are deeply accustomed to
thinking of law as created by, working through, and inextricably
bound together with the political and legal institutions of the state.
7
Our paradigmatic conception of a legal system rests on the state’s de-
finitive monopoly over the power to make and enforce law in a given
territory. Without the backing and institutional support of the sover-
eign state’s consolidation of coercive power, authoritative lawmaking,
and binding dispute resolution, legal order as we intuitively know it —
which is to say, in the form of ordinary domestic law — cannot exist.
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
5 We do not mean to suggest that the distinction between international and constitutional law
is incoherent or insignificant. We merely hope to show that the similarities between the two kinds
of legal systems, which are suppressed by the traditional divide, are more interesting and fruitful
than has been commonly recognized. To avoid any philosophical confusion, this should be under-
stood as a pragmatic claim about the utility of working with different conceptual frameworks, not
as any sort of metaphysical claim about the true joints along which reality must be cut. We are
simply offering a new conceptual tool that we believe will prove useful for some purposes.
6 We avoid the term “private law” because when it is used in contrast to public law it comes
freighted with the unnecessary (for our purposes) theoretical and historical baggage of the pub-
lic/private distinction, and because its meaning becomes even more ambiguous as it crosses the
boundary between domestic and international law. We also elide intermediate cases where state
actors are subject to contract, criminal, administrative, or tort laws. We call these cases “interme-
diate” because they implicate some of the features shared by international and constitutional law,
but not others. To the extent these legal regimes are based on statutes, for example, they are not
subject to the same problems of uncertainty regarding the authoritative sources of legal norms or
the same concerns about constraining sovereignty. They are, however, still confronted by the ab-
sence of any super-state enforcement authority. It is worth noting that much of what is commonly
described as administrative law and government contract and tort law is, in fact, straightforward
constitutional law, and thus fully encompassed by our discussion.
7 See, e.g., William Ewald, Comment on MacCormick, 82 C
ORNELL
L. R
EV
. 1071, 1072
(1997) (“[M]ost modern legal theorists, at least tacitly, accept Kelsen’s identification of law and
state. That is, they take it for granted that the primary task of legal theory is to explicate the legal
systems of the modern nation-state. The modern state is taken as the paradigm case; and on those
occasions when supra-national or international law is discussed, this form of law is generally
treated as a marginal case, if not neglected altogether.”).
1796 HARVARD LAW REVIEW [Vol. 122:1791
And yet, of course, we have systems of public law, international
and constitutional, which cast the state as the subject (and product)
rather than solely the source of law. Even as legal systems for the
state have become a familiar and ubiquitous feature of the modern
world, how these systems work remains surprisingly mysterious. Most
immediately puzzling, perhaps, is how public law regimes can effec-
tively constrain the behavior of states in the absence of any super-state
enforcement authority. “[W]hy do people with power accept limits to
their power? . . . [W]hy do people with guns obey people without
guns?”
8
But it is not just guns that the state possesses and systems of
public law lack. Public law regimes are also missing — and must bor-
row or functionally recreate — institutions with the legislative and ju-
dicial capacities to authoritatively make and interpret law. Also ab-
sent from public law is the legitimate, or at least taken-for-granted
authority of the state to exercise coercion through law, overriding sov-
ereignty-based claims of self-determination and self-government. How
public law regimes can work, and work effectively, despite these
handicaps is a puzzle that has seldom come into clear focus in Anglo-
American legal theory.
9
Public law has been relegated to the blurry
margins, we believe, largely because the artificial divide within public
law has made it easy to dismiss the international legal system as an
outlier. Constitutional law is less easy to dismiss.
By assimilating constitutional and international law and examining
how the two systems similarly manage the peculiar difficulties of run-
ning a legal system outside of the state, we hope to bring focus to the
possibilities and limitations of public law as a distinctive legal form.
More accurately, we hope to return focus, for we are far from the first
to take this perspective; our contribution can be seen more in the na-
ture of resurrection than invention. Many prominent western political
theorists conceived of what we would today call constitutional and in-
ternational law as conjoined efforts to regulate the sovereign state
from an “internal” and “external” perspective.
10
The father of the
modern conception of sovereignty, Jean Bodin, also contemplated con-
stitutional rules and a regime of international law as limits on the oth-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
8 Stephen Holmes, Lineages of the Rule of Law, in D
EMOCRACY
AND
THE
R
ULE
OF
L
AW
19, 24 (José María Maravall & Adam Przeworski eds., 2003); accord N
ICCOLÒ
M
ACHIAVELLI
,
T
HE
P
RINCE
71 (Leo Paul S. de Alvarez trans., 1981) (1532) (“[T]here cannot be good laws where
there are not good arms . . . .”).
9 At the level of jurisprudence, of course, explaining the coordinated recognition of and com-
pliance with law generally has long been identified as a fundamental problem.
10 See F.H . H
INSLEY
, S
OVEREIGNTY
126–213 (2d ed., Cambridge Univ. Press 1986) (1966).
Even before the state emerged as a distinctive form in the sixteenth century, medieval theorists
viewed the problem of regulating secular rulers in both their domestic and external affairs
through the common lens of Christian theology and natural law. See id. at 45–125, 164–78.
2009] LAW FOR STATES 1797
erwise illimitable sovereignty of the state.
11
Hugo Grotius’s effort to
ground public international law in the will and practice of states was
built upon an elaborate and influential conception of domestic sover-
eignty and domestic constitutional constraint.
12
Thomas Hobbes was
less enthusiastic about attempts to constrain sovereignty, but his con-
ception of international relations as a war of all against all and his
concomitant skepticism about international law went hand-in-hand
with the near-limitless power of the Leviathan over its own citizens.
13
And in the nineteenth century, following in Hobbes’s footsteps, John
Austin sharply distinguished ordinary domestic law, which he con-
ceived as the command of the sovereign backed by force, from interna-
tional law and constitutional law, both of which he believed were, for
similar reasons, rules of “positive morality”
14
rather than laws “prop-
erly so called.”
15
This unifying perspective has been largely lost,
16
but it is ripe for
recovery. Political and legal developments in recent decades have
blurred the international/constitutional law divide in a number of dif-
ferent ways. Debates about whether the European Union is best un-
derstood as an international or a constitutional legal order — echoing
U.S. debates about whether the Articles of Confederation should be
classified as a treaty or a constitution
17
— both presuppose and prob-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
11 See generally J
EAN
B
ODIN
, S
IX
B
OOKS
OF
THE
C
OMMONWEALTH
25 (M.J. Tooley
trans., Basil Blackwell 1955) (1576) (conceiving of sovereignty as “that absolute and perpetual
power” vested in the commonwealth). On Bodin and constitutional constraints, see H
INSLEY
,
supra note 10, at 120–25; S
TEPHEN
H
OLMES
, P
ASSIONS
AND
C
ONSTRAINT
100–33 (1995); and
J.H. Burns, Sovereignty and Constitutional Law in Bodin, 7 P
OL
. S
TUD
. 174 (1959). On Bodin
and international law, see H
INSLEY
, supra note 10, at 179–83. See also id. at 180 (noting that
Bodin’s work produced “the doctrine of sovereignty in relation to the internal structure of the po-
litical community and, with regard to the relations between communities, the recognition
that . . . there was a need for a new category of law [separate from natural law] — for interna-
tional law”).
12 See generally H
UGO
G
ROTIUS
, T
HE
R
IGHTS
OF
W
AR
AND
P
EACE
(Richard Tuck ed.,
2005) (1625); H
INSLEY
, supra note 10, at 139–41; H
ISTORY
OF
P
OLITICAL
P
HILOSOPHY
(Leo
Strauss & Joseph Cropsey eds., Univ. of Chi. Press 1981) (1963); R
ICHARD
T
UCK
, P
HILOSOPHY
AND
G
OVERNMENT
1572–1651 (1993).
13 See generally T
HOMAS
H
OBBES
, L
EVIATHAN
(A.R. Waller ed., Cambridge Univ. Press
1935) (1651).
14 J
OHN
A
USTIN
, T
HE
P
ROVINCE
OF
J
URISPRUDENCE
D
ETERMINED
141 (Wilfrid E.
Rumble ed., Cambridge Univ. Press 1995) (1832).
15 Id. at 143; see also id. at 141–43; 254–64.
16 Why it has been lost is an interesting and, as best we can tell, unanswered question. We
suspect that the rise of positivism and written constitutionalism in the late eighteenth and early
nineteenth centuries led to a conceptual splintering of international and constitutional law, al-
though this is a story that has not yet been fully told. For an enlightening historical analysis of
the relationship between international strategic affairs and domestic constitutional orders that is
orthogonal to our project, see generally P
HILIP
B
OBBITT
, T
HE
S
HIELD
OF
A
CHILLES
: W
AR
,
P
EACE
,
AND
THE
C
OURSE
OF
H
ISTORY
(2002).
17 Compare Bruce Ackerman & Neal Katyal, Our Unconventional Founding, 62 U. C
HI
. L.
R
EV
. 475, 478–87 (1995) (arguing that the Articles were a constitution), with Akhil Reed Amar,
1798 HARVARD LAW REVIEW [Vol. 122:1791
lematize a qualitative distinction between the two kinds of legal re-
gimes and lead some to wonder what should turn on the difference.
18
Similar questions are pressed by the so-called movement toward
“global constitutionalism”
19
and the increasingly common characteriza-
tion of international arrangements like the WTO as “constitutional.”
20
In the United States, judicial and political debates surrounding the
war on terrorism have brought to the fore the complicated overlapping
relationship between constitutional and international rights and obliga-
tions, and more abstractly, between constitutional and global justice.
21
Responding to these and other proliferating transpositions of the inter-
national and the constitutional, political scientists have begun to ana-
lyze some of the parallels between the architecture of international and
constitutional regimes.
22
Our aim in this Article is to clarify, deepen,
and extend these arguments, and to explain their relevance to modern
legal theory, by analyzing constitutional and international law as con-
ceptually linked forms of public law.
The Article proceeds as follows. Each of the next three Parts takes
a standard critique of international law, shows how it also applies to
constitutional law, and then discusses how the problem is structurally
symptomatic of public law, in contrast to ordinary domestic law, and
how international law and constitutional law attempt to deal with the
problem in similar ways. Part II addresses the problem of legal uncer-
tainty, which arises from the absence of centralized legislative and ju-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
The Consent of the Governed: Constitutional Amendment Outside Article V, 94 C
OLUM
. L. R
EV
.
457, 464–69 (1994) (arguing that the Articles were a treaty).
18 For overviews of the European Union debate, see generally Mattias Kumm, Beyond Golf
Clubs and the Judicialization of Politics: Why Europe Has a Constitution Properly So Called, 54
A
M
. J. C
OMP
. L. S
UPPLEMENT
505 (2006); and J.H.H. Weiler, The Transformation of Europe, 100
Y
ALE
L.J. 2403 (1991).
19 See generally Ernest A. Young, The Trouble with Global Constitutionalism, 38 T
EX
. I
NT
’
L
L.J. 527 (2003).
20 See John O. McGinnis & Mark L. Movsesian, Commentary, The World Trade Constitution,
114 H
ARV
. L. R
EV
. 511 (2000); see also Jeffrey L. Dunoff, Constitutional Conceits: The WTO’s
‘Constitution’ and the Discipline of International Law, 17 E
UR
. J. I
NT
’
L
L
.
647 (2006) (collecting
and critically examining the meaning of “constitutional” characterizations of the WTO).
21 See, e.g., Boumediene v. Bush, 128 S. Ct. 2229 (2008); Hamdan v. Rumsfeld, 548 U.S. 557
(2006). On the relationship between constitutional and global justice, see David Golove, Incorpo-
rating Global Justice into the U.S. Constitution (Nov. 19, 2007) (unpublished manuscript, on file
with the Harvard Law School Library).
22 Important efforts in this regard include G. J
OHN
I
KENBERRY
, A
FTER
V
ICTORY
(2001);
Alec Stone, What Is a Supranational Constitution? An Essay in International Relations Theory,
56 R
EV
. P
OL
. 441 (1994); Clifford James Carrubba, A Model of the Endogenous Development of
Judicial Institutions in Federal and International Systems (unpublished manuscript, on file with
the Harvard Law School Library); and Jeffrey K. Staton & Will H. Moore, The Last Pillar To
Fall? Domestic and International Legal Institutions (Nov. 22, 2008) (unpublished manuscript, on
file with the Harvard Law School Library). We are especially indebted to, and build directly
upon, Christopher A. Whytock, Thinking Beyond the Domestic-International Divide: Toward a
Unified Concept of Public Law, 36 G
EO
. J. I
NT
’
L
L. 155 (2004).
2009] LAW FOR STATES 1799
dicial institutions with widely recognized authority to determine the
meaning and application of legal rules. International and constitu-
tional law have both gone some distance toward reducing legal uncer-
tainty by co-opting or substituting for ordinary domestic legislatures
and courts, but compared to the benchmark of ordinary domestic law,
their success remains only partial. Part III focuses on the absence of
any super-state enforcement authority to compel compliance with legal
rules. It shows that the mechanisms through which compliance is
achieved must be different from the prototypical threat of coercion or
punishment by the state that backs, and to some extent explains, com-
pliance with the statutory and common law rules of ordinary domestic
law. Part III also demonstrates how an alternative set of “self-
enforcement” mechanisms have been developed in both international
and constitutional law and theory. Part IV examines the normative
problems that arise in both international law and constitutional law
from the attempt to bind states and governments to law. Notwith-
standing philosophical concerns about legitimate political authority
over private individuals, the authority of domestic law in a well-
ordered state is generally taken for granted. Not so the authority of
international and constitutional law, each of which is perpetually ques-
tioned or resisted on grounds of sovereignty, self-determination, and
democracy. The Article concludes by suggesting some constructive
implications of assimilating international and constitutional law into a
more unified vision of public law.
Three preliminary methodological points will further clarify the
scope and ambition of this project. First, while we draw upon the
classics of positivist jurisprudence to frame our analysis, we do not in-
tend to make or endorse any claims at the level of jurisprudence. The
conventional diagnoses of the deficiencies of international law that we
take as our starting point might be understood to implicate jurispru-
dential claims about what should count as law or a legal system — in-
cluding some claims that are controversial or have been discredited at
a philosophical level. Few contemporary jurisprudes would join
Hobbes and Austin in casting sanction-based commands or sovereignty
in the central roles these concepts play in the conventional wisdom
about the exceptionalism of international law (and therefore in Parts
III and IV of this Article, respectively). In bludgeoning international
law with its Austinian deficits, conventional legal culture may display
some measure of philosophical naiveté in failing to internalize H.L.A.
Hart’s refutation of Austin — not to mention a selective blindness to-
ward the applicability of the Austinian criteria to constitutional law.
But another possibility is that Hobbesian and Austinian features of the
legal order, or their absence, remain relevant for functional and norma-
tive reasons beyond the jurisprudential one of distinguishing law from
other kinds of normative order. For those who believe that sanctions
are an empirically important determinant of legal compliance, for ex-
1800 HARVARD LAW REVIEW [Vol. 122:1791
ample, Hart’s analysis of the nature of law and legal systems provides
no reason to stop caring about them. The same is true of those who
believe that sovereignty or democratic self-determination are norma-
tively significant values that legal constraints compromise. The com-
mand theory of law may be dead jurisprudentially, but some of the
state-centric features of legal systems that Hobbes and Austin empha-
sized — including sanctions and sovereignty — retain central signifi-
cance at less rarified levels of legal theory and practice.
23
Whatever
jurisprudential progress has been made toward working out a concep-
tion of law that is relatively autonomous of the state does nothing to
obviate the functional and normative imperatives that the state has
long been understood to serve.
Second, consistent with our perhaps exceptionally American per-
spective on the apparent differences between international and consti-
tutional law, our focus throughout is on the United States’s system of
constitutional law. Nonetheless, we believe that our fundamental
points apply to other constitutional systems as well. Indeed, we be-
lieve that the general features of constitutional law in our focus are
analytically necessary to the ambitions of any sort of constitutional re-
gime that aspires to limit (as well as constitute) political authority.
24
Finally, in case it does not go without saying, our argument is not
that international and constitutional law are the same in all respects.
Some formal differences between the two kinds of legal regime are ob-
vious (if not entirely clear-cut). International law predominantly ad-
dresses relations between and among states, whereas constitutional law
predominantly addresses the political structure of a single state; rules
of international law are created primarily through treaties entered into
by states or by customary state practice, whereas rules of constitu-
tional law are created primarily by popular ratification of an authori-
tative text or conventions of political life that have achieved normative
status as higher law; and so on. And while our analysis of the two le-
gal regimes along the functional dimensions of uncertainty and en-
forcement and the normative dimension of sovereignty emphasizes im-
portant similarities, we also pause to notice differences. From a
functional perspective, the size and heterogeneity of the international
community may make it more difficult for the international legal sys-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
23 We do not touch upon natural law theories at all, simply because these theories offer no re-
sources for elaborating what we take to be the significant and interesting differences between
public and ordinary domestic law (though they do offer resources for portraying international and
constitutional law as fundamentally alike).
24 Where the analytic generalizations are least transparent — particularly in Part II, with re-
spect to the problem of legal uncertainty — we sketch out the comparative extensions. Our dis–
cussions of the problems of enforcement and sovereignty in Parts III and IV respectively should
translate straightforwardly to other constitutional systems.
2009] LAW FOR STATES 1801
tem to develop institutional mechanisms for specifying and enforcing
legal rules than for constitutional systems of smaller and more ho-
mogenous states to do the same.
25
The fact that American constitu-
tional law is made, interpreted, and implemented exclusively by
Americans may make a normative difference to those who believe that
sharing governance authority with a broader political community will
invariably threaten American sovereignty, or that a politico-legal
community can only be sustained at the level of the nation-state. We
recognize these and other differences, but the ambition of this project
is to reveal an important set of similarities that such differences may
have masked.
II. T
HE
P
ROBLEM
OF
U
NCERTAINTY
For positivists, a defining feature of law is broad agreement in soci-
ety on what counts as a legal rule and on what identifiable legal rules
require in concrete cases. A defining feature of the state is that its in-
stitutions foster this agreement. It has been common ground among
positivist legal theorists for centuries that a well-functioning legal sys-
tem requires something like the institutional apparatus of the modern
state — legislatures with widely acknowledged authority to enact and
modify legal norms and courts with widely acknowledged authority to
adjudicate disputes about the proper interpretation of those norms —
in order to coordinate understandings of what the law requires.
26
Thus, for Hobbes, the state comes into being to resolve disagreement
about what counts as law by serving as the singular and decisive
source of legal norms.
27
Without the centralized authority of the Le–
viathan, Hobbes insisted, the divergent interests, values, and perspec-
tives of individuals in the anarchical state of nature would make it
impossible to coordinate any legal order. Bringing this insight to bear
on modern political order, Hart emphasized that legal systems solve
the problem of “uncertainty” by providing institutions and procedures
for resolving what counts as law, “either by reference to an authorita-
tive text or to an official whose declarations on this point are authori-
tative.”
28
Hart famously described how mature legal systems accom-
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––
25 See E
RIC
A. P
OSNER
, T
HE
P
ERILS
OF
G
LOBAL
L
EGALISM
11 (forthcoming 2009).
26 See T
ERRY
N
ARDIN
, L
AW
, M
ORALITY
,
AND
THE
R
ELATIONS
OF
S
TATES
69–83 (1983)
(describing the views of Hobbes, Locke, Rousseau, Kant, Bentham, and James Mill).
27 See H
OBBES
, supra note 13, at 189; see also J
EREMY
W
ALDRON
, L
AW
AND
D
ISAGREE-
MENT
39–41 (1999). This view was hardly limited to Hobbes. See, e.g., Jeremy Waldron, Kant’s
Legal Positivism, 109 H
ARV
. L. R
EV
. 1535 (1996).
28 H.L.A. H
ART
, T
HE
C
ONCEPT
OF
L
AW
92 (2d ed. 1994); see also id. at 93 (“Disputes as to
whether an admitted rule has or has not been violated will always occur . . . if there is no agency
specially empowered to ascertain finally, and authoritatively, the fact of violation.”). Hart calls
this second problem “inefficiency,” but its relationship to “uncertainty” in the intuitive sense is
obvious.
1802 HARVARD LAW REVIEW [Vol. 122:1791
plish this task through “secondary rules” of recognition, change, and
adjudication that determine what the primary legal rules are and when
they have been violated.
29
Within the institutional framework of the
modern state, these rules serve to identify legislatures and courts as the
authoritative sources of legal norms and the authoritative arbiters of
disputes over their meaning.
The crucial role of state institutions in coordinating public under-
standings of the sources and proper interpretation of legal norms has
long been a reason for skepticism about international law. Interna-
tional society lacks a super-state — including a super-legislature and a
super-judiciary — to create the kind of consensus about operative legal
norms that these institutions enable within states. This deficiency is
why cosmopolitan theorists have long argued for creating a global leg-
islature and court to govern world affairs.
30
In fact, the development
of international institutions has proceeded some distance, and the
number and density of international institutions continues to increase.
But the international system is still a long way from establishing any-
thing like a single, comprehensive global legislature or court. The dif-
ficulty of creating such institutions, or of duplicating their functions
through some other institutional arrangement, leads to pessimism
