Dahl Robert How Democrati Is The American Constitution

00dahlFM.
i_x 11/27/01 4:37 PM Page i
The Castle Lectures in Ethics, Politics, and Economics

00dahlFM.i_x 11/27/01 4:37 PM Page ii
00dahlFM.i_x 11/27/01 4:37 PM Page iii
How Democratic
Is the
American
Constitution?
Robert A. Dahl
Yale University Press / New Haven & London

00dahlFM.i_x 11/27/01 4:37 PM Page iv
Published with assistance from the foundation established in memory of

Philip Hamilton McMillan of the Class of 1894, Yale College.
Copyright 2001 by Yale University.

All rights reserved.
This book may not be reproduced, in whole or in part, including
illustrations, in any form (beyond that copying permitted by Sections
107 and 108 of the U.S. Copyright Law and except by reviewers for
the public press), without written permission from the publishers.
Designed by James J. Johnson and set in New Caledonia and

Bulmer types by Integrated Publishing Solutions.
Printed in the United States of America.
Library of Congress Cataloging-in-Publication Data
Dahl, Robert Alan, 1915

How democratic is the American Constitution?/
Robert A. Dahl.
p. cm. (The Castle lecture series)
Includes bibliographical references and index.
ISBN 0-300-09218-0 (cloth : alk. paper)
1. Constitutional lawUnited States. 2. Constitutional history
United States. 3. Democracy. I. Title. II. Series.
KF4550 .D34 2001
342.7302dc21 2001004323
A catalogue record for this book is available from the British Library.
The paper in this book meets the guidelines for permanence and
durability of the Committee on Production Guidelines for Book
Longevity of the Council on Library Resources.
10 9 8 7 6 5 4 3 2 1
00dahlFM.i_x 11/27/01 4:37 PM Page v
Parts of this book were given as a series

of Castle Lectures in Yales Program in
Ethics, Politics, and Economics, deliv-
ered by Robert Dahl at Yale University
in 2000.
The Castle Lectures, endowed by Mr.

John K. Castle, honor his ancestor, the
Reverend James Pierpont, one of Yales
founders. Given by prominent public
figures, the lectures are intended to
promote reflection on the moral foun-
dations of society and government and
to enhance understanding of ethical is-
sues facing individuals in our complex
modern society.
00dahlFM.i_x 11/27/01 4:37 PM Page vi
00dahlFM.i_x 11/27/01 4:37 PM Page vii
Contents
Acknowledgments ix
CHAPTER 1. Introduction: Fundamental Questions 1
CHAPTER 2. What the Framers Couldnt Know 7
CHAPTER3. The Constitution as a Model:
An American Illusion 41
CHAPTER 4. Electing the President 73
CHAPTER 5. How Well Does the Constitutional
System Perform? 91
CHAPTER 6. Why Not a More Democratic
Constitution? 121
CHAPTER 7. Some Reflections on the Prospects
for a More Democratic Constitution 141
Appendix A: On the Terms Democracy
and Republic 159
Appendix B: Tables and Figures 163
Notes 173
Index 191
00dahlFM.i_x 11/27/01 4:37 PM Page viii
00dahlFM.i_x 11/27/01 4:37 PM Page ix
Acknowledgments
THE INVITATION TO DELIVER THE CASTLE LECTURES AT

Yale provided me with an incentive for bringing into
focus views about the American Constitution that I had
gradually formed over many years. Though I had set
forth some of my arguments in various essays and book
chapters, others remained largely implicit or undevel-
oped until I drafted the lectures. This book embodies
the substance of the Castle Lectures, slightly revised
and enlarged, that I delivered in the early fall of 2000.
For extending to me the invitation to deliver the
lectures, I am indebted to Geoffrey Garrett, director
of the Program in Ethics, Politics, and Economics. To
both him and Ian Shapiro I want to express my appre-
ciation for their warm endorsement of the subject I
proposed for my lectures.
00dahlFM.i_x 11/27/01 4:37 PM Page x
x ac k n ow l e d g m e n t s
For her invaluable research assistance I am in-

debted to Jennifer Smith.
For their helpful queries, corrections, or contribu-
tions I also want to thank Wendell Bell, Kai Erikson,
Fred Greenstein, Steven Hill, Malcolm Jewell, Joseph
LaPalombara, Rogers Smith, and, at Yale University
Press, Ali Peterson for her sensitive editing and Larisa
Heimert for her skillful and energetic help in conduct-
ing the manuscript through the publishing process.
Finally, let me take this opportunity to express my
thanks to those members of the audience whose ques-
tions and comments enabled me to discover aspects of
my presentation that would benefit, in this published
version, from greater clarification or more extensive
treatment.
01dahl.001_006 11/27/01 4:38 PM Page 1
chapter 1
Introduction:
Fundamental Questions
M
Y AIM IN THIS BRIEF BOOK IS NOT TO PROPOSE
changes in the American Constitution but to
suggest changes in the way we think about
our constitution. In that spirit, Ill begin by posing a
simple question: Why should we Americans uphold
our Constitution?
Well, an American citizen might reply, it has been
our constitution ever since it was written in 1787 by a
group of exceptionally wise men and was then ratified
by conventions in all the states.1 But this answer only
leads to a further question.
To understand what lies behind that next question,
I want to recall how the Constitutional Convention that
met in Philadelphia during the summer of 1787 was
made up. Although we tend to assume that all thirteen
01dahl.001_006 11/27/01 4:38 PM Page 2
2 i n t ro d u c t i o n
states sent delegates, in fact Rhode Island refused to

attend, and the delegates from New Hampshire didnt
arrive until some weeks after the Convention opened.
As a result, several crucial votes in June and July were
taken with only eleven state delegations in attendance.
Moreover, the votes were counted by states, and al-
though most of the time most state delegations agreed
on a single position, on occasion they were too divided
internally to cast a vote.
My question, then, is this: Why should we feel
bound today by a document produced more than two
centuries ago by a group of fifty-five mortal men, actu-
ally signed by only thirty-nine, a fair number of whom
were slaveholders, and adopted in only thirteen states
by the votes of fewer than two thousand men, all of
whom are long since dead and mainly forgotten? 2
Our citizen might respond that we Americans are
free, after all, to alter our constitution by amendment
and have often done so. Therefore our present consti-
tution is ultimately based on the consent of those of us
living today.
But before we accept this reply, let me pose an-
other question: Have we Americans ever had an op-
portunity to express our considered will on our consti-
tutional system? For example, how many readers of
these lines have ever participated in a referendum that
asked them whether they wished to continue to be
governed under the existing constitution? The answer,
of course, is: none.
01dahl.001_006 11/27/01 4:38 PM Page 3
i n t ro d u c t i o n 3
Our citizen might now fall back on another line of

argument: Why should we change a constitution that
has served and continues to serve us well?
Although this is surely a reasonable line of argu-
ment, it does suggest still another question: By what
standards does our constitution serve us well? In par-
ticular, how well does our constitutional system meet
democratic standards of the present day? Ill turn to
this question in the next chapter.
And if our constitution is as good as most Ameri-
cans seem to think it is, why havent other democratic
countries copied it? As well see in Chapter 3, every
other advanced democratic country has adopted a con-
stitutional system very different from ours. Why?
If our constitutional system turns out to be unique
among the constitutions of other advanced democratic
countries, is it any better for its differences, or is it
worse? Or dont the differences matter? Ill explore
this difficult question in the fourth chapter.
Suppose we find little or no evidence to support
the view that our constitutional system is superior to
the systems of other comparable democratic countries,
and that in some respects it may actually perform
rather worse. What should we conclude?
As one part of an answer, I am going to suggest
that we begin to view our American Constitution as
nothing more or less than a set of basic institutions
and practices designed to the best of our abilities for
the purpose of attaining democratic values. But if an
01dahl.001_006 11/27/01 4:38 PM Page 4
important democratic value is political equality, wont

political equality threaten the rights and liberties we
prize? In Chapter 5, Ill argue that this viewfa-
mously defended by Tocqueville, among othersis
based on a misunderstanding of the relationship be-
tween democracy and fundamental rights.
Yet the question remains: if our constitution is in
some important ways defective by democratic stan-
dards, should we change it, and how? As I said, my
aim here is not so much to suggest changes in the ex-
isting constitution as to encourage us to change the
way we think about it, whether it be the existing one,
an amended version of it, or a new and more demo-
cratic constitution. That said, in my final chapter Ill
comment briefly on some possible changes and on the
obstacles to achieving them.
! ! !
BEFORE TURNING TO THESE QUESTIONS, I NEED TO DIS-

pose of two matters. One is purely terminological. In
discussing the formation of the constitution at the
Convention in 1787, I shall refer to the delegates as
the Framers, not, as is more common, the Founding
Fathers. I do so because many of the men who reason-
ably might be listed among the Founding Fathers
including such notables as John Adams, Samuel Adams,
Tom Paine, and Thomas Jeffersonwere not at the
Convention. (By my count, only eight of the fifty-five
01dahl.001_006 11/27/01 4:38 PM Page 5
i n t ro d u c t i o n 5
delegates to the Convention had also signed the Dec-

laration of Independence.)
The second matter is both terminological and sub-
stantive. Some readers may argue that the Founding
Fathers (including the Framers) intended to create a
republic, not a democracy. From this premise, accord-
ing to a not uncommon belief among Americans, it fol-
lows that the United States is not a democracy but a
republic. Although this belief is sometimes supported
on the authority of a principal architect of the Consti-
tution, James Madison, it is, for reasons I explain in
Appendix A, mistaken.
But even more important, the conclusion does not
follow from the premise. Whatever the intentions of the
Framers may have been, we would hardly feel bound
by them today if we believed that they were morally,
politically, and constitutionally wrong. Indeed, more
than two centuries of experience demonstrates that
whenever a sufficiently large and influential number
of Americans conclude that the views of the Framers
were wrong, they will change the constitution. Even if
the Framers did not intend their constitution to abol-
ish slavery, when later generations concluded that slav-
ery could no longer be tolerated and must be abol-
ished, they changed the constitution to conform with
their beliefs.
Even if some of the Framers leaned more toward
the idea of an aristocratic republic than a democratic
republic, they soon discovered that under the leader-
01dahl.001_006 11/27/01 4:38 PM Page 6
ship of James Madison, among others, Americans would

rapidly undertake to create a more democratic repub-
lic, and in doing so they would begin almost immedi-
ately to change the constitutional system the Framers
had created.
02dahl.007_040 11/27/01 4:39 PM Page 7
chapter 2
What the Framers Couldnt Know
W
ISE AS THE FRAMERS WERE, THEY WERE
necessarily limited by their profound igno-
rance.
I say this with no disrespect, for like many others I
believe that among the Framers were many men of ex-
ceptional talent and public virtue. Indeed, I regard
James Madison as our greatest political scientist and
his generation of political leaders as perhaps our most
richly endowed with wisdom, public virtue, and devo-
tion to lives of public service. In the months and weeks
before the Constitutional Convention assembled on
Monday the 14th of May, A.D. 1787. [sic] and in the
eleventh year of the independence of the United States
of America, at the State-House in the city of Philadel-
02dahl.007_040 11/27/01 4:39 PM Page 8
8 w h at t h e f r a m e r s c o u l d n t k n ow
phia,1 Madison studied the best sources as carefully

as a top student preparing for a major exam.2 But even
James Madison could not foresee the future of the
American republic, nor could he draw on knowledge
that might be gained from later experiences with de-
mocracy in America and elsewhere.
It is no detraction from the genius of Leonardo
da Vinci to say that given the knowledge available in
his time he could not possibly have designed a work-
able airplanemuch less the spacecraft that now
bears his name. Nor, given the knowledge available in
1903, could the Wright brothers have built the Boeing
707. Although like many others I greatly admire Ben-
jamin Franklin, I recognize that his knowledge of elec-
tricity was infinitesimal compared with that of a first-
year student in electrical engineeringor, for that
matter, the electrician who takes care of my occasional
wiring problems. In fact, on that famous first experi-
ment with the kite, Franklin was lucky to have escaped
alive. None of us, I expect, would hire an electrician
equipped only with Franklins knowledge to do our
wiring, nor would we propose to make a trip from
New York to London in the Wright brothers aircraft.
Leonardo, Franklin, the Wright brothers were great
innovators in their time, but they could not draw on
knowledge that was still to be accumulated in the
years and centuries to come.
The knowledge of the Framerssome of them,
certainlymay well have been the best available in
02dahl.007_040 11/27/01 4:39 PM Page 9
w h at t h e f r a m e r s c o u l d n t k n ow 9
1787. But reliable knowledge about constitutions

appropriate to a large representative republic was, at
best, meager. History had produced no truly relevant
models of representative government on the scale
the United States had already attained, not to mention
the scale it would reach in the years to come. As
much as many of the delegates admired the British
constitution, it was far from a suitable model. Nor
could the Roman Republic provide much of a guide.
The famous Venetian Republic, illustrious though it
had been, was governed by a hereditary aristocracy of
fewer than two thousand men and was already totter-
ing: a decade after the Convention an upstart Corsican
would knock it over in a featherweight military attack.
Whatever knowledge the delegates could gain from
historical experience was, then, only marginally rele-
vant at best.
Leaping into the Unknown
Among the important aspects of an unforeseeable fu-

ture, four broad historical developments would yield
some potential knowledge that the Framers necessar-
ily lacked and that, had they possessed it, might well
have led them to a different constitutional design.
First, a peaceful democratic revolution was soon to
alter fundamentally the conditions under which their
constitutional system would function.
02dahl.007_040 11/27/01 4:39 PM Page 10
Second, partly in response to that continuing revo-

lution, new democratic political institutions would fun-
damentally alter and reconstruct the framework they
had so carefully designed.
Third, when democratization unfolded in Europe
and in other English-speaking countries during the two
centuries to come, constitutional arrangements would
arise that were radically different from the American
system. Within a generation or two, even the British
constitution would bear little resemblance to the one
the Framers knewor thought they knewand in
many respects admired and hoped to imitate.
Fourth, ideas and beliefs about what democracy re-
quires, and thus what a democratic republic requires,
would continue to evolve down to the present day and
probably beyond. Both in the way we understand the
meaning of democracy and in the practices and insti-
tutions we regard as necessary to it, democracy is not
a static system. Democratic ideas and institutions as
they unfolded in the two centuries after the American
Constitutional Convention would go far beyond the
conceptions of the Framers and would even transcend
the views of such early democrats as Jefferson and
Madison, who helped to initiate moves toward a more
democratic republic.
I shall consider each of these developments in
later chapters. But first I want to indicate some of the
practical limitations on what the Framers could rea-
sonably achieve.
02dahl.007_040 11/27/01 4:39 PM Page 11
11
What the Framers Couldnt Do
The Framers were not only limited by, so to speak,

their inevitable ignorance. They were also crucially
limited by the opportunities available to them.
We can be profoundly grateful for one crucial re-
striction: the Framers were limited to considering only
a republican form of government. They were con-
strained not only by their own belief in the superiority
of a republican government over all others but also by
their conviction that the high value they placed on re-
publicanism was overwhelmingly shared by American
citizens in all the states. Whatever else the Framers
might be free to do, they well knew that they could not
possibly propose a monarchy or a government ruled by
an aristocracy. As the Massachusetts delegate Elbridge
Gerry put it, There was not a one-thousand part of our
fellow citizens who were not against every approach to-
ward monarchy.3 The only delegate who was recorded
by Madison4 as looking with favor on monarchy was
Alexander Hamilton, whose injudicious expression of
support for that heartily unpopular institution may have
greatly reduced his influence at the Convention, as it
was to haunt him later.5 Hardly more acceptable was an
adaptation of aristocratic ideas to an American constitu-
tion. During the deliberations about the Senate, Gou-
verneur Morris of Pennsylvania explored the possibility
of drawing its members from an American equivalent
of the British aristocracy.6 But it soon became obvious
02dahl.007_040 11/27/01 4:39 PM Page 12
that the delegates could not agree on just who these

American aristocrats might be, and in any case they
well knew that the overwhelming bulk of American cit-
izens would simply not tolerate such a government.
A second immovable limit was the existence of the
thirteen states, with still more states to come. A consti-
tutional solution that would be available in most of the
countries that were to develop into mature and stable
democraciesa unitary system with exclusive sover-
eignty lodged in the central government, as in Britain
and Sweden, for examplewas simply out of the ques-
tion. The need for a federal rather than a unitary re-
public was therefore not justified by a principle ad-
duced from general historical experience, much less
from political theory. It was just a self-evident fact. If
Americans were to be united in a single country, it was
obvious to all that a federal or confederal system was
inescapable. Whether the states would remain as fun-
damental constituents was therefore never a serious
issue at the Convention; the only contested question
was just how much autonomy, if any, they would yield
to the central government.7
The delegates had to confront still another stub-
born limit: the need to engage in fundamental com-
promises in order to secure agreement on any consti-
tution at all. The necessity for compromise and the
opportunities this gave for coalitions and logrolling
meant that the Constitution could not possibly reflect
a coherent, unified theory of government. Compro-
mises were necessary because, like the country at large,
02dahl.007_040 11/27/01 4:39 PM Page 13
members of the convention held different views on

some very basic issues.
Slavery. One, of course, was the future of slav-
ery. Most of the delegates from the five southern states
were adamantly opposed to any constitutional provi-
sion that might endanger the institution. Although the
delegates from the other seven states were hardly of
one mind about slavery, it was perfectly obvious to them
that the only condition on which coexistence would be
acceptable to the delegates from the southern states
would be the preservation of slavery. Consequently, if
these delegates wanted a federal constitution they
would have to yield, no matter what their beliefs about
slavery. And so they did. Although some delegates who
signed the final document abhorred slavery, they nev-
ertheless accepted its continuation as the price of a
stronger federal government.
Representation in the Senate. Another conflict of
views that could not be settled without a one-sided
compromise resulted from the adamant refusal of the
delegates from the small states to accept any constitu-
tion that did not provide for equal representation in
the Senate. The opponents of equal representation in-
cluded two of the most illustrious members of the
Convention, James Madison and James Wilson, who
were also among the chief architects of the Constitu-
tion. Both men bitterly opposed what seemed to them
an arbitrary, unnecessary, and unjustifiable limit on
national majorities. As Alexander Hamilton remarked
about this conflict: As states are a collection of indi-
02dahl.007_040 11/27/01 4:39 PM Page 14
vidual men which ought we to respect most, the rights

of the people composing them, or the artificial beings
resulting from the composition. Nothing could be more
preposterous or absurd than to sacrifice the former to
the latter. It has been sd. that if the smaller States re-
nounce their equality, they renounce at the same time
their liberty. The truth is it is a contest for power, not
for liberty. Will the men composing the small States be
less free than those composing the larger.8
Let me give you a flavor of the elevated discussion
that preceded the victory of the small states. Here is
Gunning Bedford of Delaware on June 30:
The large states dare not dissolve the Confederation. If

they do the small ones will find some foreign ally of
more honor and good faith, who will take them by the
hand and do them justice.
To which Rufus King of Massachusetts replied:
I cannot sit down, without taking some notice of the lan-

guage of the honorable gentleman from Delaware. . . .
It was not I who with a vehemence unprecedented in
this House, declared himself ready to turn his hopes
from our common Country, and court the protection of
some foreign hand. . . . I am grieved that such a thought
has entered into his heart. . . . For myself whatever
might be my distress, I would never court relief from a
foreign power.9
Faced with the refusal of the small states to ac-

cept anything less, Madison, Wilson, Hamilton, and
the other opponents of equal representation finally ac-
02dahl.007_040 11/27/01 4:39 PM Page 15
cepted compromise of principle as the price of a con-

stitution. The solution of equal representation was not,
then, a product of constitutional theory, high prin-
ciple, or grand design. It was nothing more than a
practical outcome of a hard bargain that its opponents
finally agreed to in order to achieve a constitution.10
Incidentally, this conflict illustrates some of the
complexities of voting coalitions at the Constitutional
Convention, for the faction opposed to equal repre-
sentation in the Senate included four strange bed-
fellows: Madison, Wilson, Hamilton, and Gouverneur
Morris. Although all four generally supported moves
to strengthen the federal government, Madison and
Wilson usually endorsed proposals that leaned toward
a more democratic republic, while Hamilton and Mor-
ris tended to support a more aristocratic republic.
Undemocratic Elements in the

Framers Constitution
It was within these limits, then, that the Framers con-

structed the Constitution. Not surprisingly, it fell far
short of the requirements that later generations would
find necessary and desirable in a democratic republic.
Judged from later, more democratic perspectives, the
Constitution of the Framers contained at least seven
important shortcomings.
Slavery. First, it neither forbade slavery nor em-
powered Congress to do so. In fact, the compromise
02dahl.007_040 11/27/01 4:39 PM Page 16
on slavery not only denied Congress the effective

power to prohibit the importation of slaves before
180811 but it gave constitutional sanction to one of the
most morally objectionable byproducts of a morally re-
pulsive institution: the Fugitive Slave laws, according
to which a slave who managed to escape to a free state
had to be returned to the slaveholder, whose property
the slave remained.12 That it took three-quarters of a
century and a sanguinary civil war before slavery was
abolished should at the least make us doubt whether
the document of the Framers ought to be regarded as
holy writ.
Suffrage. Second, the constitution failed to guar-
antee the right of suffrage, leaving the qualifications
of suffrage to the states.13 It implicitly left in place
the exclusion of half the populationwomenas well
as African Americans and Native Americans.14 As we
know, it took a century and a half before women were
constitutionally guaranteed the right to vote, and nearly
two centuries before a president and Congress could
overcome the effective veto of a minority of states in
order to pass legislation intended to guarantee the vot-
ing rights of African Americans.
Election of the president. Third, the executive
power was vested in a president whose selection, ac-
cording to the intentions and design of the Framers,
was to be insulated from both popular majorities and
congressional control. As well see, the Framers main
design for achieving that purposea body of presi-
dential electors composed of men of exceptional wis-
02dahl.007_040 11/27/01 4:39 PM Page 17
dom and virtue who would choose the chief executive

unswayed by popular opinionwas almost immedi-
ately cast into the dustbin of history by leaders sympa-
thetic with the growing democratic impulses of the
American people, among them James Madison him-
self. Probably nothing the Framers did illustrates more
sharply their inability to foresee the shape that politics
would assume in a democratic republic. (I shall say
more about the electoral college in a later chapter.)
Choosing senators. Fourth, senators were to be
chosen not by the people but by the state legislatures,
for a term of six years.15 Although this arrangement
fell short of the ambitions of delegates like Gouver-
neur Morris who wanted to construct an aristocratic
upper house, it would help to ensure that senators
would be less responsive to popular majorities and
perhaps more sensitive to the needs of property hold-
ers. Members of the Senate would thus serve as a
check on the Representatives, who were all subject to
popular elections every two years.16
Equal representation in the Senate. The attempt
to create a Senate that would be a republican version
of the aristocratic House of Lords was derailed, as we
have seen, by a prolonged and bitter dispute over an
entirely different question: Should the states be equally
represented in Congress or should members of both
houses be allocated according to population? This ques-
tion not only gave rise to one of the most disruptive is-
sues of the Convention, but it resulted in a fifth unde-
mocratic feature of the constitution. As a consequence
02dahl.007_040 11/27/01 4:39 PM Page 18
of the famousor from a democratic point of view, in-

famousConnecticut Compromise each state was,
as we have seen, awarded the same number of sena-
tors, without respect to population. Although this
arrangement failed to protect the fundamental rights
and interests of the most deprived minorities, some
strategically placed and highly privileged minorities
slaveholders, for examplegained disproportionate
power over government polices at the expense of less
privileged minorities. (I shall come back to this ele-
ment in the constitution in a later chapter.)
Judicial power. Sixth, the constitution of the
Framers failed to limit the powers of the judiciary to
declare as unconstitutional laws that had been prop-
erly passed by Congress and signed by the president.
What the delegates intended in the way of judicial re-
view will remain forever unclear; probably many dele-
gates were unclear in their own minds, and to the ex-
tent that they discussed the question at all, they were
not in full agreement. But probably a majority ac-
cepted the view that the federal courts should rule on
the constitutionality of state and federal laws in cases
brought before them. Nevertheless, it is likely that
a substantial majority intended that federal judges
should not participate in making government laws and
policies, a responsibility that clearly belonged not to
the judiciary but to the legislative branch. Their oppo-
sition to any policy-making role for the judiciary is
strongly indicated by their response to a proposal in
the Virginia Plan that the Executive and a convenient
02dahl.007_040 11/27/01 4:39 PM Page 19
number of the National Judiciary, ought to compose a

council of revision empowered to veto acts of the Na-
tional Legislature. Though this provision was vigor-
ously defended by Madison and Mason, it was voted
down, 6 states to 3.17
A judicial veto is one thing; judicial legislation is
quite another. Whatever some of the delegates may
have thought about the advisability of justices sharing
with the executive the authority to veto laws passed by
Congress, I am fairly certain that none would have
given the slightest support to a proposal that judges
should themselves have the power to legislate, to
make national policy. However, the upshot of their
work was that in the guise of reviewing the constitu-
tionality of state and congressional actions or inac-
tions, the federal judiciary would later engage in what
in some instances could only be called judicial policy-
makingor, if you like, judicial legislation.18
Congressional power. Finally, the powers of Con-
gress were limited in ways that could, and at times did,
prevent the federal government from regulating or
controlling the economy by means that all modern
democratic governments have adopted. Without the
power to tax incomes, for example, fiscal policy, not to
say measures like Social Security, would be impos-
sible. And regulatory actionsover railroad rates, air
safety, food and drugs, banking, minimum wages, and
many other policieshad no clear constitutional au-
thorization. Although it would be anachronistic to
charge the Framers with lack of foresight in these
02dahl.007_040 11/27/01 4:39 PM Page 20
matters,19 unless the constitution could be altered by

amendment or by heroic reinterpretation of its provi-
sionspresumably by what I have just called judicial
legislationit would prevent representatives of later
majorities from adopting the policies they believed
were necessary to achieve efficiency, fairness, and se-
curity in a complex post-agrarian society.
Enlightened as the Framers constitution may have
been by the standards of the eighteenth century, fu-
ture generations with more democratic aspirations
would find some of its undemocratic features objec-
tionableand even unacceptable. The public expres-
sion of these growing democratic aspirations was not
long in coming.
Even Madison did not, and probably could not,
predict the peaceful democratic revolution that was
about to begin. For the American revolution was soon
to enter into a new and unforeseen phase.
The Framers Constitution Meets

Emergent Democratic Beliefs
We may tend to think of the American republic and its

constitution as solely the product of leaders inspired
by extraordinary wisdom and virtue. Yet without a citi-
zenry committed to republican principles of govern-
ment and capable of governing themselves in accor-
dance with those principles, the constitution would
soon have been little more than a piece of paper. As
02dahl.007_040 11/27/01 4:39 PM Page 21
historical experience would reveal, in countries where

democratic beliefs were fragile or absent, constitutions
did indeed become little more than pieces of paper
soon violated, soon forgotten.
The American democratic republic was not cre-
ated nor could it have been long maintained by lead-
ers alone, gifted as they may have been. It was they, to
be sure, who designed a framework suitable, as they
thought, for a republic. But it was the American people,
and the leaders responsive to them, who ensured that
the new republic would rapidly become a democratic
republic.
The proto-republican phase. The ideas, practices,
and political culture necessary to sustain a republican
government were by no means unfamiliar to Ameri-
cans. Unlike some countries that have moved almost
overnight from dictatorship to democratic forms, and
often soon thereafter to chaos and back to dictator-
ship, by 1787 the Americans had already accumulated
a century and a half of experience in the arts of gov-
ernment.
The long colonial period had provided opportuni-
ties to both leaders and many men of ordinary rank
to become acquainted with the requirements of self-
government, both in the direct form of a town meeting
and through electing representatives to the colonial
legislatures.20 We easily forget that although in its two
famous opening paragraphs the Declaration of Inde-
pendence laid down some new and audacious claims,
in the rest of that documentthe part few people
02dahl.007_040 11/27/01 4:39 PM Page 22
bother to read todaythe authors mainly protested

against the British king for violating rights that, with
some exaggeration, they had previously enjoyed as
Englishmen.
The republican phase. The next phase, creating a
popular republic, had begun with the astounding dec-
laration on July 4, 1776, that all Men are created
equal. The Declaration marks the beginning of a se-
ries of events that went much further than simply
gaining independence from Britain. In what the histo-
rian Gordon Wood has called the greatest Utopian
movement in American history,21 the Declaration also
triggered a democratic revolution in beliefs, practices,
and institutionsor better, an evolutionthat has con-
tinued ever since. The two decades since independ-
ence had provided still more, and deeper, experience
in the practices of self-government. Nor was this expe-
rience limited to a tiny minority. In some of the thir-
teen states, a fairly high proportion of adult males had
acquired the franchise.22
Toward a democratic republic. The lengthy colo-
nial and post-independence experience provided a
sturdy foundation for the efforts that Americans now
undertook in the next phase of the revolution, when
the new republic was transformed into a more demo-
cratic republic. To be sure, at the end of the eigh-
teenth century few Americans were ready to concede
that the principles of the Declaration, much less dem-
ocratic citizenship, applied to everyone.23 It would take
two more centuries of evolution in democratic beliefs
02dahl.007_040 11/27/01 4:39 PM Page 23
before most Americans would be inclined to agree

that the famous claim in the Declaration might be re-
phrased: not just all men, but all persons are cre-
ated equal.
Yet always keeping in mind the huge and persistent
exceptions, by the standards prevailing elsewhere in
the world the extent of equality among Americans was
extraordinary. Alexis de Tocqueville, who observed
Americans during his years visit in 183132, opened
his famous work with these words:
Among the novel objects that attracted my attention
during my stay in the United States, nothing struck me
more forcibly than the general equality of conditions. I
readily discovered the prodigious influence which this
primary fact exercises on the whole course of society, by
giving a certain direction to public opinion, and a cer-
tain tenor to the laws; by imparting new maxims to the
governing powers, and peculiar habits to the governed.
I speedily perceived that the influence of this fact
extends far beyond the political character and the laws
of the country, and that it has no less empire over civil
society than over the Government. . . .
The more I advanced in the study of American so-
ciety, the more I perceived that the equality of condi-
tion is the fundamental fact from which all others seem
to be derived, and the central point at which all my ob-
servations constantly terminated.24
During the three decades before Tocqueville ar-

rived, under the leadership of Jefferson, Madison, and
others, supporters of a more democratic republic had
already made some changes. The seismic shift from
02dahl.007_040 11/27/01 4:39 PM Page 24
the views of the Framers and the Federalists is sym-

bolized by the changing name of the party that won
both the presidency and Congress in the election that
Jefferson calledas have later historiansthe Revo-
lution of 1800. To defeat the Federalists, win the elec-
tion, and gain control of the new government, Jeffer-
son and Madison had created a political party that
they appropriately named the Democratic-Republican
Party. By 1832, with Andrew Jackson as its winning
candidate, the Democratic-Republican party became
the Democratic Party, plain and simple.25 The name
has stuck ever since.
Conservative delegates among the Framerslater
the core of the Federalist Partyhad feared that if or-
dinary people were given ready access to power they
would bring about policies contrary to the views and in-
terests of the more privileged classes, which, as the
conservative delegates viewed their interests, were also
the best interests of the country. These conservative
fears were soon confirmed. Within a decade the emi-
nent Federalist leaders were pushed aside and the Fed-
eral Party became a minority party. A generation later
had seen the demise of both the party and its leaders.
If these changes justified some of the pessimism
about popular majorities of many of the Framers, their
pessimism proved unjustified in another important re-
spect. A substantial number of the Framers believed
that they must erect constitutional barriers to popular
rule because the people would prove to be an unruly
mob, a standing danger to law, to orderly government,
02dahl.007_040 11/27/01 4:39 PM Page 25
and to property rights. Contrary to these pessimistic

appraisals, when American citizens were endowed with
the rights and opportunities to support demagogues
and rabble rousers, they chose instead to support law,
orderly government, and property rights. White male
Americans were, after all, mainly farmers who owned
their own land; or, where farm land was not easily
available because most of it had already been occu-
pied, they could count on the ready availability of good
farm land farther westoften obtained, to be sure,
at the expense of its earlier inhabitants, the Native
Americans.
White Americans in vast numbers bought western
land and settled down on their own farms. Two-thirds
of the landless white men of Virginia moved West in
the 1790s. . . . Between 1800 and 1820, the trans-
Appalachian population grew from a third of a million
to more than two million.26 In foreseeing a democratic
republic based on a citizen body consisting predomi-
nantly of independent farmers, mainly property owners
cultivating their own lands, Jefferson reflected the re-
ality of his time.27 Outside the South, and even in the
southern piedmont, a predominant number of Ameri-
can citizens were free farmers who stood to benefit
from an orderly government dependent on their votes.
Ordinary citizens also revealed strong beliefs in
democratic values and procedures. Presented with the
opportunity to do so, they would choose leaders who
cultivated democratic values and procedures. Just such
an opportunity was soon presented by four acts passed
02dahl.007_040 11/27/01 4:39 PM Page 26
in 1798 by the Federalists, who were alarmed not only

by the seemingly subversive activities of France but
also by the rapidly growing influence of boisterous, ir-
reverent, and sometimes libelous opponents in the
new Republican party. In particular, the Federalists
employed one of these new laws, the Sedition Act, in
an effort to silence Republican critics. Notable among
the fourteen who were prosecuted was a bombastic and
somewhat unsavory Republican congressman, the Irish
immigrant Mathew Lyon, whose only memorable con-
tribution to American history was his conviction for
sedition, which carried a fine of a thousand dollarsa
huge amount in those daysand four months in jail.28
To the Republicans, the Sedition Act was a flagrant vio-
lation of the newly adopted First Amendment. After
they gained the presidency and control of Congress in
the election of 1800, the Sedition Act was allowed to
lapse, despite the vigorous efforts of the Federalists.
Democratic Changes to the

Framers Constitution: Amendments
The fate of the Alien and Sedition Acts symbolizes a

larger change at work in the country. The democratic
revolution, fitful and uncertain though it would for-
ever remain, not only helped to democratize the for-
mal constitution itself by amendments, it generated
new democratic political institutions and practices
within which the constitutional system would operate.
02dahl.007_040 11/27/01 4:39 PM Page 27
The constitutional system that has emerged is no

longer that of the Framers, nor is it one they had in-
tended to create.
The Bill of Rights. To be sure, the first ten amend-
ments to the Constitutionthe Bill of Rightscannot
be attributed to the democratic revolution that fol-
lowed the Convention. They resulted instead from de-
mands within the Convention itself by delegates who
generally favored a more democratic system than their
colleagues could then accept. Among the most influen-
tial of these was George Mason, who wrote the Virginia
constitution and its Declaration of Rights. Responding
to the insistent demands of Mason and several others,
as well as to similar voices outside the Convention,
Masons fellow Virginian, James Madison, drafted ten
amendments that were ratified in 178990 by eleven
states, more than a sufficient number for their adop-
tion. (Incidentally, the two laggards, Georgia and Con-
necticut, finally did come aroundbut not until 1939!)
Thus, for all practical purposes the Bill of Rights was
a part of the original constitution. In any case, the
amendments have proved to be a veritable cornucopia
of expanding rights necessary to a democratic order.29
Other Amendments
As I have mentioned, the most profound violation of

human rights permitted by the original constitution,
slavery, was not corrected until the adoption of the
02dahl.007_040 11/27/01 4:39 PM Page 28
Thirteenth, Fourteenth, and Fifteenth Amendments

between 1865 and 1870. In 1909 the Sixteenth Amend-
ment in 1913 gave Congress the power to enact in-
come taxes. The election of U.S. senators by state legis-
latures finally gave way to direct election with the
adoption of the Seventeenth Amendment in 1913.
Women were finally guaranteed the right of suffrage in
federal and state elections with the passage of the
Nineteenth Amendment in 1919. Although the effort
to add an Equal Rights Amendment failed, the Four-
teenth Amendment was later interpreted to provide
a constitutional basis for eliminating discrimination
against women as well as certain minorities whose
members suffered from discriminatory practices. The
iniquitous poll tax that had continued to bar African
Americans from voting in some southern states was fi-
nally forbidden in 1964 by the Twenty-Fourth Amend-
ment. Finally, in a move toward a more inclusive elec-
torate, in 1971 the Twenty-Sixth Amendment reduced
the voting age to eighteen.
In this halting fashion, the democratic revolution
belatedly worked its way through the Constitution to
overcome the veto power of long-entrenched minorities
and to eliminate some of the most flagrantly undemo-
cratic features of the constitution. As Alan Grimes ob-
served some years ago, of the twenty-six (now twenty-
seven) amendments to the constitution, Twenty-one
amendments may be said to affirm either the principle
of democratic rights or that of democratic processes.30
02dahl.007_040 11/27/01 4:39 PM Page 29
29
Democratic Changes in
Political Practices and Institutions
The constitution of the Framers was changed not only

by formal amendments. It was also fundamentally al-
tered by political practices and institutions that the
Framers did not foresee, even though they were un-
avoidableindeed, highly desirablein a democratic
republic.
Political parties. Perhaps the most important of
these was the political party. The Framers feared and
detested factions, a view famously expressed by Madison
in Federalist No. 10.31 Probably no statement has been
so often cited to explain and justify the checks against
popular majorities that the Framers attempted to build
into the constitution. It is supremely ironic, therefore,
that more than anyone except Jefferson, it was Madison
who helped to create the Republican Party in order to
defeat the Federalists. Although the system would not
settle down for some years, Jefferson and Madison
helped to inaugurate the competitive two-party system
that has pretty much remained in place ever since.
Which suggests other questions. Despite the claim
of every political party everywhere in the world that it
truly represents the general interest, arent political
parties really factions in Madisons sense? So did the
Framers fail after all to prevent government by fac-
tions? And did they succeed only in making it more
difficult for a majority faction to prevailthat is, a
party reflecting the interests of a majority coalition?
02dahl.007_040 11/27/01 4:39 PM Page 30
Whatever the best answers to these hard ques-

tions, it cannot be denied that partisan politics trans-
formed the constitution. Despite their familiarity with
the role of the Tories and Whigs in Britain and nascent
parties in their own legislatures, the Framers did not
fully foresee that in a democratic republic political
parties are not only possible, they are also inevitable
and desirable. As Jefferson and Madison soon came to
realize, without an organized political party to mobi-
lize their voters in the states and their fellow support-
ers in the Congress, they could not possibly overcome
the entrenched political domination of their political
adversaries, the Federalists. The democratic rights in-
corporated in the Bill of Rights made parties possible;
the need to compete effectively made them inevitable;
the ability to represent citizens who would otherwise
not be adequately represented made them desirable.
Today we take for granted that political parties and
party competition are essential to representative de-
mocracy: we can be pretty sure that a country wholly
without competitive parties is a country without democ-
racy. If the Framers had been aware of the central im-
portance of political parties to a democratic republic,
would they have designed their constitution differently?
They might well have. At the very least they would not
have created the absurdity of an electoral college.
The electoral college. In an outcome the Framers
had made possible by their defective design of the
electoral college, the election of 1800 produced a tie
between Jefferson and his running mate, Aaron Burr.
02dahl.007_040 11/27/01 4:39 PM Page 31
From the time the final results were known in late

December 1800, the deadlock in the electoral college
persisted, despite many attempts at persuasion and
compromise, until February 17, 1801, when shifts and
abstentions by a number of state delegations gave Jef-
ferson the presidency.32 Ironically, the very institution
that the Framers hoped would insulate the election of
the president from partisan politics was its first victim.
Although a similar fiasco was prevented in the future
by the Twelfth Amendment in 1804, even with the
amendment the electoral college was converted by par-
tisan politics into nothing more than a rather peculiar
and ritualized way of allocating the votes of the states
for president and vice president. Yet the electoral col-
lege still preserved features that openly violated basic
democratic principles: citizens of different states would
be unequally represented, and a candidate with the
largest number of popular votes might lose the presi-
dency because of a failure to win a majority in the elec-
toral college. That this outcome was more than a theo-
retical possibility had already occurred three times
before it was displayed for all the world to see in the
election of 2000. Ill come back to the democratic
shortcomings of the electoral college in a later chapter.
The Democratic Revolution:

What Madison Learnedand Taught
James Madison arrived in Philadelphia in 1787, a few
months past his thirty-sixth birthday. He was already
02dahl.007_040 11/27/01 4:39 PM Page 32
far from a political neophyte, having been elected at

the age of twenty-five to the Virginia constitutional
convention where, with George Mason, he helped to
draft the Virginia Declaration of Rights and the new
state constitution. He then became successively a mem-
ber of the Virginia legislature (though he failed to be
reelected because, it was said, he refused to treat the
voters to the customary rum punch), a delegate to the
Continental Congress, and again a member of the Vir-
ginia legislature. In the months before the Constitu-
tional Convention opened, he drafted the outline of
the proposal that would be presented in the opening
days of the Convention and that would come to be
known as the Virginia Plan. (We shall see something of
its contents in the next chapter.)
Yet, experienced as he was, like his fellow delegates
Madison brought to the Convention limited knowl-
edge of the institutions and practices that a more fully
democratized republic would require. Before his death
in 1836 at the age of eighty-five, nearly half a century
after the Convention, Madison could have looked back
on a rich body of experience that would have shaped
his constitutional views in many ways.
Following the Convention, he was elected to the
U.S. House of Representatives where he drafted and
introduced the first ten amendments to the Constitu-
tionthe Bill of Rights. With Jefferson he soon be-
came a leader of the opposition to Federalist policies
and ideas. As we have seen, they formed and led the
02dahl.007_040 11/27/01 4:39 PM Page 33
opposition party, the Democratic Republicans. After

Jeffersons election, Madison became secretary of
state. He then succeeded Jefferson in the presidency.
By the time he left that office in 1817, his views about
democratic political institutions were probably as well
informed as those of any person then alive.
However that may be, the Madison of seventy in
1821 was no longer the Madison of thirty-six in 1787.
Among other changes, the Madison of 1821 would
have trusted popular majoritiesAmerican popular
majorities, anywayfar more than the Madison of
1787. The mature and experienced Madison of 1821
might therefore have done less to check majority rule
and more to facilitate it. Let me offer several pieces of
evidence, one from a time early in his awakening to
the requirements of a democratic republic, the others
from his reflections in old age.
I have already alluded to the first: the basic alter-
ation in his views about factions, or what the two
distinguished historians of Federalism describe as
Madison Revises The Federalist.33 Madisons views in
Federalist No. 10, influenced by his reading of David
Hume, are cited endlessly: the dangers of factions, the
threat from majorities united on principles contrary to
the general interest, political parties as at best a neces-
sary evil. But these were not his more mature views.
In January 1792, less than five years after the close
of the Convention, Madison begins to publish a series
of essays in The Gazette, an opposition newspaper
02dahl.007_040 11/27/01 4:39 PM Page 34
published by Philip Freneau. The first is entitled On

Parties. In every political society, he writes, parties
are unavoidable. To combat their dangers, Madison
offers five proposals that might well serve us better in
our own time than the anti-majoritarian biases dis-
played in Federalist No. 10. Whatever dangers politi-
cal parties may pose can be overcome
By establishing political equality among all.
By withholding unnecessary opportunities from a few,
to increase the inequality of property by an immoder-
ate, and especially unmerited, accumulation of riches.
By the silent operation of the laws, which, without vio-
lating the rights of property, reduce extreme wealth to-
wards a state of mediocrity, and raise extreme indigence
toward a state of comfort.
By abstaining from measures which operate differently
on different interests, and particularly favor one inter-
est, at the expense of another.
By making one party a check on the other, so far as the
existence of parties cannot be prevented, nor their
views accommodated.34
If this is not the language of reason, he went on to

say, it is that of republicanism.
Nearly thirty years later (around 1821), when he is
preparing his notes on the constitutional debates for
publication, he records some of his later reflections.
As to the right of suffrage, he remarks that his obser-
vations at the Convention do not convey the speakers
[Madisons] more full and matured view of the sub-
02dahl.007_040 11/27/01 4:39 PM Page 35
ject. The right of suffrage, he now insists, is a fun-

damental Article in Republican Constitutions. He
also makes explicit his view of political parties: No
free Country, he says, has ever been without parties,
which are a natural offspring of Freedom. But politi-
cal parties and a broad suffrage may create a conflict
over property. An obvious and permanent division of
every people is into the owners of the Soil, and the
other inhabitants. Consequently, if the suffrage is ex-
tended to citizens who are not freeholders, a majority
might threaten the property rights of the freeholders.
Madison then considers a number of possible solu-
tions to this problem, of which the first would be to re-
strict the suffrage to freeholders, and to such as hold
an equivalent property. He rejects this solution with
an observation that might well have been a central
principle of the Second Phase of the American Revo-
lution. The objection to this regulation, he writes, is
obvious. It violates the vital principle of free Govt. that
those who are to be bound by laws, ought to have a
voice in making them. And the violation wd. be more
strikingly unjust as the lawmakers became the minor-
ity. A second option is confining the right of suffrage
for one branch to the holders of property, and for the
other Branch to those without property. But to do so
wd. not in fact be either equal or fair. Nor prudent:
The division of the State into the two Classes . . .
might lead to contests & antipathies not dissimilar to
those between the Patricians and Plebeians at Rome.
After examining other possibilities, he concludes:
02dahl.007_040 11/27/01 4:39 PM Page 36
Under every view of the subject, it seems indispensable

that the Mass of Citizens not be without a voice, in
making the laws which they are to obey, & in chusing
the Magistrates, who are to administer them, and if the
only alternative be between an equal & universal right
of suffrage for each branch of the Govt. and a confine-
ment of the entire right to a part of the Citizens, it is
better that those having the greater interest at stake
namely that of property & persons both, should be de-
prived of half their share in the Govt. than, that those
having the lesser interest, that of personal rights only,
should be deprived of the whole.35
The older Madison is also more favorable to ma-

jority rule. Like most of his contemporaries, Madison
believes that all power in human hands is liable to be
abused. But taking that assumption as axiomatic to-
gether with the need for government, the relevant
question becomes: what kind of government is best?
His answer remains unchanged:
In Governments independent of the people, the rights
and views of the whole may be sacrificed to the views of
the Government. In Republics, where the people gov-
ern themselves, and where, of course, the majority gov-
ern, a danger to the minority arises from opportunities
tempting a sacrifice of their rights to the interest, real
or supposed, of a majority. No form of government,
therefore, can be a perfect guard against the abuse of
power. The recommendation of the republican form is,
that the danger of abuse is less than any other.36
What has changed is his greater confidence in ma-

jority rule. Compared with its alternatives at least, the
02dahl.007_040 11/27/01 4:39 PM Page 37
mature Madison is confident that majority rule, in the

words of Marvin Meyers, promises the least imper-
fect government.37
[E]very friend to Republican Government, he
writes in 1833, ought to raise his voice against the
sweeping denunciation of majority Governments as the
most tyrannical and intolerable of all Governments.
It has been said that all Government is an evil. It would
be more proper to say that the necessity of any govern-
ment is a misfortune. This necessity however exists; and
the problem to be solved is, not what form of govern-
ment is perfect, but which of the forms is least imper-
fect; and here the general question must be between a
republican Government in which the majority rule the
minority, and a government in which a lesser number or
the least number rule the majority.
The result . . . is, that we must refer to the moni-
tory reflection that no government of human device
and human administration can be perfect; that that
which is the least imperfect is therefore the best gov-
ernment; that the abused of all other governments have
led to the preference of republican government as the
best of all governments, because the least imperfect;
that the vital principle of republican government is the
lex majoris parties, the will of the majority.38
! ! !
I HAVE LITTLE DOUBT THAT IF THE AMERICAN CONSTI-

tutional Convention had been held in 1820, a very dif-
ferent constitution would have emerged from the de-
liberationsalthough, I hasten to add, we can never
02dahl.007_040 11/27/01 4:39 PM Page 38
know what shape that constitution might have taken.

We can be reasonably sure, however, that the dele-
gates would have attempted to provide more support
for, and fewer barriers to, a democratic republic.
As to the undemocratic features of the constitution
created in 1787, let me suggest four conclusions.
First, the aspects of the constitution that are most
defective from a democratic point of view do not nec-
essarily all reflect the intentions of the Framers, inso-
far as we may surmise them. Though the flaws are
traceable to their handiwork, they are in some cases
flaws resulting from the inability of these superbly tal-
ented craftsmen to foresee how their carefully crafted
instrument of government would work under the chang-
ing conditions that were to followand most of all,
under the impact of the democratic revolution in which
Americans were, and I hope still are, engaged.
Second, some of the undemocratic aspects of the
original design also resulted from the logrolling and
compromises that were necessary to achieve agree-
ment. The Framers were not philosophers searching
for a description of an ideal system. Norand we may
be forever grateful to them for thiswere they
philosopher kings entrusted with the power to rule.
They were practical men, eager to achieve a stronger
national government, and as practical men they made
compromises. Would the country have been better off
if they had refused to do so? I doubt it. But in any
case, they did compromise, and even today the consti-
tution bears the results of some of their concessions.
02dahl.007_040 11/27/01 4:39 PM Page 39
Ill have more to say on that point in my next chapter.

Third, undemocratic aspects that were more or
less deliberately built into the constitution overesti-
mated the dangers of popular majoritiesAmerican
popular majorities, at any rateand underestimated
the strength of the developing democratic commit-
ment among Americans. As a result, in order to adapt
the original framework more closely to the require-
ments of the emerging democratic republic, with the
passage of time some of these aspects of the original
constitution were changed, sometimes by amendment,
sometimes, as with political parties, by new institu-
tions and practices.
Finally, though the defects seem to me serious and
may grow even more serious with time, Americans are
not much predisposed to consider another constitu-
tion, nor is it clear what alternative arrangements
would serve them better.
As a result, the beliefs of Americans in the legiti-
macy of their constitution will remain, I think, in con-
stant tension with their beliefs in the legitimacy of
democracy.
For my part, I believe that the legitimacy of the
constitution ought to derive solely from its utility as an
instrument of democratic governmentnothing more,
nothing less. In my last chapter, Ill reflect further on
the meaning of that judgment.
02dahl.007_040 11/27/01 4:39 PM Page 40
03dahl.041_072 11/27/01 4:39 PM Page 41
chapter 3
The Constitution as a Model:

An American Illusion
M
ANY AMERICANS APPEAR TO BELIEVE THAT OUR
constitution has been a model for the rest of
the democratic world.1 Yet among the coun-
tries most comparable to the United States and where
democratic institutions have long existed without break-
down, not one has adopted our American constitutional
system. It would be fair to say that without a single ex-
ception they have all rejected it. Why?
Before I explore that question, I need to clarify two
matters. As you may have noticed, rather than speaking
simply of the constitution, Ive sometimes used the
phrase the constitutional system. I do so because I
want to include in a constitutional system an important
set of institutions that may or may not be prescribed in
the formal constitution itself: these are its electoral
03dahl.041_072 11/27/01 4:39 PM Page 42
42 t h e c o n st i t u t i o n a s a m o d e l
arrangements. As well see, electoral systems can in-

teract in crucial ways with the other political institu-
tions and thereby determine the way they function.
Also, Ive just referred to the countries where
democracy is oldest and most firmly established. We
could call them the older democracies, the mature
democracies, the stable democratic countries, and so
on, but Ill settle on the advanced democratic coun-
tries. Whatever we choose to call them, in order to
compare the characteristics and performance of the
American constitutional system with the characteris-
tics and performance of the systems in other demo-
cratic countries, we need a set of reasonably compa-
rable democratic countries. In short, we dont want to
compare apples and orangesor good apples and rot-
ten apples.
Ive noticed that we Americans often assure our-
selves of the superiority of our American political sys-
tem by comparing it with political systems in countries
ruled by nondemocratic regimes or in countries that
suffer from violent conflict, chronic corruption, fre-
quent chaos, regime collapse or overthrow, and the
like. On voicing or hearing criticism of political life in
the United States, an American not infrequently adds,
Yes, but just compare it with X!, a favorite X being
the Soviet Union during the Cold War and, after its
collapse, Russia. One could easily pick more than a
hundred other countries with political systems that by
almost any standard are unquestionably inferior to our
own. But comparisons like this are absurdly irrelevant.
03dahl.041_072 11/27/01 4:39 PM Page 43
t h e c o n st i t u t i o n a s a m o d e l 43
To my mind, the most comparable countries are

those in which the basic democratic political institu-
tions have functioned without interruption for a fairly
long time, lets say at least half a century, that is, since
1950. Including the United States, there are twenty-
two such countries in the world.2 (See Appendix B, Ta-
bles 1 and 2.) Fortunately for our purposes, they are
also comparable in their relevant social and economic
conditions: not a rotten apple in the bunch. Not sur-
prisingly, they are mostly European or English speak-
ing, with a few outliers: Costa Rica, the only Latin
American country; Israel, the only Middle Eastern
country; and Japan, the only Asian country.
When we examine some of the basic elements in
the constitutional structures of the advanced demo-
cratic countries, we can see just how unusual the Amer-
ican system is. Indeed, among the twenty-two older de-
mocracies, our system is unique.3
Federal or Unitary
To begin with, among the other twenty-one countries

we find only six federal systems, in which territorial
unitsstates, cantons, provinces, regions, Lnder
are endowed by constitutional prescription and prac-
tice with a substantial degree of autonomy and with
significant powers to enact legislation. As in the United
States, in these federal countries the basic territorial
units, whether states, provinces, or cantons, are not
03dahl.041_072 11/27/01 4:39 PM Page 44
simply legal creatures of the central government with

boundaries and powers that the central government
could, in principle, modify as it chooses. They are
basic elements in the constitutional design and in the
political life of the country.
As with the United States, so too in these other five
countries federalism was not so much a free choice as
a self-evident necessity imposed by history. In most,
the federal unitsstates, provinces, cantonsexisted
before the national government was fully democra-
tized. In the extreme case, Switzerland, the consti-
tuent units were already in place before the Swiss
Confederation itself was formed from three Alpine
cantons in 1291, five centuries before America was
born. Throughout the following seven centuries the
Swiss cantons, now twenty in number,4 have retained a
robust distinctiveness and autonomy. In the outlier,
Belgium, federalism followed long after a unitary gov-
ernment had been imposed on its diverse regional
groups. As the brilliant period of Flemish painting,
weaving, commerce, and prosperity in the sixteenth
and seventeenth centuries reminds us, profound terri-
torial, linguistic, religious, and cultural differences be-
tween the predominantly Flemish and Walloon areas
existed long before Belgium itself became an inde-
pendent country in 1830. Despite the persistent cleav-
ages between the Flemish and Walloons, however,
federalism did not arrive until 1993 when the three
regionsWallonia, Flanders, and Brusselswere fi-
nally given constitutional status. I should point out
03dahl.041_072 11/27/01 4:39 PM Page 45
that the deep divisions between Walloons and Flemish

continue to threaten the survival of Belgium as a single
country.
The second and third features follow directly from
the existence of federalism.
Strong Bicameralism
A natural, if not strictly necessary, consequence of fed-

eralism is a second chamber that provides special rep-
resentation for the federal units. To be sure, unitary
systems may also have, and historically all have had, a
second chamber. However, in a democratic country
with a unitary system, the functions of a second cham-
ber are far from obvious. The question that was posed
during the American constitutional convention is bound
to arise: Exactly whom or whose interests is a second
chamber supposed to represent? And just as the Fram-
ers could provide no rationally convincing answer, so
too as democratic beliefs grow stronger in democratic
countries with unitary governments, the standard an-
swers become less persuasivein fact, so unpersua-
sive to the people of the three Scandinavian countries
that they have all abolished their second chambers.
Like the state of Nebraska, Norway, Sweden, and
Denmark also seem to do quite nicely without them.
Even in Britain, the gradual advance of democratic
beliefs created an inexorable force opposed to the his-
torical powers of the House of Lords. As early as 1911
03dahl.041_072 11/27/01 4:39 PM Page 46
the Liberals wiped out the power of the Lords to veto

money bills passed by the Commons. The continu-
ing advance of democratic beliefs during the past cen-
tury led in 1999 to the abolition of all but ninety-two
hereditary seats, whose occupants would be elected by
hereditary peers.5 The future of that ancient chamber
remains in considerable doubt.
By the end of the twentieth century, then, a
strongly bicameral legislature continued to exist in
only four of the advanced democratic countries, all of
them federal: in addition to the United States, these
were Australia, Germany, and Switzerland. Their exis-
tence poses a question: What functions can and should
a second chamber perform in a democratic country?
And in order to perform its proper functions, if any,
how should a second chamber be composed? As the
deliberations of the Parliamentary Commission on the
future of the House of Lords indicate, these questions
admit of no easy answer. It would not be surprising,
then, if Britain ends up with no real second chamber at
all, even if a ghostly shade of the upper house persists.
Unequal Representation
A third characteristic of federal systems is significant

unequal representation in the second chamber. By un-
equal representation I mean that the number of mem-
bers of the second chamber coming from a federal
unit such as a state or province is not proportional to
03dahl.041_072 11/27/01 4:39 PM Page 47
its population, to the number of adult citizens, or to

the number of eligible voters. The main reason, per-
haps the only real reason, why second chambers exist
in all federal systems is to preserve and protect un-
equal representation. That is, they exist primarily to
ensure that the representatives of small units cannot
be readily outvoted by the representatives of large
units. In a word, they are designed to construct a bar-
rier to majority rule at the national level.
To make this clear, let me extend the range of the
term unequal representation to include any system
where, in contrast to the principle of one person one
vote, the votes of different persons are given unequal
weights. Whenever the suffrage is denied to some per-
sons within a system, we might say that their votes are
counted as zero, whereas the votes of the eligible citi-
zens are counted as one. When women were denied
the vote, a mans vote effectively counted for one, a
womans for nothing, zero. When property require-
ments were required for the suffrage, property owners
were represented in the legislature, those below the
property threshold were not: like women their votes
counted for zero. Some privileged members of Parlia-
ment, like Edmund Burke, referred to virtual repre-
sentation, where the aristocratic minority repre-
sented the best interests of the entire country. But the
bulk of the people who were excluded easily saw
through that convenient fiction, and as soon as they
were able to they rejected these pretensions and
gained the right to vote for their own M.P.s. In nine-
03dahl.041_072 11/27/01 4:39 PM Page 48
teenth-century Prussia, voters were divided into three

classes according to the amount of their property
taxes. Because each class of property owners was given
an equal number of votes irrespective of the vast dif-
ference in numbers of persons in each class, a wealthy
Prussian citizen possessed a vote that was effectively
worth almost twenty times that of a Prussian worker.6
To return now to the United States: as the Ameri-
can democratic credo continued episodically to exert
its effects on political life, the most blatant forms of
unequal representation were in due time rejected. Yet,
one monumental though largely unnoticed form of un-
equal representation continues today and may well
continue indefinitely. This results from the famous
Connecticut Compromise that guarantees two sena-
tors from each state.
Imagine a situation in which your vote for your
representative is counted as one while the vote of a
friend in a neighboring town is counted as seventeen.
Suppose that for some reason you and your friend
each change your job and your residence. As a result
of your new job, you move to your friends town. For
the same reason, your friend moves to your town.
Presto! To your immense gratification you now dis-
cover that simply by moving, you have acquired six-
teen more votes. Your friend, however, has lost sixteen
votes. Pretty ridiculous, is it not?
Yet that is about what would happen if you lived
on the western shore of Lake Tahoe in California and
moved less than fifty miles east to Carson City, Nevada,
03dahl.041_072 11/27/01 4:39 PM Page 49
while a friend in Carson City moved to your commu-

nity on Lake Tahoe. As we all know, both states are
equally represented in the U.S. Senate. With a popula-
tion in 2000 of nearly 34 million, California had two
senators. But so did Nevada, with only 2 million resi-
dents. Because the votes of U.S. senators are counted
equally, in 2000 the vote of a Nevada resident for the
U.S. Senate was, in effect, worth about seventeen times
the vote of a California resident. A Californian who
moved to Alaska might lose some points on climate,
but she would stand to gain a vote worth about fifty-
four times as much as her vote in California.7 Whether
the trade-off would be worth the move is not for me
to say. But surely the inequality in representation it
reveals is a profound violation of the democratic idea
of political equality among all citizens.
Some degree of unequal representation also exists
in the other federal systems. Yet the degree of unequal
representation in the U.S. Senate is by far the most
extreme. In fact, among all federal systems, including
those in more newly democratized countriesa total
of twelve countrieson one measure the degree of
unequal representation in the U.S. Senate is exceeded
only by that in Brazil and Argentina.8
Or suppose we take the ratio of representatives in
the upper chamber to the populations of the federal
units. In the United States, for example, the two sena-
tors from Connecticut represent a population of slightly
above 3.4 million, while the two senators from its neigh-
bor New York represent a population of 19 million:
03dahl.041_072 11/27/01 4:39 PM Page 50
a ratio of about 5.6 to 1. In the extreme case, the ratio

of over-representation of the least populated state,
Wyoming, to the most populous state, California, is just
under 70 to 1.9 By comparison, among the advanced
democracies the ratio runs from 1.5 to 1 in Austria to
40 to 1 in Switzerland. In fact, the U.S. disproportion
is exceeded only in Brazil, Argentina, and Russia.10
On what possible grounds can we justify this ex-
traordinary inequality in the worth of the suffrage?
A brief digression: rights and interests. A common
response is to say that people in states with smaller
populations need to be protected from federal laws
passed by congressional majorities that would violate
their basic rights and interests. Because the people in
states like Nevada or Alaska are a geographical minor-
ity, you might argue, they need to be protected from
the harmful actions of national majorities. But this re-
sponse immediately raises a fundamental question. Is
there a principle of general applicability that justifies
an entitlement to extra representation for some indi-
viduals or groups?
In searching for an answer, we need to begin with
an eternal and elementary problem in any governmen-
tal unit:11 whether the unit is a country, state, munici-
pality, or whatever, virtually all of its decisions will in-
volve some conflict of interests among the people of
the relevant political unit. Inevitably, almost any gov-
ernmental decision will favor the interests of some cit-
izens and harm the interests of others. The solution to
this problem, which is inherent in all governmental
03dahl.041_072 11/27/01 4:39 PM Page 51
units, is ordinarily provided in a democratic system by

the need to secure a fairly broad consent for its deci-
sions by means, among other things, of some form of
majority rule. Yet if decisions are arrived at by majority
rule, then the possibility exists, as Madison and many
others have observed, that the interests of any minor-
ity will be damaged by a majority. Sometimes, fortu-
nately, mutually beneficial compromises may be found.
But if the interests of a majority clash irreconcilably
with those of a minority, then the interests of that mi-
nority are likely to be harmed.
Some interests, however, may be protected from
the ordinary operation of majority rule. To a greater or
lesser degree, all democratic constitutions do so.
Consider the protections that all Americans enjoy,
not just in principle but substantially in practice as
well. First, the Bill of Rights and subsequent amend-
ments provide a constitutional guarantee that certain
fundamental rights are protected whether a citizen
lives in Nevada or California, Rhode Island or Massa-
chusetts, Delaware or Pennsylvania. Second, an im-
mense body of federal law and judicial interpretation
based on constitutional provisions enormously extends
the domain of protected rightsprobably far beyond
anything the Framers could have foreseen. Third, the
constitutional division of powers in our federal system
provides every state with an exclusive or overlapping
domain of authority on which a state may draw in
order to extend even further the protections for the
particular interests of the citizens of that state.
03dahl.041_072 11/27/01 4:39 PM Page 52
The basic question. Beyond these fundamental

and protected rights and interests, do people in the
smaller states possess additional rights or interests that
are entitled to protection from policies supported by
national majorities? If so, what are they? And on what
general principle can their special protection be justi-
fied? Surely they do not include a fundamental right
to graze sheep or cattle in national forests or to extract
minerals from public lands on terms that were set
more than a century ago. Why should geographical lo-
cation endow a citizen or group with special rights and
interests, above and beyond those I just indicated, that
should be given additional constitutional protection?
If these questions leave me baffled, I find myself
in good company. Can we forget for whom we are
forming a government? James Wilson asked at the
Constitutional Convention. Is it for men, or for the
imaginary beings called States? Madison was equally
dubious about the need to protect the interests of
people in the small states. Experience, he said, sug-
gests no such danger. . . . Experience rather taught a
contrary lesson. . . . The states were divided into dif-
ferent interests not by their differences in size, but by
other circumstances.12
Two centuries of experience since Madisons time
have confirmed his judgment. Unequal representation
in the Senate has unquestionably failed to protect the
fundamental interests of the least privileged minorities.
On the contrary, unequal representation has some-
times served to protect the interests of the most privi-
03dahl.041_072 11/27/01 4:39 PM Page 53
leged minorities. An obvious case is the protection of

the rights of slaveholders rather than the rights of their
slaves. Unequal representation in the Senate gave ab-
solutely no protection to the interests of slaves. On the
contrary, throughout the entire preCivil War period
unequal representation helped to protect the interests
of slave owners. Until the 1850s equal representation
in the Senate, as Barry Weingast has pointed out, gave
the the South a veto over any policy affecting slavery.
Between 1800 and 1860 eight anti-slavery measures
passed the House, and all were killed in the Senate.13
Nor did the Southern veto end with the Civil War. After
the Civil War, Senators from elsewhere were compelled
to accommodate to the Southern veto in order to secure
the adoption of their own policies. In this way the
Southern veto not only helped to bring about the end of
Reconstruction; for another century it prevented the
country from enacting federal laws to protect the most
basic human rights of African Americans.
So much for the alleged virtues of unequal repre-
sentation in the Senate.
Suppose for a moment we try to imagine that we
actually wanted the constitution to provide special pro-
tection to otherwise disadvantaged minorities by giv-
ing them extra representation in the Senate. What mi-
norities most need this extra protection? How would
we achieve it? Would we now choose to treat certain
states as minorities in special need of protection sim-
ply because of their smaller populations? Why would
we want to protect these regional minorities and not
03dahl.041_072 11/27/01 4:39 PM Page 54
other, far weaker minorities? To rephrase James Wil-

sons question in 1787: Should a democratic govern-
ment be designed to serve the interests of the imagi-
nary beings called States, or should it be designed
instead to serve the interests of all its citizens consid-
ered as political equals?
As I have said, the United States stands out among
twenty-two comparable democratic countries for the
degree of unequal representation in its upper chamber.
Of the half dozen that have federal systems and an
upper house designed to represent the federal units,
none come even close to the United States in the ex-
tent of its unequal representation in its upper house.
We begin to see, then, that our constitutional sys-
tem is unusual. As we continue our exploration we shall
discover that it is not merely unusual. It is one of a kind.
Strong Judicial Review of National Legislation
Not surprisingly, other federal systems among the older

democracies also authorize their highest national courts
to strike down legislation or administrative actions by
the federal unitsstates, provinces, and the like
that are contrary to the national constitution. The case
for the power of federal courts to review state actions
in order to maintain a federal system seems to me
straightforward, and I accept it here. But the authority
of a high court to declare unconstitutional legislation
that has been properly enacted by the coordinate con-
03dahl.041_072 11/27/01 4:39 PM Page 55
stitutional bodiesthe parliament or in our system the

Congress and the presidentis far more controversial.
If a law has been properly passed by the law-
making branches of a democratic government, why
should judges have the power to declare it unconstitu-
tional? If you could simply match the intentions and
words of the law against the words of the constitution,
perhaps a stronger case could be made for judicial re-
view. But in all important and highly contested cases,
that is simply impossible. Inevitably, in interpreting the
constitution judges bring their own ideology, biases,
and preferences to bear. American legal scholars have
struggled for generations to provide a satisfactory ra-
tionale for the extensive power of judicial review that
has been wielded by our Supreme Court. But the
contradiction remains between imbuing an unelected
bodyor in the American case, five out of nine jus-
tices on the Supreme Courtwith the power to make
policy decisions that affect the lives and welfare of
millions of Americans. How, if at all, can judicial re-
view be justified in a democratic order? Ill discuss
that question in my last chapter.
Meanwhile, let me return to another aberrant as-
pect of the American constitutional system.
Electoral Systems
Earlier I explained that I wanted to use the term con-

stitutional system because some arrangements that are
03dahl.041_072 11/27/01 4:39 PM Page 56
not necessarily specified in a countrys constitutional

document interact so strongly with the other institu-
tions that we can usefully regard them as a part of the
countrys constitutional arrangements. In that spirit, we
might want to reflect on the peculiarities of our elec-
toral system, which, natural as it may seem to us, is of a
species rare to the vanishing point among the advanced
democratic countries. Closely allied with it is an equally
rare bird, our much revered two-party system.
To be sure, our electoral system was not the doing
of the Framers, at least directly, for it was shaped less
by them than by British tradition. The Framers simply
left the whole matter to the states and Congress,14
both of which supported the only system they knew,
one that had pretty much prevailed in Britain, in the
colonies, and in the newly independent states.
The subject of electoral systems is fearfully com-
plex and for many people fearfully dull as well. I shall
therefore employ a drastic oversimplification, but one
sufficient for our purposes. Let me simply divide elec-
toral systems into two broad types, each with a variant
or two. In the one we know best, typically you can cast
your vote for only one of the competing candidates,
and the candidate with the most votes wins. In the
usual case, then, a single candidate wins office by gain-
ing at least one more vote than any of his or her oppo-
nents. We Americans tend to call this one-vote margin
a plurality; elsewhere, to distinguish it from an ab-
solute majority it may be called a relative majority.
To describe our system, American political scientists
03dahl.041_072 11/27/01 4:39 PM Page 57
sometimes employ the cumbersome expression single

member district system with plurality elections. I pre-
fer the British usage: on the analogy of a horse race
where the winner needs only a fraction of a nose-
length to win, the British tend to call it the first-past-
the-post system.
If voters were to cast their ballots in the same pro-
portion in every district, the party with the most votes
would win every seat. In practice, as a result of varia-
tions from district to district in support for candidates,
a second party generally manages to gain some seats,
although its percentage of seats will ordinarily be
smaller than its percentage of votes. But the represen-
tation of third parties usually diminishes to the vanish-
ing point. In short, first-past-the-post favors two-party
systems.
The main alternative to first-past-the-post is pro-
portional representation. As the name implies, propor-
tional representation is designed to ensure that voters
in a minority larger than some minimal sizesay, 5
percent of all voterswill be represented more or less
in proportion to their numbers. For example, a group
consisting of 20 percent of all voters might win pretty
close to 20 percent of the seats in the parliament.
Consequently, countries with proportional represen-
tation systems are also very likely to have multiparty
systems in which three, four, or more parties are rep-
resented in the legislature. In short, although the rela-
tionship is somewhat imperfect, in general a country
with first-past-the-post is likely to have a two-party
03dahl.041_072 11/27/01 4:39 PM Page 58
system and a country with proportional representation

is likely to have a multiparty system.
In the most common system of proportional repre-
sentation, each party presents voters with a list of its
candidates; voters cast their votes for a partys candi-
dates; each party is then awarded a number of seats
roughly in proportion to its overall share of the vote.
Countries with a list system may also permit voters to
indicate their preferences among the partys candi-
dates. The partys seats are then filled by the candi-
dates who are most preferred by the voters. Twelve of
the twenty-two advanced democratic countries employ
the list system of proportional representation, and an-
other six use some variant of it. (See Appendix B,
Table 3.)
Of the four countries without proportional repre-
sentation, France avoids one of the defects of single-
member districts by providing that in parliamentary
districts where no candidate receives an absolute ma-
jority of votes, a second election will be held in which
the two candidates with the highest number of votes
compete. This run-off, two-round, or double-ballot sys-
tem, as it is variously called, thereby ensures that all
the members have been elected by a majority of the
voters in their constituency.
This leaves the three oddballs with first-past-the-
post, a plurality system in single member districts:
Canada, the United Kingdom, and the United States.
Even in the United Kingdom, the original source on
which the Americans drew, the traditional system was
03dahl.041_072 11/27/01 4:39 PM Page 59
replaced by proportional representation in the 1999

elections to the newly created legislative bodies in
Scotland and Wales. Four parties won seats in the
Scottish Parliament, and four too in the Welsh Assem-
bly. What is more, the Independent Commission on
the Voting System set up by the Labor Party in 1997 to
recommend an alternative to first-past-the-post pro-
posed in its report a year later that members of the
House of Commons be elected by means of a propor-
tional representation systema hybrid, to be sure,
but one that would ensure greater proportionality be-
tween votes and seats in that ancient house.15 It is al-
together possible that one day not far off, Britain will
be added to the list of proportional representation
countries, leaving only Canada and the United States
among the advanced democracies with first-past-the-
post.
Although few Americans know much about experi-
ence in the other advanced democratic countries with
proportional representation and multiparty systems,
they seem to have strong prejudices against both. Un-
willing to conceive of an alternative to first-past-the-
post and under pressure to ensure fairer representa-
tion for minorities in state legislatures and Congress,
our legislatures and federal courts in recent years have
sometimes gerrymandered weirdly shaped districts. . .
well, yes, rather like a salamander. But neither legisla-
tures nor courts seem willing to give serious thought
to some form of proportional representation as quite
possibly a better alternative.
03dahl.041_072 11/27/01 4:39 PM Page 60
The extent to which we take first-past-the-post for

granted was clearly revealed in 1993, when it was dis-
covered that a well-qualified candidate to head the
Civil Rights Division of the Department of Justice had
written an article in a law journal suggesting that a
rather sensible system of proportional representation
might be worth considering as a possible solution to the
problem of securing more adequate minority repre-
sentation.16 From the comments the authors innocent
heresy generated, you might have thought that she had
burned the American flag on the steps of the Supreme
Court. Her candidacy, naturally, was stone dead.
First-past-the-post was the only game in town in
1787 and for some generations thereafter. Like the
locomotive, proportional representation had not yet
been invented. It was not fully conceived until the
mid-nineteenth century when a Dane and two Eng-
lishmenone of them John Stuart Millprovided a
systematic formulation. Since then it has become the
system overwhelmingly preferred in the older democ-
racies.
After more than a century of experience with other
alternatives, isnt it time at last to open our minds to
the possibility that first-past-the-post may be just fine
for horse races but might not be best for elections in a
large and diverse democratic country like ours? Might
we not also want to consider the possible advantages
of a multiparty system?
I do not say that we should necessarily make these
choices. But should we not at least give them serious
03dahl.041_072 11/27/01 4:39 PM Page 61
consideration? Shouldnt we ask ourselves this ques-

tion: What kind of electoral and party systems would
best serve democratic ends?
Party Systems
Nearly a half-century ago, a French political scientist,

Maurice Duverger, proposed what came to be called
Duvergers Law: first-past-the-post electoral systems
tend to result in two-party systems. Conversely, pro-
portional representation systems are likely to produce
multiparty systems.17 Although the causal relation may
be more complex than my brief statement of Du-
vergers Law suggests,18 a country with a proportional
representation system is likely to require coalition gov-
ernments consisting of two or more parties. In a coun-
try with a first-past-the-post electoral system, however,
a single party is more likely to control both the execu-
tive and the legislature. Thus in countries with propor-
tional representationmultiparty systems and coalition
governments, minorities tend to be represented more
effectively in governing. By contrast, in countries with
first-past-the-post and two-party systems, the govern-
ment is more likely to be in the hands of a single party
that has gained a majority of seats in the parliament
and the most popular votes, whether by an outright
majority, or more commonly, a plurality. To distinguish
the two major alternatives, Ill refer to the propor-
tional representationmultiparty countries as propor-
03dahl.041_072 11/27/01 4:39 PM Page 62
tional and countries with first-past-the-post electoral

systems and only two major parties as majoritarian.19
Where does the United States fit in? As usual: in
neither category. It is a mixed system, a hybrid, nei-
ther predominantly proportional nor predominantly
majoritarian. (See Appendix B, Table 4.) I am going to
return to the American hybrid in Chapter 5, but three
brief observations may help to put it in perspective
here. First, the Framers had no way of knowing about
the major alternatives to first-past-the-post, much less
fully understanding them. Second, since the Framers
time most of the older and highly stable democratic
countries have rejected first-past-the-post and opted
instead for proportional systems. Third, our mixed de-
sign contributes even further to the unusual structure
of our constitutional system.
Our Unique Presidential System
As we make our way through the list of countries that

share some constitutional features with the United
States, the list, short to begin with, diminishes even fur-
ther. By the time we reach the presidency the United
States ceases to be simply unusual. It becomes unique.
Among the twenty-two advanced democracies, the
United States stands almost alone in possessing a single
popularly elected chief executive endowed with im-
portant constitutional powersa presidential system.
Except for Costa Rica, all the other countries govern
03dahl.041_072 11/27/01 4:39 PM Page 63
themselves with some variation of a parliamentary sys-

tem in which the executive, a prime minister, is chosen
by the national legislature. In the mixed systems of
France and Finland, most of the important constitu-
tional powers are assigned to the prime minister, but
an elected president is also provided with certain pow-
erschiefly over foreign relations. This arrangement
may lead, as in France, to a president from one major
party and a prime minister from the opposing party, a
situation that with a nice Gallic touch the French call
cohabitation. Yet even allowing for the French and
Finnish variations, none of the other advanced demo-
cratic countries has a presidential system like ours.
Why is this? The question breaks down into sev-
eral parts. Why did the Framers choose a presidential
system? Why didnt they choose a parliamentary sys-
tem? Why have all the other advanced democratic
countries rejected our presidential system? Why have
they adopted some variant of a parliamentary system
instead, or as in France and Finland a system that is
predominantly parliamentary with an added touch of
presidentialism?
To answer these questions in detail would go be-
yond our limits here. But let me sketch a brief answer.
Before I do so, however, I want to admonish you
not to cite the explanation given in the Federalist
Papers. These were very far from critical, objective
analyses of the constitution. If we employ a dictionary
definition of propaganda as information or ideas me-
thodically spread to promote or injure a cause, nation,
03dahl.041_072 11/27/01 4:39 PM Page 64
etc., then the Federalist Papers were surely propa-

ganda. They were written post hoc by partisans
Alexander Hamilton, John Jay, and James Madison
who wanted to persuade doubters of the virtues of the
proposed constitution in order to secure its adoption
in the forthcoming state conventions. Although they
were very fine essays indeed, and for the most part
much worth reading today, they render the work of
the convention more coherent, rational, and compel-
ling than it really was. Ironically, by the way, the task
of explaining and defending the Framers design for
the presidency was assigned to Hamilton, who had
somewhat injudiciously remarked in the Convention
that as to the executive, The English model was the
only good one on this subject, because the heredi-
tary interest of the king was so interwoven with that of
the nation. . . and at the same time was both suffi-
ciently independent and sufficiently controuled [sic],
to answer the purpose. He then proposed that the ex-
ecutive and one branch of the legislature hold their
places for life, or at least during good behavior.20 Per-
haps as a result of these remarks, Hamilton seems to
have had only a modest influence in the Convention
on that matter or any other.
How it came about. What is revealed in the most
complete record of the Convention21 is a body floun-
dering in its attempts to answer an impossibly difficult
question: How should the chief executive of a republic
be selected, and what constitutional powers should be
assigned to the executive branch? The question was im-
03dahl.041_072 11/27/01 4:39 PM Page 65
possibly difficult because, as I emphasized in the pre-

vious chapter, the Framers had no relevant model of
republican government to give them guidance. Most
of all, they lacked any suitable model for the executive
branch. To be sure, they could draw on the sacred doc-
trine of separation of powers. Not surprisingly, the
references to that doctrine recorded in Madisons notes
were all positive. And up to a point, its implications
were obvious: a republic would need an independent
judiciary, a bicameral legislature consisting of a popular
house and some kind of second chamber to check the
popular house, and an independent executive.
But how was the independent executive to be cho-
sen? How independent of the legislature and of the
people should he be? How long should his term of of-
fice be? (He is, of course, the language of Article II
and, like most Americans until recently, the only way
the Framers could conceive of the office.) The British
constitution was a helpful model for the Framers in
some respects. But as a solution to the problem of the
executive, it utterly failed them. Despite the respect of
the delegates for many aspects of the British constitu-
tion, a monarchy was simply out of the question.22
Even so, they might have chosen a democratic ver-
sion of the parliamentary system, as the other evolving
European democracies were to do. Although they were
unaware of it, even in Britain a parliamentary system
was already evolving. Why then didnt the Framers
come up with a republican version of a parliamentary
system?
03dahl.041_072 11/27/01 4:39 PM Page 66
Well, they almost did. It has been too little empha-

sized, I think, that the Framers actually came very
close to adopting something like a parliamentary sys-
tem. What is more, it is far from clear, to me at least,
why they rejected it and ended up instead with a pres-
idential system.23 One obvious solutioneven more
obvious to us today than it would have been in 1787
was to allow the national legislature to choose the ex-
ecutive. In fact, throughout most of the Convention
this was their favored solution. Right off the bat on
June 2, only two weeks after the Convention opened,
the Virginia delegation, which contained some of the
best minds and most influential delegates, proposed
that the national executive should be chosen by the na-
tional legislature. In Madisons notes, the subsequent
course of that proposal and the alternatives to it has
left a fascinating and often mystifying trail.
The meandering trail they pursued, as best I can
reconstruct it, looks something like this.24 On three
occasionsJuly 17, July 24, and July 26the dele-
gates vote for the selection of the president by the na-
tional legislature, the first time by a unanimous vote,
the last by a vote of 63. With one exception every
other alternative is defeated by substantial majorities:
in a puzzling detour on July 19, with Massachusetts di-
vided, they vote 63 for electors appointed by the
state legislatures. On July 26, their favored solution,
election by the national legislature, is forwarded to a
Committee on Detail. On August 6 the committee duly
reports in favor of election by the national legislature.
03dahl.041_072 11/27/01 4:39 PM Page 67
On August 24 two other alternatives fail once again. A

new committee to consider the issue reports back on
September 4. By now the delegates are eager to wind
up a convention that has already gone on for three
months. In contradiction to the recommendation of
the previous committee, however, this one recom-
mends that the executive be chosen by electors ap-
pointed by the state legislatures. Two days later, with
nine states in favor and only two opposed, the impa-
tient delegates adopt this solution.
Well, not exactly. What they adopt actually states
that: Each state shall appoint, in such manner as the
legislature thereof may direct, a number of electors,
equal to the whole number of Senators and representa-
tives to which the State may be entitled in Congress.
Whatever the Framers intend by these words, they will
offer a huge opportunity for the democratic phase of
the American revolution to democratize the presidency.
Ten days after they agree on this provision, the
constitution is signed and the Convention adjourns.
What this strange record suggests to me is a group
of baffled and confused men who finally settle on a so-
lution more out of desperation than confidence. As
events were soon to show, they had little understand-
ing of how their solution would work out in practice.
So the question remains with no clear answer: Why,
finally, did they fail to adopt the solution they had
seemed to favor, a president elected by the Congress, a
sort of American version of a parliamentary system? The
standard answer no doubt has some validity: they feared
03dahl.041_072 11/27/01 4:39 PM Page 68
that the president might be too beholden to Congress.

And all the other alternatives seemed to them worse.
Among these alternatives was election by the
people, which had been twice rejected overwhelm-
ingly. Yet it was this twice-rejected solution, election
by the people, that was quickly adopted de facto dur-
ing the democratic phase of the American revolution.
How their solution failed. Perhaps in no part of
their work did the Framers fail more completely to
design a constitution that would prove acceptable to a
democratic people. As I have mentioned, their hope
for a group of electors who might exercise their inde-
pendent judgments about the best candidate to fill the
office came a cropper following the election of 1800.
But as I shall describe in the next chapter, more was
still to come. If the election of 1800 first revealed how
inappropriate the electoral college was in a democratic
order, the presidential election of 2000, two centuries
later, dramatized for all the world to witness the con-
flict between the Framers constitution and the demo-
cratic ideal of political equality.
Ironically, had they adopted the Virginia Plan and
placed the choice of the chief executive in the hands of
the legislature, as would become the practice in parlia-
mentary systems, the Framers would have put a bit
more distance between the people and the president
than their solution provided in practice. Here again, in
1787 they could not anticipate a constitutional design
that was yet to evolve fully in Britain and, even later, in
other countries on the path to democracy.
03dahl.041_072 11/27/01 4:39 PM Page 69
The continuing democratic revolution would bring

about an even more profound change in the presi-
dency. However deftly Jefferson steered the Congress
as he rode the tide of the democratic revolution, he
never publicly challenged the standard view that the
only legitimate representative of the popular will was
the Congress, not the president. Nor did any of his
successors, Madison, Monroe, John Quincy Adams, lay
down such a claim.
Andrew Jackson did just that. In justifying his use
of the veto against Congressional majorities, as the
only national official who had been elected by all the
people and not just by a small fraction, as were Sena-
tors and Representatives, Jackson insisted that he alone
could claim to represent all the people. Thus Jackson
began what I have called the myth of the presidential
mandate: that by winning a majority of popular (and
presumably electoral) votes, the president has gained
a mandate to carry out whatever he had proposed
during the campaign.25 Although he was bitterly at-
tacked for this audacious assertion, which not all later
presidents supported, it gained credibility from its re-
assertion by Lincoln, Cleveland, Theodore Roosevelt,
and Wilson and was finally nailed firmly in place by
Franklin Roosevelt.
Whatever we may think of the validity of the
claimI am inclined to think it is little more than a
myth created to serve the political purposes of ambi-
tious presidentsit is simply one part of a transforma-
tion of the presidency in response to democratic ideas
03dahl.041_072 11/27/01 4:39 PM Page 70
and beliefs that has produced an office completely dif-

ferent from the office that the Framers thought they
were creating, vague and uncertain as their intentions
may have been.
And a good thing, too, you may say. But if you ap-
prove of the democratization of the presidencyor, as I
would prefer to say, its pseudo-democratizationarent
you suggesting in effect that the constitutional system
should be altered to meet democratic requirements?
Why other countries became parliamentary de-
mocracies. There is still one more reason why the
Framers didnt choose a parliamentary system. They
had no model to inspire them. One hadnt yet been in-
vented.
The British constitutional system they knew, and in
some respects admired, was already on its way to his-
torys attic of abandoned or failed constitutions. Al-
though no one saw it clearly in 1787, even at the time
of the Convention the British constitution was under-
going rapid change. Most important, the monarch was
swiftly losing the power to impose a prime minister on
the parliament. The contrary assumption was gaining
strength: that a prime minister must receive a vote of
confidence from both houses of parliament, and that
he must resign if and whenever he lost their confi-
dence. But this profound change in the British consti-
tution did not become fully manifest until 1832, too
late for the Framers to see its possibilities.
In addition, there was the problem of a monarch.
How could a country have a parliamentary system
03dahl.041_072 11/27/01 4:39 PM Page 71
without a symbolic head of state who would perform

ceremonial functions, symbolize the unity of the coun-
try, and help to confer legitimacy on the parliaments
choice by anointing him as prime minister? After the
evolution of a parliamentary system in Britain, in due
time monarchies also helped the Swedes, the Danes,
and the Norwegiansand much later Japan and
Spainto move to a parliamentary system that the
monarchy helped to legitimize. But in 1787 the full
development of parliamentary democracy in countries
with a monarchy was still a long way off. For Ameri-
cans, a monarch, even a ceremonial monarch, was com-
pletely out of the question. So why didnt they split the
two functions, ceremonial and executive, by creating a
titular head of state to serve in the place of a ceremo-
nial monarch, and a chief executive, the equivalent of
a prime minister, to whom executive functions would
be assigned? Although that arrangement may seem
obvious enough to us now, for the Framers in 1787 it
was even more distant than the system that was gradu-
ally evolving in Britain, the country they knew best. It
was not until after 1875 and the installation of the
Third Republic in France that the French evolved a
solution that would later be adopted in many other de-
mocratizing countries: a president elected by the par-
liament, or in some cases by the people, who serves as
formal head of state, and a prime minister chosen by
and responsible to the parliament, who serves as the
actual chief executive. But for the Framers this inven-
tion, which now seems obvious enough to us, was al-
03dahl.041_072 11/27/01 4:39 PM Page 72
most as far off and about as difficult to imagine, per-

haps, as a transcontinental railroad.
Without intending to do so, then, the Framers cre-
ated a constitutional framework that under the driving
impact of the continuing American Revolution would
develop a presidency radically different from the one
they had in mind. In time American presidents would
gain office by means of popular electionsa solution
the Framers rejected and fearedand by combining
the functions of a head of state with those of a chief
executive the president would be the equivalent of
monarch and prime minister rolled into one.
I cant help wondering whether the presidency
that has emerged is appropriate for a modern demo-
cratic country like ours.
! ! !
SO: AMONG THE OLDER DEMOCRACIES OUR CONSTITU-

tional system is not just unusual. It is unique.
Well, you might say, being unique isnt necessarily
bad. Perhaps our constitutional system is better for it.
Better by what standards? Is it more democratic?
Does it perform better in many ways? Or worse?
These questions are by no means easy to answer
probably impossible to answer with finality. But before
turning to them, we need to take one more look at that
anomalous vestige of the Framers work, the electoral
college.
04dahl.073_090 11/27/01 4:40 PM Page 73
chapter 4
Electing the President
O
N THE NIGHT OF NOVEMBER 7, 2000, A DRAMA
opened in the United States that absorbed the
attention of millions of people until the cur-
tain came down six weeks later. The nation was fo-
cused once again on an anomalous institution that had
originated in the Framers search for a suitable way to
elect the new republics chief executive. This was the
electoral college, by means of which the presidency
was wonnot for the first time and perhaps not the
lastby a candidate with fewer votes than his rival.1
As we saw in the previous chapter, the Framers
were baffled by how to conceive of the executive in
a republic. How should a republican executive be
chosen? During the final debates over the electoral
college, James Wilson commented: This subject has
04dahl.073_090 11/27/01 4:40 PM Page 74
74 electing the president
greatly divided the House, and will also divide the

people out of doors. It is in truth the most difficult of
all on which we have had to decide.2 Three months
after the Convention had adjourned, when his memo-
ries were still fresh, he made the same observation to
his fellow Pennsylvanians who had assembled to ratify
the new constitution: The convention, sir, were per-
plexed with no part of this plan, so much as with the
mode of choosing the president of the United States.3
Every solution seemed worse than the rest. The
arrangement they finally cobbled together at the last
minute was adopted more out of desperation, perhaps,
than out of any great confidence in its success. So why
did the delegates finally give their approval to the
electoral college? Probably the best answer to our
question would be: the Framers settled on an electoral
college because they had run out of alternatives.
How the Electoral College Came About
We have discussed how during the three summer

months the delegates considered and rejected the most
obvious possibilities. From the scanty information avail-
able, it is possible to add a few more facts.
On August 6, the Committee on Detail brings in a
draft of the Constitution that calls for the election of
the president by Congress. This proposal finds no tak-
ers. As late as August 24, despite mounting pressures
04dahl.073_090 11/27/01 4:40 PM Page 75
electing the president 75
to complete their work and adjourn, the delegates vote

down every plausible alternative. Even a motion that
the president shall be chosen by electors fails to gain
a majority of votes. Unable to agree, on the last day of
August the delegates turn the problem of choosing
the president over to yet another committee, this one
drawn from each of the eleven state delegations. Four
days later the committee offers one of the solutions the
delegates had already rejected earlier: Each State shall
appoint in such manner as its Legislature may direct, a
number of electors equal to the whole number of Sen-
ators and members of the House of Representatives to
which the State may be entitled in the Legislature.4
The delegates, no doubt weary of their task and
eager to finish, find one remaining flaw. The commit-
tee has proposed that in case of a tie vote in the elec-
toral college, the choice will be made by the Senate
from the five highest candidates. Objection! Many del-
egates believe, it appears, that the competition for the
presidency will usually lead to more than two leading
candidatesthree, four, five, more. If their conjec-
tures are correct, then the election of the president
will frequently, perhaps regularly, fall to the Senate, a
body they expect to be highly powerful and rather
aristocratic. Wilson forcefully puts the case against
conferring that power on the Senate. Combining this
additional power with other parts of the plan, he
said, he was obliged to consider the whole as having a
dangerous tendency to aristocracy; as throwing a dan-
04dahl.073_090 11/27/01 4:40 PM Page 76
gerous power into the hands of the Senate.5 Others

agree, and by a lopsided vote the delegates relocate
the power in the hands of the more popular branch.
Thus was the electoral college born.
Why the Electoral College?
To return now to our question. Why did the Framers

settle on this solution? The usual answer runs like this:
they wanted to remove the choice of the President
from the hands of popular majorities and to place the
responsibility in the hands of a select body of wise,
outstanding, and virtuous citizensas they clearly saw
themselves, a cynic might add. The main source of the
standard view seems to be the Federalist No. 68, writ-
ten by Hamilton. The immediate election should be
made by men most capable of analyzing the qualities
adapted to the station, and acting under circumstances
favorable to deliberation, and to a judicious combina-
tion of all the reasons and inducements which were
proper to govern their choice.6 Nothing, of course,
could be further from the reality of the electoral col-
lege as it swiftly developed. And even Hamilton ap-
pears to have misunderstood the mechanics of the
electoral college in one respect: He assumed that the
people of each State shall choose . . . [the] electors.7
But what the Constitution actually provided, as we just
saw, was that the power to determine how the electors
were to be chosen was assigned to the state legisla-
04dahl.073_090 11/27/01 4:40 PM Page 77
tures. And most states did not initially assign that re-
sponsibility to the people.
The fullest available explanation for the commit-
tees proposal is one offered by Gouverneur Morris, a
member (and probably a highly influential member)
of the committee, who ran through the now familiar
objections to alternative solutions:
Congress: The danger of intrigue & faction if the ap-
pointmt. [sic] Should be made by the Legislature.
No body appeared to be satisfied with an appoint-
ment by the Legislature. The indispensable neces-
sity of making the Executive independent of the Leg-
islature.
The People: Many were anxious [i.e., in fear]8 even for
an immediate choice by the people.
Cabals and Corruption: As the Electors would vote at
the same time throughout the U.S. and at so great a
distance from each other, the great evil of cabal
would be avoided. It would be impossible to corrupt
them.9
Nothing here of Hamiltons later rationalization,

though something like it might have been on the
minds of some delegates.
Failure
No part of the constitution revealed the flaws in its de-

sign more quickly than the provision for the electoral
college. Within a dozen years, the election of 1800 had
already displayed two of its defects. The more serious
04dahl.073_090 11/27/01 4:40 PM Page 78
at the time, but easily corrected, was the failure of the

Framers to distinguish adequately between electing the
president and electing the vice-president.10 At a con-
gressional caucus in May 1800, the Republicans had
unanimously agreed on the nominations of Thomas
Jefferson as president and Aaron Burr as vice presi-
dent. But later on in the electoral college, the votes
were split among five candidates: Jefferson and Burr
tied for president with 73 votes each; of two Federal-
ist candidates, the incumbent, President John Adams,
gained 65, and Charles Cotesworth Pinckney, 64; John
Jay, the governor of New York, received 1 vote. As the
Constitution prescribed, the deadlock was referred to
the House. After thirty-six ballots taken over a weeks
time, Jefferson finally prevailed with the votes of ten
of the sixteen states.
What had not been foreseen by the Framersa tie
between the top two candidateswas now made ob-
vious.11 The solution was equally obvious. The Twelfth
Amendment, requiring separate ballots for president
and vice president, was swiftly adopted, in time for the
election of 1804.12
Although this failure in the Framers design of the
electoral college was easily corrected, the other that
was clearly revealed by the election of 1800 remained.
It remains still. That presidential election shattered
whatever hopes the delegates to the Convention may
have entertained that the electoral college would serve
as an independent body free of the supposed vices of
popular election. Party politicspartisan politics, if you
04dahl.073_090 11/27/01 4:40 PM Page 79
willhad transformed the electors into party agents, a

role that except for an aberrant member would con-
tinue to be played. The privilege of serving as an elec-
tor would typically be awarded not to leading citizens
willing to express their independent judgments, as the
Framers may have assumed, but to party loyalists, and
usually minor ones at that. Thus the development of
political parties and party loyalties turned the elabo-
rate machinery of the electoral college into little more
than a way of counting votes.
What is more, in a change that Ill come back to in
a moment, the democratizing phase of the American
revolution speedily put the choice of the president
where the Framers had explicitly refused to lodge it: in
the hands of the people (that is, the white male voters).
So endeth the aristocratic pretensions of the elec-
toral college.
Inherent Democratic Defects
Even when the electors were chosen by popular elec-

tion, however, three undemocratic features remained
inherent features of the electoral college.
Popular votes versus electoral votes. First, the can-
didate with the greatest number of popular votesa
plurality or even an outright majoritymight not re-
ceive a majority of electoral votes and thus might fail
to be chosen president. Four presidential elections
including that of 2000have led to just such an out-
04dahl.073_090 11/27/01 4:40 PM Page 80
come. In the most famous instance, the election of

1876, a candidate with an outright majority of popular
votes lost the presidency. The crisis began when Samuel
J. Tilden, the Democratic candidate, won 51 percent
of the popular votes and yet failed to gain a clear and
uncontested majority of votes in the electoral college
against his Republican opponent, Rutherford B. Hayes.
In the complex political shenanigans that ensued,13
Southern Democrats in Congress extorted from Hayes
a promise to withdraw federal troops from the South,
and they promised in turn to respect Negro rights
(which, needless to say, they never did). Hayes was
then awarded 185 electoral votes to Tildens 184. In the
words of one account: The country acquiesced. Thus
ended a crisis that could have resulted in civil war.14
In an additional three elections in which the win-
ning candidate gained fewer popular votes than his
rival, none of the candidates gained a majority of popu-
lar votes. In these cases, which include the election of
2000, the votes cast for third party candidates deprived
both major party candidates of a popular majority.15
Winning with a minority of popular votes. In fact,
winning the presidency with only a minority of popu-
lar votes has been a fairly common occurrence. In a
total of eighteen elections, candidates have gained the
presidency without winning a majority of popular
votes. (See Appendix B, Figure 1.) Overall, then, in
one out of every three presidential elections the high-
est office in the land has been awarded to a candidate
chosen by a minority of voters. In a close election
04dahl.073_090 11/27/01 4:40 PM Page 81
where third party candidates pick up some votes, as in

the election of 2000, this outcome is highly probable.
Losing though preferred by a majority. In some
cases where no candidate receives a majority of popu-
lar votes, if voters second choices had been taken into
account (as they can be in some electoral systems), or
if a run-off had occurred between the two highest can-
didates, it is altogether possible that the outcome
might have gone the other way. If voters had been
provided with an opportunity to express their second
choices in the contested election of 2000, most votes
of the major third party candidate, Ralph Nader,
would likely have gone to Vice President Gore, in
which case Gore would have gained the presidency.
Unequal representation of voters. To these de-
fects in the electoral college we can add yet one more.
Because each state is entitled to a number of electors
equal to the whole number of Senators and Represen-
tatives from that state, unequal representation in the
Senate plays itself out once again. Although the effects
are somewhat diluted by the inclusion of Representa-
tives in the total, the inequality in the weight of votes
in the electoral college is still strong. The vote of a
Wyoming resident, for example, is worth almost four
times the vote of a California resident in the electoral
college. Where the number of residents for each elec-
tor runs from 165 thousand to a little more than 300
thousand in the ten smallest states, in the ten largest it
ranges from 586 thousand in Georgia to 628 thousand
in California. The ten smallest states each choose two
04dahl.073_090 11/27/01 4:40 PM Page 82
to three times as many electors as they would if a

states electors were strictly in proportion to its popu-
lation.16 (See Appendix B, Figure 2.)
A Remediable Defect: Winner Take All
The democratic deficiencies inherent in the electoral

college were compounded by another item that was de-
liberately tacked on. For the first elections the states ex-
perimented with a variety of methods for choosing the
electors. An extreme example was Massachusetts, which
altered its system of selecting electors no fewer than
seven times in the first ten presidential elections, often
to suit short-term partisan interests.17 The two major
options were to lodge the choice of electors in the legis-
lature or to give the choice to the people; popular
choice in turn could be made either in districtsone
elector per districtor at large, with the winner taking
all the states electoral votes. Giving the legislature the
power to choose the delegates ran sharply contrary to
the democratizing currents of the time, and popular
election soon predominated. By 1832 only South Car-
olina continued to lodge the choice with the legislature,
a practice it finally abandoned during the Civil War.18
But of the two systems of popular election, winner-take-
all soon predominated as political leaders concluded
that by concentrating all the states electoral votes in a
single slate, they could enhance their weight in the elec-
toral college and thus their influence on the elections.
04dahl.073_090 11/27/01 4:40 PM Page 83
If the winner-take-all system provides some states

with strategic advantages in presidential elections, it
also has at least three disadvantages. First, it reduces
the incentives of a presidential candidate to compete
for votes in safe states that are clearly going to be
taken by one of the two major party candidates. As a
consequence, candidates tend to compete most heav-
ily for votes in swing states that could reasonably be
expected to go either way in the election. Second, it
severely reduces the incentives of potential third-party
candidates to undertake the costly task of running for
president, since they cannot realistically hope to win
any votes at all in the electoral college. Finally, for
both reasons it may weaken the incentives of many
voters in safe states to go to the polls: Why bother to
vote if you know that a majority of voters in your state
will, in effect, choose the entire slate of presidential
electors?
Unlike the problems in the electoral college aris-
ing from features that are constitutionally prescribed,
winner-take-all, as the history of its development clearly
demonstrates, can be altered by state legislatures.
Should We Alter It or Abolish It?
In view of its many deficiencies, what should be done

with the electoral college? Before turning to that ques-
tion, let me consider a common objection to changing
it. As we saw, the constitutional allocation of electors
04dahl.073_090 11/27/01 4:40 PM Page 84
gives a considerable advantage to a voter in a less pop-

ulated state and correspondingly reduces the influ-
ence of a voter in a larger state. This marked inequal-
ity in representation is often defended, as with the
more extreme case of the Senate itself, on the ground
that the small states need and are entitled to protection
from the large states.
Entitlement. Why should the interests of voters
in the small states be entitled to extra protection? The
objections I raised in the previous chapter about the
Senate are also germane to the electoral college, so Ill
repeat them here:
James Wilson at the Constitutional Convention:
Can we forget for whom we are forming a govern-
ment? Is it for men [persons, we would want to say
today] or for the imaginary beings called States?
James Madison: Experience suggests [that] . . .
the states were divided into different interests not by
their differences in size, but by other circumstances.
Beyond the protections provided by the Bill of
Rights, the constitutional division of powers in our
federal system, and an immense body of legislative
and judicial protections for basic rights, do people in
the less populated states possess additional rights or
interests that are entitled to special protection by
means of unequal representation? If so, specifically
what are they?
Is there a principle of general applicability that
justifies an entitlement to extra representation for
some individuals or groups? If so, what is it?
04dahl.073_090 11/27/01 4:40 PM Page 85
If we were to formulate a general principle and

apply it fairly, would not those most entitled to protec-
tion be the least-privileged minoritiesrather than
people who happen to live in the smaller states?
I am not aware of any convincing answers to these
questions and objections.
Need. Behind the belief that small states need
special protection against the larger states may lurk an
image of the oversized bully intimidating his weaker
fellows. But once again the concrete issue comes down
to one of legitimate rights and interests. If the presi-
dent were elected by popular vote, would the legiti-
mate rights and interests of citizens in the small states
suffer from neglect or abuse? The belief that they
would suffer rests on the assumption that presidential
candidates would have little incentive to compete for
the votes of citizens in the small states, and as a result
their interests would tend to be neglected in national
policy-making.
But this assumption seems to me mistaken. In a
system of direct election where every citizens votes
are given equal weight, presidential candidates will be
even more eager than they are now to win votes wher-
ever they might be available; and the closer they ex-
pect the election to be, the more eagerly they will
search out those votes. It is true, of course, that inter-
est groups with many members are likely, as they are
with the electoral college, to secure greater considera-
tion than interest groups with few members. But let us
imagine two relatively small, similarly sized groups of
04dahl.073_090 11/27/01 4:40 PM Page 86
potential voters, one concentrated in a small state and

the other in a large state; both groups, let us assume,
possess interests and express demands that are equally
compatible with a candidates general program. Other
things being roughly comparable, under a system of
popular election the incentives of a presidential can-
didate to win the votes of the two groups would be
equally strong, and the geographical location of each
group would be largely irrelevant, conspicuously so in
this age of television.
I see no reasonable grounds on which to conclude
that the legitimate rights, interests, and demands of
voters in small states should be privileged or that they
would be unjustifiably impaired if the president were
popularly elected.
So what should we do? And what can we do?19
What Should We Do About the Electoral College?
From a democratic perspective, the most desirable

change would be a constitutional amendment that
would replace the electoral college with the direct elec-
tion of the president by popular vote; if no candidate
were to receive more than 50 percent of the popular
vote, a run-off election between the top two candi-
dates would be held shortly thereafter.20
A second possibility is a constitutional amendment
that would maintain the electoral college but require
that a states electoral votes be allocated to candidates
04dahl.073_090 11/27/01 4:40 PM Page 87
in direct proportion to their share of the popular votes

in the state.
Finally, even without a constitutional amendment,
in response to a ground swell of popular opinion, state
legislatures might choose the second solutionand
thus return to the district system that, as we saw, was
adopted in a number of states in the earliest presiden-
tial elections.
What Can We Do?
Evidence from surveys indicates that a substantial ma-

jority of Americans would support a reform of the
electoral college.21 In 1989, a proposal to amend the
Constitution in order to abolish the electoral college
and provide instead for the direct popular election of
the president passed the House of Representatives
with overwhelming support33870, or 83 percent
of the votes in the House.22
Yet desirable and popular as a reform might be, the
most realistic answer to our second questionwhat
can we do?is: Probably not much. By one count, over
seven hundred proposals have been introduced in the
House to modify or abolish the electoral college. None
have succeeded. As might be expected, the graveyard of
constitutional amendments altering the electoral college
is the Senatewhich, as we have seen, is the citadel of
unequal representation. When the proposal for direct
election that had passed the House with 83 percent of
04dahl.073_090 11/27/01 4:40 PM Page 88
the votes in 1989 reached the floor of the Senate a year

later, it ran into a filibusterunlimited debate by oppo-
nents of a measure they do not want to come to a vote.
Under Senate rules, to shut off the debate and get on
with the vote would have required the support of 60
Senators60 percent of the total. The motion to end
debate actually did win a majority of votes54 out of
100; but it failed to gain the 60 percent that the rules re-
quired.23 And even if the proposed amendment had
come to a vote, it could not have gained the necessary
two-thirds67 Senatorsrequired for an amendment.
Thus the requirement that an amendment must
gain the votes of two-thirds of the members of the
Senate gives a veto power to Senators from the small
states, and these Senators may act in concert with
other colleagues who foresee a reduction in the influ-
ence of their states on the presidency.24
The relative desirability of three possible solutions
to the problems of the electoral college appears to be
inversely related to the likelihood of their enactment.
Consequently, however deluded the Framers were
about to the way their last-minute improvisation would
work in practice, it seems unlikely that we shall man-
age to erase this undemocratic blemish on the Ameri-
can constitutional system.
! ! !
THE FRAMERS ATTEMPT TO INSULATE THE CHIEF EX-

ecutive from popular choice provides the most telling
04dahl.073_090 11/27/01 4:40 PM Page 89
example of their failure to provide a constitutional sys-

tem that would be appropriate for a democratic repub-
lic. The fate of the electoral college illustrates three
aspects of this failure.
First, almost from the beginning the electoral col-
lege wholly failed to operate as the Framers had in-
tended. It was, so to speak, swiftly subverted by the
emerging democratic forces.
Second, even the more democratized electoral col-
lege retained features that could and sometimes did
lead to undemocratic outcomes.
Finally, the requirements laid down by the Fram-
ers for amending the constitution made it extraordi-
narily difficult to bring about changes that might be
supported by a majority of American citizens.
04dahl.073_090 11/27/01 4:40 PM Page 90
05dahl.091_120 11/27/01 4:40 PM Page 91
chapter 5
How Well Does the Constitutional

System Perform?
L
ET ME REPEAT THE QUESTION I RAISED AT THE
beginning: Why should we uphold the Ameri-
can Constitution? One response might be: Be-
cause it performs better than any feasible alternative.1
If the unique properties of our constitutional sys-
tem enable it to perform better than the systems of
other democratic countries, then it merits our pride
and confidence. If these peculiarities dont matter, per-
haps we should ignore them. But if it performs worse,
then shouldnt we begin to consider possible changes?
Questions about the relative performance of dif-
ferent constitutional systems are easy to pose but ex-
traordinarily difficult to answer responsibly. True, we
can find today, as only a generation ago or more we
05dahl.091_120 11/27/01 4:40 PM Page 92
92 h ow w e l l d o e s t h e c o n st i t u t i o na l sy st e m p e r f o r m ?
could not, many good indicators of how different coun-

tries systems perform in a variety of important ways:
from literacy, education, health, and life expectancy to
political and civil rights, incomes, income distribution,
and others. It is, however, not easy to determine the ex-
tent to which a countrys constitutional arrangements
influence that countrys performance on such matters.
As one scientific commonplace puts it: Correlations
dont prove causation. If a greater degree of income
inequality exists in the United States than in most of
our twenty-two established democracies, is this a con-
sequence of our unique constitutional system?
As difficult as questions like this are and although
constitution-making is still far from an exact science,
we do have more knowledge at our disposal today
about different constitutional systems than the Fram-
ers could have dreamed ofindeed, more than any
generation in history could possibly have assembled.
In the years to come we could acquire even better
knowledge, if we are determined to do so.
With due respect for uncertainty, then, I want to
assess how well our constitutional arrangements per-
form in comparison with those of the other countries
where democracy is well established. Ill use five crite-
ria. To what extent, if at all, do constitutional arrange-
ments help to:
1. maintain the democratic system;
2. protect fundamental democratic rights;
3. ensure democratic fairness among citizens;
05dahl.091_120 11/27/01 4:40 PM Page 93
h ow w e l l d o e s t h e c o n st i t u t i o na l sy st e m p e r f o r m ? 93
4. encourage the formation of a democratic consensus;

and
5. provide a democratic government that is effective in
solving problems?
Maintaining Democratic Stability
Do different constitutional arrangements significantly

affect the chances that a country will preserve its basic
democratic institutionsthat it will, in short, remain a
democracy? This question opens up a vast subject that
has been extensively explored in recent years.
Unfortunately for my purposes here, but fortu-
nately for democracy, the experience of our twenty-
two democratic countries cannot provide the evidence
we need to answer our question about stability. I se-
lected these countries as suitable for comparison with
the United States precisely because they are the only
countries in the world today that have fully maintained
their basic democratic institutions for a half-century or
more. Since democratic institutions have never col-
lapsed in any of these countries during that period (or
longer), we lack any basis for comparing their perform-
ance in maintaining basic democratic stability. For
example, if twenty-two persons of greatly varying but
moderate diets all remain about equally healthy, we
couldnt draw any conclusions about the effects of their
diets on their health. So, too, with our twenty-two coun-
05dahl.091_120 11/27/01 4:40 PM Page 94
tries: they have all performed equally well in maintain-

ing the existence of their democratic systems.
From our hypothetical study of diets, however, we
could derive one conclusion that is far from trivial:
good health is possible within a considerable range of
moderate diets. Similarly, from the experience of our
twenty-two countries we can draw at least one impor-
tant conclusion: the differences in their constitutional
arrangements evidently have not affected the survival
of their basic democratic institutions. Because all of
these countries have remained steadily democratic, it
follows that within the rather wide range of constitu-
tional variations they exhibit, their differences simply
havent mattered for democratic survival.2
How can we explain this surprising conclusion?
Let me offer three general propositions.
First, if the conditions in a country are highly fa-
vorable for democracy, constitutional differences like
those that exist among our twenty-two countries will
not affect the stability of the basic democratic institu-
tions. To return to the analogy with diet and health:
among a group of people who otherwise live healthy
lives, the variations in their moderate diets wont
matter much. It would take us too far afield to de-
scribe the conditions that favor democratic stability,
but we can say that they appear to include such things
as the effective control by elected leaders over the mil-
itary and police, a political culture supportive of dem-
ocratic beliefs, and a relatively well-functioning eco-
nomic order, among others.
05dahl.091_120 11/27/01 4:40 PM Page 95
Although a country with favorable conditions like

these is likely to maintain its democratic institutions
under a variety of constitutional possibilities, no con-
stitutional system can preserve democracy in a country
where these and other favorable conditions are absent.
As I said, none of our twenty-two democratic coun-
tries have suffered a breakdown of democracy during
the past half-century. But if we move back to the pre-
vious century, we do find that in one of our countries
where all the basic democratic institutions (except for
inclusive citizenship) had been in place for more than
half a century, the constitutional arrangements could
not prevent a breakdown into a civil wara conflict,
moreover, that resulted in vastly greater casualties than
the American or the French Revolution. That country
was the United States. The conditions necessary for
preserving national unity had become so unfavorable
that probably no constitutional arrangements could
have prevented both secession and civil war. Given the
extreme polarization in interests, values, and ways of
life between the citizens of the slave states and those
of the free states, I cannot imagine any democratic
constitution under which the two sections could have
continued to coexist peacefully in one country.
But suppose that, unlike our twenty-two countries,
we have a country in which some conditions are fa-
vorable for democracy while others are unfavorable.
Might the particular features of a constitution matter
in a country where the underlying conditions make
democracy rather chancy? It seems possible that in
05dahl.091_120 11/27/01 4:40 PM Page 96
situations of uncertainty, constitutional arrangements

might just tip the balance one way or the other, toward
democratic stability or democratic breakdown. What
does the evidence indicate?
A much debated issue is the effect on democratic
stability of presidential versus parliamentary systems.
Perhaps because they were impressed by American
stability and power, developing countries have often
adopted some version of a presidential system. As two
scholars have noted, A remarkable fact . . . is the ex-
tent to which presidentialism is a Third World phe-
nomenon.3 So is the likelihood of breakdown. Are the
two connectedpresidentialism and breakdown? The
answer is a subject of dispute. Some scholars have con-
cluded that in countries where the conditions for a sta-
ble democracy are mixedsome favorable, some un-
favorablea presidential system is more likely than a
parliamentary system to put a greater strain on demo-
cratic survival.4 Others argue, however, that parlia-
mentarism has not fared any better in the third world
than has presidentialism; arguably, it has fared worse.5
To explore this controversial question would take
us far outside the purposes of this book, thus I leave
it unsettled here but accompanied by four brief ad-
monitions: One, the intricate American constitutional
system is probably not suitable for export to other
countries. Two, insofar as we Americans can directly
influence decisions in newly democratizing countries,
we should avoid trying to impose it on them. Three,
there is probably no single best constitutional system.
05dahl.091_120 11/27/01 4:40 PM Page 97
And finally, democratic constitutions need to be tai-

lored to fit the culture, traditions, needs, and possibili-
ties of a particular country.
Protecting Fundamental Rights
If the evidence about democratic stability is inconclu-

sive, what does the evidence show about democratic
rights? How well do the constitutional systems of dem-
ocratic countries protect the rights, opportunities, and
duties of both majorities and minorities?
Here we again run into a methodological problem.
As I shall explain in more detail in the next chapter,
democracy and its fundamental institutions presup-
pose the existence of certain fundamental rights, such
as freedom of speech and freedom of the press. We
can reasonably classify all twenty-two of our countries
as democratic only because, among other things, they
all have maintained a high level of protection for basic
democratic rights and liberties. As with stability and
breakdown, in selecting our twenty-two countries we
have necessarily excluded any in which massive and
persistent violations of fundamental political rights
have occurred.
Nonetheless, even if we assume that all of them
have maintained political rights at or above the basic
threshold for democracy, we do find some smaller
variations. The important point, however, is that there
is no discernible relationship between constitutional
05dahl.091_120 11/27/01 4:40 PM Page 98
systems, broadly defined, and these variations in rights

and freedoms. Freedom House, an independent non-
profit organization that since 1973 has provided an-
nual evaluations of the conditions of liberty among the
countries of the world, assigns the identical score on
political rights to all twenty-two democratic countries.
On civil liberties, seven countries Belgium, Costa
Rica, France, Germany, Israel, Italy, and the United
Kingdom fall just below the top score.6 Yet if we go
back to our examination of the variations in broad con-
stitutional features that might explain why these seven
fall below the rest, none can account for the differ-
ence. Federalism, strong bicameralism, unequal rep-
resentation in the upper house, strong judicial review,
the electoral and party systems, and parliamentarism
or presidentialism: none provide an explanation. Or
consider freedom in print and broadcasting media.
At the top, with a near-perfect score in the Freedom
House evaluations, is Norway: a nonfederal country
with a parliamentary system, a unicameral parliament,
proportional representation, multiple parties, coalition
governments, and no judicial review of parliamentary
enactments. Halfway down, just below the United
States, is the Netherlands, another nonfederal country
with a parliamentary system, proportional representa-
tion, multiple parties, coalition governments, and no
judicial review. Why the difference? Or compare four
of the federal countriesSwitzerland, Australia, the
United States, and Germany. Federalism can hardly
account for the variations in their scores.7
05dahl.091_120 11/27/01 4:40 PM Page 99
The most relevant conclusion we can draw is that

among mature democratic countries, where the con-
ditions for democracy are generally favorable, differ-
ences in rights and liberties cannot be attributed to
constitutional systems. But if not constitutional sys-
tems, then what?
The answer will be found, I believe, in differences
in national histories, political cultures, and perceptions
of internal and strategic threats to survival. If this is the
case, then in the end a democratic country cannot de-
pend on its constitutional systems for the preservation
of its liberties. It can depend only on the beliefs and
cultures shared by its political, legal, and cultural elites
and by the citizens to whom these elites are responsive.
Democratic Fairness
How does the American constitutional system com-

pare with those of other mature democracies in the
fairness with which it treats different citizens? As we
all know, the question of fairness or justice has been a
source of endless debate among the best minds since
ancient times. Indeed, differences in views about jus-
tice seem to be built into the human condition. I, how-
ever, want to bypass these perennial controversies and
focus instead on one aspect of fairness that bears di-
rectly on the question at hand.
If I may now use the term constitutional system in
its broader sense to include electoral arrangements,
05dahl.091_120 11/27/01 4:40 PM Page 100
we can arbitrarily reduce the alternatives to two. In

one, which Ill call a proportional system, as a result of
proportional representation the percentage of seats
won by a party in the legislature will roughly mirror the
percentage of votes cast for candidates of that party. In
the other, which Ill call a majoritarian system,8 candi-
dates receiving the most votes in a particular district
win that districts single seat, and the other candidates,
therefore, win no seats at all. In a proportional system,
all minority parties that gain votes above some thresh-
old, such as 5 percent, will be represented in the legis-
lature. In a majoritarian system, if the candidate of
one party were to win a plurality (relative majority) of
votes in every district, then that party would win all
the seats. Although such an extreme outcome is only
a theoretical possibility, in majoritarian systems the
party with a majority of votes does ordinarily win a
disproportionately large number of seats; the second
largest party gains a disproportionately small number
of seats; and all third parties gain few if any seats.
In an earlier chapter I pointed out that propor-
tional representation is likely to produce a multiparty
system and coalition governments; first-past-the-post
is likely to produce two dominant parties; and in a par-
liamentary system with two dominant parties, the
prime minister and cabinet are likely to be drawn from
a single party with a majority of seats, as is typically the
case in Britain.
In the debate over the relative desirability of pro-
portionality versus majoritarianism,9 virtually no one
05dahl.091_120 11/27/01 4:40 PM Page 101
questions that proportionality is fairer to citizens than

majoritarianism. Proportionality doesnt necessarily
mean, however, that the principle of majority rule en-
tirely stops operating. In the legislature, for example,
elected representatives will ordinarily make their deci-
sions by majority rule. But because the governing coali-
tion will generally include representatives from minor-
ity parties, governing majorities are likely to be more
inclusive than in a majoritarian system. Thus, a propor-
tional system comes closer than a majoritarian system to
providing equal representationan equal sayfor all.
Advocates of majoritarianism may concede that
proportionality is fairer; but they might argue that a
majoritarian system offers two advantages that consid-
erably outweigh its unfairness. For one thing, its de-
fenders often say, proportionality tends to produce gov-
erning coalitions that are more unstable and, therefore,
more ineffective than governments in majoritarian sys-
tems. Does the experience of the large number of ma-
ture democratic countries with proportional system
confirm that their governments are less effective? In a
moment Ill turn to evidence bearing on this question.
But just suppose we were to find that proportional
systems are, in general, no less effective than govern-
ments in countries with majoritarian systems. On what
grounds could we then reject proportionality?
We might still reject proportionality if we conclude
that having two dominant parties rather than the mul-
tiplicity of parties typical of proportional systems helps
to make governments more accountable to voters. Our
05dahl.091_120 11/27/01 4:40 PM Page 102
supposition might run something like this: Two-party

majoritarian systems help voters hold governments
more accountable because they are better at simplify-
ing and clarifying the alternatives open to voters. Con-
sequently, during campaigns and elections, voters can
fix responsibility for the decisions and policies that the
government has recently adopted. What is more, be-
cause voters in proportional systems face a multiplicity
of parties and possible governing coalitions, they may
find it difficult to guess what their vote will actually
mean. Forming a majority coalition in a multiparty
parliament can be a tricky business. If a voters party
wants to be included in the governing coalition, what
compromises will it be compelled to make in order to
find a place? And what policies will the coalition fi-
nally manage to agree on and carry through? In con-
trast, because voters in majoritarian systems ordinarily
have only two realistic choices, they can make more
informed guesses about the direction the government
is likely to take under one major party or the other.10
A justification along these lines would provide
strong support for majoritarian systems. But as appeal-
ing as it is, the majoritarian vision isnt easily trans-
formed into reality. For one thing, in the small num-
ber of countries with nominally majoritarian systems,
as Powell points out, we find a persistent refusal of
voters to deliver majority support for a single party or
even a preelection coalition. In forty-five elections
from 1969 through 1994 in six predominantly majori-
tarian countries, only in Australia in 1975 and in
05dahl.091_120 11/27/01 4:40 PM Page 103
France in 1981 did a party or preelection coalition win

a clear voter majority. In short, just as happens in
American presidential elections, majoritarianism often
fails to produce a government that reflects the choices
of a majority of voters. Second, the distortion between
seats and votes in majoritarian systems sometimes cre-
ates a majority of seats for a party that has failed to
win even a plurality of votes and thus has actually
come in second. In these cases, the minority party
among voters becomes the majority party in the legis-
lature. Third, even in majoritarian systems, in prac-
tice, purely two-party politics is a rare phenomenon
and often not robust when it appears. That is, a third
partylike the Liberal Democrats in Britainmay
prevent either of the two major parties from gaining a
majority of votes, even though one of them may gain a
majority of seats.11
Encouraging Consensus
Even if proportionality is fairer than majoritarianism,

many Americans will say that the price of fairness is
too high. Any country in which multiple parties com-
pete for office, they assume, will surely be divided and
contentious and suffer from government by unstable
and ineffective coalitions. How valid is this common
American view?
In direct contradiction to that view, Arend Lijp-
hart, the scholar who pioneered the comparative analy-
05dahl.091_120 11/27/01 4:40 PM Page 104
sis of proportionality and majoritarianism in democratic

countries, refers to proportional systems as consensus
governments.12 And rightly so, for experience shows
that even if proportionality cannot always overcome
deep political, social, cultural, or economic cleavages
(as in Israel, for example), a proportional system can
sometimes help to maintain internal peace, provide
opportunities for compromise among opponents, and
produce a broad consensus in favor of not only gov-
ernment policies but the countrys political arrange-
ments as well.
Let me offer three examples. In the Netherlands,13
religious and ideological differences led to a profound
division of the country into four basic groups: Prot-
estants, Catholics, Liberals, and Socialists. The four
groups became fairly distinctive subcultures pretty
much walled off from one another by their own insti-
tutions, from newspapers and radio to schools, trade
unions, hospitals, marriages, residences, and more.
After proportional representation was introduced
early in the twentieth century, each of the groups also
supported its own separate political party. Not sur-
prisingly, in the late nineteenth and early twentieth
centuries these basic cleavages among the four subcul-
tures led to severe conflicts over education, the fran-
chise, and the rights of labor. By 1910 the disputes had
became so intense that leaders of the four groups be-
came alarmed for the future of the country. Spurred
by their concern, from 1913 to 1917 they not only
managed to negotiate acceptable compromises, but
05dahl.091_120 11/27/01 4:40 PM Page 105
they also agreed that the political parties representing

the four groups would all be represented in the cabi-
net. In short, they created a consensus government.
Despite the persisting cleavages among the four sub-
cultures, a highly institutionalized system of full inclu-
sion endured for half a century, when demographic
changes and a decline in the intensity of the differ-
ences reduced the need for complete inclusion of all
four parties in every cabinet. Even so, and down to the
present day, Dutch governments have continued to
emphasize inclusion and consensus rather than ma-
joritarian control of the government.
Or consider Switzerland with its four national
languages German, French, Italian, and the tiny
Romansh-speaking population; its two major religions,
Protestantism and Catholicism, which were the source
of sanguinary conflict until the middle of the nine-
teenth century; and its two dozen or so cantons, many
of which are internally rather homogeneous in lan-
guage and religion. If you were to reflect on the possi-
bilities of conflict among these subcultures you might
conclude that, like the Balkans, Switzerland must be
forever bubbling over with intense disputes and might
even be on the verge of national disintegration. But the
pragmatism, common sense, and national attachments
of the Swiss enabled them in 1959 to create a propor-
tional system in which representatives of the four
major parties representing the different subcultures
are usually all included in the executive branchthe
federal council or Bundesrat.
05dahl.091_120 11/27/01 4:40 PM Page 106
It is a reasonable conclusion that majoritarian sys-

tems in Switzerland and in the Netherlands would have
made the search for governments based on a broad con-
sensus among different subcultures not only extraordi-
narily difficult but probably downright impossible.
A very different situation exists in Sweden. A highly
homogeneous people (until the recent influx of immi-
grants, at any rate), Swedes have a long and settled
tradition of consensus politics. Although the origins of
the Swedish parliament go back some centuries, de-
mocratization arrived comparatively late. Not until
1917 was the power to choose the prime minister
shifted from the king to the parliament. In that sense,
Swedish democracy dates back to 1917. Proportional
representation had already been introduced in parlia-
mentary elections, however, at the beginning of the
twentieth century. Yet neither proportional represen-
tation nor democratization diminished the long-stand-
ing Swedish tradition of consensus. As a Swedish polit-
ical scientist has written:
In the Swedish political tradition . . . accountability is
seldom mentioned as a value. Instead, legitimacy is pro-
moted by another strategy. By sharing power with the
parties in opposition and including them in the rule of
the country, the government is supposed to be regarded
as representative for the people as a whole and conse-
quently one that all can feel loyal to. To reach consen-
sus, to find a common policy, to capture the will of
the people have been the declared motives of Swedish
politicians. Representativeness is the central norm of
political culture.14
05dahl.091_120 11/27/01 4:40 PM Page 107
From an American perspective, the result can be

unbelievable. In stark contrast to Holland and Switzer-
land, Swedish cabinets have frequently been drawn
from a party or coalition that has actually lacked a ma-
jority of seats in parliament. In the past century, mi-
nority governments have been by far the most com-
mon. The average parliamentary support enjoyed by
governments between 1920 and 1994 has been 41.5
percent. You might well wonder how minority gov-
ernments could ever get anything accomplishedor,
for that matter, remain very long in office. The answer
seems to be that in order to gain a broad consensus
both in the parliament and in the country at large,
even minority governments negotiate with representa-
tives of parties outside the government. In short, even
minority governments govern by consensus.
If you wonder why the Dutch, the Swiss, and the
Swedes prefer proportionality to majoritarianism, the
answer is fairly clear: not only does it seem to them
much fairer but it also helps to achieve and maintain a
broad consensus for government policies.
What is more, proportionality can strengthen con-
sensus not just for policies but for democracy as well.
The reason appears to be that proportionality results
in fewer losers. To clarify this point let me overstate it:
In a majoritarian system the only winners in elections
are the citizens who happen to be in the majority; all
the other citizens, being in the defeated minority, are
losers. By contrast, in proportional systems with con-
sensus governments, everyonewell, almost every-
05dahl.091_120 11/27/01 4:40 PM Page 108
onecan win, not everything they hoped for perhaps

but enough to leave them basically satisfied with their
government.
Lest you suppose that these judgments are nothing
more than interesting speculations, let me cite some
persuasive supporting evidence.15
In a 1990 survey of citizens views in eleven Euro-
pean democracies, respondents were asked how satis-
fied they were with the way democracy works in their
country. They also reported how they voted in their
countrys last national election. Knowing the outcome
of that election in each country, the authors of the study
classified the respondents as winners or losers. The
eleven countries were then arranged from the most ma-
joritarian, Britain, to the most consensual, the Nether-
lands. The results, which the authors of the study
describe as robust, were pretty clear: In the more
consensual countries, losers were almost as satisfied as
the winners with the way democracy worked in their
country. By contrast, in the more majoritarian coun-
tries, losers were much more likely to be dissatisfied.
To describe the results another way, suppose that in
one country 70 percent of the winners but only 40 per-
cent of the losers are satisfied with the working of
democracy, a difference of 30 percent. In another coun-
try, say, 70 percent of the winners and 65 percent of the
losers are satisfied, a difference of only 5 percent. In
the study of eleven European democracies that I just
mentioned, this difference in satisfaction with the way
05dahl.091_120 11/27/01 4:40 PM Page 109
democracy worked decreased steadily from large differ-

ences in the most majoritarian countries like Britain
around 25 percentto almost negligible differences
in the most consensual countries like the Netherlands,
where it was less than 5 percent.16 Moreover, these re-
sults held up even when variations in such influences
as economic performance, socioeconomic status, and
political interest were taken into account.17
In short, if you live in a majoritarian country and
your party comes in second or worse, youre likely to
be dissatisfied with the way democracy works in your
country. But if you live in a democratic country with a
more consensual system and your party comes in sec-
ond, or third, or maybe even fourth, youre likely to
feel satisfied with the way democracy works because
you know your views will still be represented in the
government.
This is all fine, you might say, but can a consensus
system produce effective governmentsgovernments
that can solve the problems that concern citizens?
Might not majoritarian governments perform more ef-
fectively? In particular, hasnt our American constitu-
tional system been just as effective as many consensus
governments and perhaps even more effective than
most in getting things done that citizens want? Ill turn
to this question in a moment.
But before I do so, I want to call your attention to
one salient aspect of our constitutional system: The
American constitutional system is not majoritarian.
05dahl.091_120 11/27/01 4:40 PM Page 110
110
The American Hybrid
Whatever its hypothetical advantages might be, the

majoritarian vision is not applicable to the American
system of government. Although our system is not pro-
portional, neither is it majoritarian. Whether by rational
intention or bv an understandable inability to foresee
the consequences, or both, James Madison and his fel-
low delegates created a constitutional system that is a
hybrid of proportionality and majoritarianism.
Three majorities. It is true that two parties domi-
nate our political landscape more completely than in
any other established democracy. Yet even when one
party wins not only the presidency but also majorities
in both the Senate and House, three different popular
majorities are at work; the composition of each of the
three majorities does not match the others; and their
representatives do not necessarily agree. I would not
say that this feature is necessarily undesirable, but un-
desirable consequences do show up when it is com-
bined with other aspects of our system.
Divided government. To begin with, one party
may fail to win control of all three branches. Indeed,
during the past half-century, control of the presidency
and both houses of Congress by a single party has be-
come a rarity. As David Mayhew comments in Divided
We Govern, Since World War II, divided party con-
trol of the American national government has come
to seem normal.18 From 1946 to 2000, the three
05dahl.091_120 11/27/01 4:40 PM Page 111
branches have been divided between the two parties

more than six years out of every ten. Our constitution
not only permits divided government, it cannot pre-
vent divided government. And it provides no way out
except by elections at fixed intervalselections that
may only reproduce the existing divisions or inaugu-
rate new ones.
Does divided government matter? In particular, do
periods of divided government make it more difficult
for the three branches to agree on national policies re-
quiring legislationin short, do these periods make a
stalemate easier? The evidence is mixed. In a land-
mark study of the period between the elections of
1946 and 1990, David Mayhew found no relation
worth crediting between the incidence of important
laws and whether party control was unified or di-
vided.19 A subsequent analysis of the period from
1947 to 1994, however, concluded that important leg-
islation is more likely to be passed in periods of uni-
fied government. The difference between divided and
unified government is particularly marked when all
three branches are under the unified control of the
more activist of the two parties, the Democrats.20
The president: consensual, majoritarian, neither,
both? At the apex of this complex structure of politi-
cal institutions sitsor stands, as the case may be
the American presidency, an office with no equivalent
in any of the other established democracies or, so far
as I am aware, in any other democratic country.
05dahl.091_120 11/27/01 4:40 PM Page 112
It is difficult, indeed impossible, to fit the pres-

idency into the simple categories of consensual or
majoritarian. One obstacle to straightforward classifi-
cation is the presidents combination of roles. Most no-
tably, whereas in the other older democracies the roles
of prime minister and ceremonial head of state are
separated, in our system they are blended, not only
constitutionally but also in popular expectations. We
expect our president to serve both as chief executive
and as a sort of ceremonial, dignified, American-style
elected monarch and moral exemplar.
The mix of roles was present from the beginning.
Although during the early years the vituperation of
presidents in the press often far exceeded boundaries
now regarded as acceptable, in order to maintain the
dignity of the office presidents rarely addressed the
general public, except possibly on official occasions;
and when they did they rarely employed popular rhet-
oric or discussed their own policies. In these respects,
if no others, they acted less like politicians than mon-
archs or ceremonial heads of state. Indeed, until the
1830s, presidential candidates did not make campaign
speeches at all; and until Woodrow Wilson broke a
century-old taboo in 1912, no president had ever
stumped on his own behalf.21
Beginning with Andrew Jackson, however, presi-
dents had already begun to make the audacious claim
that by virtue of their election, they alone represented
the entire people, or at least a majority. Some would
05dahl.091_120 11/27/01 4:40 PM Page 113
even assert that their election endowed them with a

mandate for their policies. To the extent that the
claim of a mandate was accepted, it would increase the
acceptability of a presidents policies by clothing them
with the legitimacy of popular sovereignty.
Despite the frequency with which newly elected
presidents stake a claim to a mandate, the closer you
inspect the chain of assumptions that are supposed to
support the claim, the more fragile the links appear.22
It requires an extraordinary leap of faith to infer the
views of voters from nothing more than the way they
cast their votes for president. Although systematic
opinion surveys provide a much firmer basis for un-
derstanding public attitudes and expectations, the
claim to a presidential mandate based on nothing
more than the outcome of an election antedated sys-
tematic surveys by more than a century. And even
since the beginning of systematic opinion surveys in
the 1940s, presidents and their followers (and pundits)
have typically based their shaky claim to a mandate on
nothing more than the election returns, which could
not possibly reveal whether presidential policies ac-
corded with voters preferences. Since the 1940s that
concordance is more likely to come from scrupulous
attention to public opinion surveys than from reading
the tea leaves of the election results.
Presidential claims to represent the American
people together with efforts to promote particular na-
tional policies are elements in the general mixture of
05dahl.091_120 11/27/01 4:40 PM Page 114
roles that distinguishes the American presidency and

makes it neither simply majoritarian nor simply con-
sensual.
This mix of roles seems to be generally accepted
by Americans. We want our presidents, it seems, to be
simultaneously shrewd politicians and gifted states-
men. We expect them to live in both the real world of
daily politics and an imaginary world above politics.
Most of us understand that to succeed in office a pres-
ident must be an active and forceful partisan, a party
leader, and the negotiator and deal-maker who mas-
sages, cajoles, bribes, threatens, and coerces Congress
to secure both votes and support if promises and poli-
cies are to be realized.
But we also expect our president to serve as a
moral example for us all, to stand as an icon on which
we can devoutly project qualities of intelligence, knowl-
edge, understanding, compassion, and character far
above those we expect in ordinary beings like our-
selves. Because no mortal can meet these exalted stan-
dards, we have often savaged a president while he is in
office and then exalted him in memory. In office, we
may portray a president as a cartoonish bungler. But
after departing the White House, or this world, we ig-
nore the warts and scars and paint an idealized por-
trait of a noble and exemplary figure.
Ambivalence toward the presidency is deeply in-
grained in our American culture. As children we learn
to worship our presidents for their greatness;23 as
adults we deride them for failing to achieve the great-
05dahl.091_120 11/27/01 4:40 PM Page 115
ness of their mythic predecessors. In choosing among

candidates for the presidency we yearn for perfection;
yet our only realistic choices are among flawed human
beings who live with all the moral ambiguities re-
quired by the life of politics. In short, the impossible
mix of roles an American president is expected to play
places a heavy burden not only on the incumbent, but
more important, on American voters.
Accountability. Holding the government account-
able for its actions may be an even greater burden for
voters. Where are we to place responsibility for the
conduct of our government? When we go to the polls,
whom can we hold accountable for the successes
and failures of national policies? The president? The
House? The Senate? The unelected Supreme Court?
Or, given our federal system, the states, where govern-
ments are, in their complexity, a microcosm of the na-
tional government?
Even for those who spend their lives studying poli-
tics, these can be extremely difficult questions to an-
swer. I, for one, am inclined to think that compared
with the political systems of the other advanced demo-
cratic countries, ours is among the most opaque, com-
plex, confusing, and difficult to understand.
We see, then, that our hybrid system, which is nei-
ther majoritarian nor proportional, may possess the
advantages of neither and the defects of both. If it fails
to ensure the fairness promised by the proportional vi-
sion, it also fails to provide the clear accountability
promised by the majoritarian vision.
05dahl.091_120 11/27/01 4:40 PM Page 116
116
Democratic Effectiveness
To all this you might say: Even if the American hybrid

may have some shortcomings when it is viewed in a
comparative perspective, isnt it just as effective as
other governments in dealing with the issues that con-
cern American citizens?
Once again, we cannot answer this question re-
sponsibly without facing up to some severe method-
ological problems. Our twenty-two democratic coun-
tries differ in so many ways that teasing out the effects
that we can reasonably attribute to constitutional sys-
tems is a pretty formidable task. Take size. The popu-
lation of the United States is sixty times larger than
Norways, fifty times Denmarks, thirty-seven times
Switzerlands, thirty times Swedens, and nearly a thou-
sand times larger than Icelands, which is somewhat
smaller in population than Tampa, Florida.
Although the effects of population size on demo-
cratic political life are extremely difficult to measure,
they can hardly be ignored.24
Or consider diversity: in a very general way, diver-
sity tends to increase with size.25 Yet can we truly say
that the United States is more diverse than Switzer-
land, or than our neighbor Canada?
Add in one more variable: relative affluence. Al-
though Norway and Costa Rica are both relatively small
in populationthere are around 4.5 million Norwe-
gians and around 3.7 million Costa RicansNorways
GNP per capita is fourteen times that of Costa Rica.26
05dahl.091_120 11/27/01 4:40 PM Page 117
How much do these differences in size, diversity,

and relative affluence affect political life and public
policy? Despite the difficulties presented by national
variations like these, comparative data can help us gain
some appreciation of the way in which American per-
formance compares with that of other advanced dem-
ocratic countries.27 When the United States is ranked
with other established democracies on such matters as
the rate of incarceration, the ratio of poor to rich, eco-
nomic growth, social expenditures, energy efficiency,
foreign aid and the like, its performance is something
less than impressive. (See Appendix B, Table 5.) Two
areas in which our country ranks highest are hardly
achievements of which we can be proud. On the per-
centage of the population we incarcerate, we come out
a clear winner, while our ratio of rich to poor is higher
than that of most other countries. We rank in the bot-
tom thirdand on some measures close to the bottom
of the bottom thirdon voter turnout, state welfare
measures, energy efficiency, and the representation of
women in the national legislature. What is more, in
spite of our good showing on economic growth, we are
almost dead last in our social expenditures. Finally,
even though many Americans believe that we are too
generous in our economic aid to other countries,
among nineteen democratic countries we are at the
very bottom.
In his comparison of consensual and majoritarian
systems in thirty-six countries, Arend Lijphart con-
cluded that majoritarian democracies do not out-
05dahl.091_120 11/27/01 4:40 PM Page 118
perform the consensus democracies on macroeconomic

management and the control of violencein fact, the
consensus democracies have the slightly better rec-
ordbut the consensus democracies do clearly out-
perform the majoritarian democracies with regard to
the quality of democracy and democratic representa-
tion as well as with regard to what I have called the
kindness and gentleness of their public policy orienta-
tions.28
Paraphrasing Lijpharts conclusion, I find no con-
vincing evidence that our hybrid outperforms systems
that are either more fully consensual systems or more
fully majoritarian. On the contrary, compared with
other democratic countries our performance appears,
on balance, to be mediocre at best.
How much does our performance have to do with
our constitutional system? To tease out the extent of
that connection would be extraordinarily difficult, per-
haps impossible, and I am going to leave that task to
others.29
It seems reasonably clear, however, that a constitu-
tional system better designed to achieve such demo-
cratic goals as the protection of fundamental rights,
fair representation, and greater consensus does not
necessarily come at the price of governmental effec-
tiveness, much less the stability of the democratic sys-
tem itself.
If this is so, then do we not have every reason to
undertake a serious and responsible examination of pos-
05dahl.091_120 11/27/01 4:40 PM Page 119
sible alternatives to our present American Constitution?

Or, at the very least, isnt it timewell-past timethat
we stop thinking of our Constitution as a sacred text
and begin to think of it as nothing more, or less, than a
means for achieving democratic goals?
05dahl.091_120 11/27/01 4:40 PM Page 120
06dahl.121_140 11/27/01 4:40 PM Page 121
chapter 6
Why Not a More Democratic

Constitution?
I
BEGAN BY POSING THIS QUESTION: WHY SHOULD WE
Americans uphold our Constitution? Let me now
change the question slightly: What kind of consti-
tution should we feel obliged to uphold?
I mean, of course, an American constitutionnot
necessarily our present Constitution, but a constitution
that, after careful and prolonged deliberation, we and
our fellow citizens conclude is best designed to serve
our fundamental political ends, goals, and values.
The Constitution as National Icon
I am well aware that in expressing reservations about

the Constitution, as I have in these essays, I may be
06dahl.121_140 11/27/01 4:40 PM Page 122
122 w h y n ot a m o r e d e m o c r at i c c o n st i t u t i o n ?
judged guilty of casting stones at a national icon.

From the time of the founding fathers, a historian
has recently remarked, there [has] been a sacred aura
about the Constitution, manifest in holiday political
rhetoric. During the years between the two World
Wars, worship of the Constitution acquired the trap-
pings of a religious cult.1 This reverential attitude con-
tinues. In a telephone survey of one thousand adult
U.S. citizens in 1997, 71 percent said they strongly
agreed with the statement that they were proud of the
Constitution; another 20 percent somewhat agreed.2
In a 1999 survey, 85 percent said they thought the
Constitution is a major reason why America has been
successful during this past century.3
I dont dismiss the importance of icons for strength-
ening beliefs, religious or political, nor do I dismiss
the utility of myth and ritual in helping to foster na-
tional cohesion. But a faith that rests on little more
than a general conformity to conventional beliefs is a
fragile foundation for nationhoodnot to say for de-
mocracy. I want therefore to suggest an alternative.
The only legitimate constitution for a democratic
people, it seems to me, is one crafted to serve demo-
cratic ends. Viewed from this perspective an American
constitution ought to be the best that we can design for
enabling politically equal citizens to govern themselves
under laws and government policies that have been
adopted and are maintained with their rational consent.
This is hardly a novel view. What I am suggesting is
that a constitution derives its legitimacy from a moral
06dahl.121_140 11/27/01 4:40 PM Page 123
w h y n ot a m o r e d e m o c r at i c c o n st i t u t i o n ? 123
and political judgment announced to the world more

than two centuries ago. This judgment (slightly modi-
fied from the original) asserts:
That all human beings are created equal, that they are
endowed by their Creator with certain inalienable
Rights, that among these are Life, Liberty, and the Pur-
suit of Happiness. That to secure these rights, Govern-
ments are instituted among a people, deriving their just
powers from the consent of the governed. That when-
ever a Form of government becomes destructive of these
ends, it is the Right of the people to alter or abolish it,
and to institute new government, laying its foundation
on such principles and organizing its powers in such a
form, as to them shall seem most likely to effect their
Safety and Happiness.
But two questions immediately arise. First, is polit-

ical equality a realistic goal? Second, is it really a desir-
able goal?4
Is Political Equality a Realistic Goal?
Some of you may dismiss the noble words I have just

quoted as obviously false. If anything about equality is
self-evident, you might object, it is that human beings
arent equal. Whether by genes, birth, luck, achieve-
ments, or whatever, we arent equal in education, cul-
tural endowments, social and communication skills, in-
telligence, motor skills, incomes, wealth, the country
in which we live, and so on. Though this objection is a
06dahl.121_140 11/27/01 4:40 PM Page 124
commonplace, it wholly misses the point. The men who

wrote and adopted the American Declaration of Inde-
pendence hardly needed to be reminded of such ele-
mentary matters. They knew too much about the ways
of the world to make assertions that were obviously con-
tradicted by everyday human experience. But, of course,
they didnt mean the Declaration to be understood as
a statement of fact. They meant it to be understood as
a moral statement. Human equality, they were insist-
ing, is a moral and even a religious standard against
which it is right and proper to judge a political system.
Yet ideal standards might rise so far beyond human
reach as to be irrelevant. Is political equality so remote
from human possibilities that we might just as well
forget it?
I need hardly remind you of the enormous and
persistent barriers to political equality and, indeed, to
human equality in general.5 Consider that elemental
and age-old barrier arising from differences in the
treatment of men and women. The authors of the fa-
miliar words about equality that I just quoted and the
fifty-five delegates to the Second Continental Con-
gress who voted to adopt the Declaration in July 1776,
were, of course, all men, none of whom had the slight-
est intention of extending the suffrage or many other
basic political and civil rights to womenwho by the
laws of that time and for a full century after were the
legal property of their fathers or husbands.
Nor did the worthy supporters of the Declaration
mean to include slaves or, for that matter, free persons
06dahl.121_140 11/27/01 4:40 PM Page 125
of African origin, who were a substantial fraction of the

population in almost all the colonies that claimed the
right to become independent self-governing republics.
The principal author of the Declaration, Thomas Jef-
ferson, owned several hundred slaves, none of whom
he freed during his life.6 It was not until more than
four score and seven years later (to borrow a poetic
phrase from Lincolns Gettysburg Address) that slav-
ery was legally abolished in the United States by force
of arms and constitutional enactment. And it took yet
another century before the rights of African Ameri-
cans to participate in political life began to be effec-
tively enforced in the American South. Now, two gen-
erations later, Americans white and black still bear the
deep wounds that slavery and its aftermath inflicted
on human equality, freedom, dignity, and respect.
Nor did our noble Declaration mean to include the
people who for thousands of years had inhabited the
lands that Europeans colonized and came to occupy.
We are all familiar with the story of how the settlers de-
nied homes, land, place, freedom, dignity, and human-
ity to these earlier peoples of America, whose descen-
dants even today continue to suffer from the effects of
their treatment throughout several centuries, when
their most elementary claims to legal, economic, and
politicalnot to say socialstanding as equal human
beings were rejected, often by violence; more recently,
this lengthy period has been followed by neglect and
indifference.
All this in a country that visitors from Europe, such
06dahl.121_140 11/27/01 4:40 PM Page 126
as Alexis de Tocqueville, portrayed (correctly, I think)

as displaying a passion for equality stronger than they
had ever observed elsewhere.
Yet, despite the fact that throughout human history
equality has often been denied in practice, throughout
the past several centuries many claims to equality, in-
cluding political equality, have come to be much more
strongly reinforced by institutions, practices, and be-
havior. Although this monumental historical move-
ment toward equality is in some respects worldwide, it
has been most conspicuous in such democratic coun-
tries as Britain, France, the United States, the Scandi-
navian countries, the Netherlands, and others.
In the opening pages of the first volume of Democ-
racy in America, Tocqueville pointed to the inexorable
increase in the equality of conditions among his
French countrymen at intervals of fifty years, begin-
ning with the eleventh century. Nor was this revolu-
tion taking place in only his own country: Whitherso-
ever we turn our eyes, he wrote, we shall witness the
same continual revolution throughout the whole of
Christendom. He goes on to say, The gradual devel-
opment of the equality of conditions is . . . a providen-
tial fact, and it possesses all the characteristics of a Di-
vine decree: it is universal, it is durable, it constantly
eludes all human interference, and all events as well as
men contribute to its progress.7
We may wish to grant Tocqueville a certain meas-
ure of hyperbole in this passage. We may also want
to note that in the second volume that he published
06dahl.121_140 11/27/01 4:40 PM Page 127
several years later, he was more troubled by what

he viewed as some of the undesirable consequences
of democracy and equality. I shall return to his con-
cerns in a moment. Even so, he did not doubt that a
continuing advance of democracy and equality was in-
evitable. And if we look back today to the changes
since his time, like Tocqueville in his own day we may
well be amazed at the extent to which ideas and prac-
tices that respect and promote political equality have
advanced across so much of the worldas, for that
matter, have some aspects of a broader human equal-
ity as well.
As to political equality, consider the incredible
spread of democratic ideas, institutions, and practices
during the century that just ended. In 1900, 48 coun-
tries were fully or moderately independent countries.
Of these, only 8 possessed all the other basic institu-
tions of representative democracy, and in only 1 of
these, New Zealand, had women gained the right to
vote. Furthermore, these 8 countries contained no
more than 1012 percent of the worlds population. At
the opening of our present century, among some 190
countries the political institutions and practices of
modern representative democracy, including universal
suffrage, exist in around 85, at levels comparable to
those in Britain, western Europe, and the United
States. These countries include almost 5 out of every
10 inhabitants of the globe today.8
In Britain, the working classes and women were
enfranchised, and more. Men and women of middle,
06dahl.121_140 11/27/01 4:40 PM Page 128
lower-middle, and working-class origins gained access

not only to the House of Commons and its facilities
but to the cabinet and even the post of prime minister.
And the hereditary peers in the House of Lords have,
after all, at last been sent packingwell, most of them
anyway. In the United States, too, women were en-
franchised. The Civil Rights Act of 1964, which pro-
tected the right of African Americans to vote, did in
fact become law; the law was actually enforced; and
African Americans have become a significant force in
American political life. I wish I could say that the mis-
erable condition of so many Native Americans had
greatly changed for the better, but that sad legacy of
human injustice remains with us.
Although we must admit to persistent failures and
continuing obstacles, if we assume that beliefs about
equality are hopelessly anemic contestants in the
struggle against the powerful forces that generate in-
equalities, we cannot account for the enormous gains in
human equality achieved over the past two centuries.
How Does Greater Political Equality

Come About?
In the face of so many obstacles, how does greater

equalityor better, a reduction in some inequalities
ever come about? Although no brief summary can do
justice to an explanation of the historical variations and
complexities in the process by which changes toward
06dahl.121_140 11/27/01 4:40 PM Page 129
equality take placeand here I have in mind mainly

political equalitya summary of the most important
elements would probably run something like this:
Despite fervent efforts by privileged elites to pro-
mote views intended to give legitimacy to their supe-
rior power and status, together with their own unques-
tioning belief in the rightness of their entitlements
(think of the Federalists!), many members of subordi-
nate groups doubt that the inferior position assigned
to them by their self-proclaimed superiors is really jus-
tified. James Scott has shown pretty convincingly that
people who have been relegated to subordinate status
by history, structure, and elite belief systems are much
less likely to be taken in by the dominant ideology
than members of the upper strata are prone to as-
sume.9 Given the open or concealed rejection of the
elite ideology by members of the subordinate groups,
a change in conditions, whether in ideas, beliefs, gen-
erations, structures, resources, or whatever, begins to
offer the subordinate groups new opportunities to ex-
press their grievances. And given these new opportu-
nities and moved by anger, resentment, a sense of
injustice, a prospect of greater individual or group
opportunities, group loyalty, or other motives, some
members of the subordinate groups begin to press for
change by whatever means are available. Some mem-
bers of the dominant group begin to support the
claims of the subordinate strata. Privileged insiders
ally themselves with outsiders. Insiders may do so for a
variety of reasons: moral convictions, compassion, op-
06dahl.121_140 11/27/01 4:40 PM Page 130
portunism, fear of the consequences of disorder, dan-

gers to property and the legitimacy of the regime aris-
ing from widening discontent, and even the real or
imagined possibility of revolution.10
So a seismic shift occurs: extension of the fran-
chise, legal protection of basic rights, political compe-
tition from leaders of hitherto subordinate groups,
election to public office, changes in law and policy,
and so on. In the United States, Civil Rights Acts were
passed in 1957, 1960, and most crucial of all, 1964.
What is more, they were enforced. African Americans
began to seize their opportunities to voteand among
other things soon tossed out the police officials who
had violently enforced their subordination. In India,
the scheduled castes have begun to vote in substantial
numbers for leaders and parties who are drawn from
their own strata and committed to reducing discrimi-
nation against them. Though changes toward equality
may be and typically are incremental, a series of incre-
mental changes can, in time, mount to a revolution.
By such processes, then, a certain measure of po-
litical equality and democracy have been obtained in
some countries despite enormous and persistent ob-
stacles to human equality.
Is Political Equality a Justifiable Goal?
Yet even if a greater degree of political equality and

democracy can be achieved, are these goals really de-
06dahl.121_140 11/27/01 4:40 PM Page 131
sirable? Whats more, are they so desirable that we

should make the constitution of a democratic country
in particular that of the United Statessubordinate to
achieving these ends?
The desirability of political equality and thus of
democracy follows, in my view, from two fundamental
judgments. One is moral, the other practical.
The moral judgment holds that all human beings
are of equal intrinsic worth; that no person is intrinsi-
cally superior in worth to another; and that the good
or interests of each person ought to be given equal
consideration.11 Let me call this the assumption of in-
trinsic equality.
Yet if we accept this moral judgment, a deeply
troublesome question immediately arises: Who or what
group is the best qualified to decide what the good or
interests of a person really are? Clearly the answer will
vary, depending on the situation, the kinds of deci-
sions, and the persons involved. But if we restrict our
focus to the government of a state, then it seems to me
that the safest and most prudent assumption would
run something like this: Among adults, none are so
better qualified than others to govern that they should
be entrusted with complete and final authority over
the government of the state.12
Although we might reasonably add refinements
and qualifications to this prudential judgment, it is dif-
ficult for me to see how a significantly different propo-
sition could be defended, particularly if we draw on
crucial historical cases in which substantial numbers of
06dahl.121_140 11/27/01 4:40 PM Page 132
persons were denied equal citizenship. Does anyone

today really believe that when the working classes,
women, and racial and ethnic minorities were excluded
from political participation, their interests were ade-
quately considered and protected by those who were
privileged to govern over them?
Does Political Equality Threaten Liberty?
Like many desirable goals, political equality might

conflict withand may indeed do harm toother im-
portant goals, ends, values. If so, shouldnt our pursuit
of political equality be tempered by our justifiable de-
sire to attain these other goals?
It is frequently said that equality conflicts with lib-
erty and fundamental rights. Like many others, Toc-
queville appears to have believed so.
But before I turn to his remarks, I cannot forgo
adding that I am amazed by a frequent assertion about
the supposed conflict between liberty and equality that
makes no mention of what would seem to me to be an
absolutely essential requirement of any reasonable dis-
cussion about the relation between the two. Whenever
we talk about liberty, freedom, or rights, are we not
obliged to answer the question: Liberty or rights for
whom? When we speak of freedom, liberty, or rights it
seems to me essential that we go beyond answering
the question, What liberty or right? An answer to
that question only specifies the domain of liberty. But
06dahl.121_140 11/27/01 4:40 PM Page 133
we are also obliged to answer the question, Liberty

for whom?13
Keeping this question in mind, let me return to
Tocqueville. His view, if I understand him correctly,
was roughly this: An equality of condition among a
people helps to make democracy possible, perhaps
even inevitable. But the very equality of condition that
makes democracy possible also carries dangers to lib-
erty. Let me paraphrase Tocqueville:
Since the very essence of democratic government is the
absolute sovereignty of the majority, which nothing in
democratic states is capable of resisting, a majority nec-
essarily has the power to oppress a minority. Just as a
man with absolute power may misuse it, so may a ma-
jority. Given an equality of condition among citizens,
we may expect that in democratic countries a wholly
new species of oppression will arise. Among citizens all
equal and alike, the supreme power, the democratic
government, acting in response to the will of the major-
ity, will create a society with a network of small compli-
cated rules, minute and uniform, that none can escape.
Ultimately, then, the citizens of a democratic country
will be reduced to nothing better than a flock of timid
and industrious animals, of which the government is the
shepherd.14
If I have fairly summarized Tocqueville, how

should we interpret his forecast in the light of subse-
quent developments? After all, we have the advantage,
as he did not, of two centuries of experience with mod-
ern democratic institutions. Some readers have inter-
preted these passages in Tocqueville as foreshadowing
06dahl.121_140 11/27/01 4:40 PM Page 134
mass society, while to others Tocqueville expected that

mass democracy would be the seed of twentieth-
century authoritarian and totalitarian systems. Yet, if
we read the passages as a forecast of the way in which
democratic countries would tend to evolve, I think we
are bound to conclude that Tocqueville was just dead
wrong. When we examine the course of democratic
development over the past two centuries, and particu-
larly over the century just ended, what we find is a
pattern of democratic development that stands in total
contradiction to such a prediction. We find instead
that as democratic institutions become more deeply
rooted in a country, so do fundamental political rights,
liberties, and opportunities. As a democratic govern-
ment matures in a country, the likelihood that it will
give way to an authoritarian regime approaches zero.
Democracy can, as we all know, collapse into dictator-
ship. But breakdowns are extraordinarily rare in ma-
ture democracies; they occur instead in countries that
encounter times of great crisis and stress when their
democratic institutions are relatively new. Crisis ap-
pears to be inevitable in the life of every country. Even
mature democratic countries have had to face wars,
economic depression, large-scale unemployment, ter-
rorism, and other challenges. But they did not collapse
into authoritarian regimes.
In the twentieth century, on something like sev-
enty occasions democracies have given way to nonde-
mocratic regimes. Yet with very few exceptions, these
breakdowns have occurred in countries where demo-
06dahl.121_140 11/27/01 4:40 PM Page 135
cratic institutions were very newless than a genera-

tion old. Indeed, the only clear-cut case of a demo-
cratic breakdown in a country where democratic insti-
tutions had existed for twenty years or more seems to
be Uruguay in 1973. In the same year, Chile provided
a less clear-cut case because of restrictions on the suf-
frage that had only recently been lifted. The Weimar
Republic had existed fewer than fourteen years before
the Nazi takeover. In all three countries the path to col-
lapse bore no relation to the Tocquevillean scenario.
Nor, as we know, is that scenario confirmed by the
older or mature democracies. As I indicated in the
previous chapter, we can find some small variations
among these countries in their protection of basic
rights. But they all maintain these rights well above
the threshold necessary for democracy. Have the fun-
damental rights and liberties of citizens grown steadily
narrower or less secure over the past half-century? I
do not see how an affirmative answer to this question
could be seriously maintained. Much as I admire Toc-
queville, on this issue, he, like the Framers, could not
foresee the future of democratic government.
Far from being a threat to fundamental rights and
liberties, political equality requires them as anchors
for democratic institutions. To see why this is so, let
me once again view democracy as, ideally at least, a
political system designed for citizens of a state who are
willing to treat one another, for political purposes, as
political equals. Citizens might view one another as
unequal in other respects. Indeed, they almost cer-
06dahl.121_140 11/27/01 4:40 PM Page 136
tainly would. But if they were to assume that all citi-

zens possess equal rights to participate, directly or in-
directly through their elected representatives, in mak-
ing the policies, rules, laws, or other decisions that
citizens are expected (or required) to obey, then the
government of their state would, ideally, have to satisfy
several criteria.
Let me list them here without amplification. To be
fully democratic, a state would have to provide: rights,
liberties, and opportunities for effective participation;
voting equality; the ability to acquire sufficient under-
standing of policies and their consequences; and the
means by which the citizen body could maintain ade-
quate control of the agenda of government policies
and decisions. Finally, as we now understand the ideal,
in order to be fully democratic, a state would have to
ensure that all, or at any rate most, permanent adult
residents under its jurisdiction would possess the rights
of citizenship.
As we know, the democratic ideal that I have just
described is too demanding to be achieved in the ac-
tual world of human society. To accomplish it as far as
may be possible under the imperfect conditions of the
real world, certain political institutions for governing
the state would be required. Moreover, since the eigh-
teenth century these institutions have had to be suit-
able for governing a state encompassing a large terri-
tory, such as a country.
There is no need to describe here the basic politi-
cal institutions of a modern democratic country; but it
06dahl.121_140 11/27/01 4:40 PM Page 137
should be obvious that just as in the ideal, so too in ac-

tual practice, democratic government presupposes that
its citizens possess a body of fundamental rights, liber-
ties, and opportunities. These include the rights to vote
in the election of officials in free and fair elections; to
run for elective office; to free expression; to form and
participate in independent political organizations; to
have access to independent sources of information;
and to have rights to other freedoms and opportunities
that may be necessary for the effective operation of
the political institutions of large-scale democracy.
Both as an ideal and as an actual set of political in-
stitutions, democracy is necessarily, then, a system of
rights, liberties, and opportunities. These are required
not merely by definition. They are required in order
for a democratic system of government to exist in the
real world. If we consider these political rights, liber-
ties, and opportunities as in some sense fundamental,
then in theory and in practice, democracy does not
conflict with liberty. On the contrary, democratic insti-
tutions are necessary for the existence of some of our
most fundamental rights and opportunities. If these
political institutions, including the rights, liberties, and
opportunities they embody, do not exist in a country,
then to that extent the country is not democratic.
When they disappear, as they did in Weimar Germany,
Uruguay, and Chile, then democracy disappears; and
when democracy disappears, as it did in these coun-
tries, then so do these fundamental right, liberties, and
opportunities. Likewise, when democracy reappeared
06dahl.121_140 11/27/01 4:40 PM Page 138
in these countries so, necessarily, did these fundamen-

tal rights, liberties, and opportunities. The connection,
then, is not in any sense accidental. It is inherent.
The links between political equality and democ-
racy, on the one hand, and fundamental rights, liber-
ties, and opportunities, on the other, run even deeper.
If a country is to maintain its democratic institutions
through its inevitable crises, it will need a body of
norms, beliefs, and habits that provide support for the
institutions in good times and bada democratic cul-
ture transmitted from one generation to the next. But
a democratic culture is unlikely to be sharply bounded.
A democratic culture will support not only the funda-
mental rights, liberties, and opportunities that demo-
cratic institutions require. People who share a demo-
cratic culture will, I think inevitably, also endorse and
support an even greater sphere of rights, liberties, and
opportunities. Surely the history of recent centuries
demonstrates that it is precisely in democratic coun-
tries that liberties thrive.
! ! !
IF WE BELIEVE THAT ALL HUMAN BEINGS ARE CREATED

equal, that they are endowed with certain inalienable
rights, that among these are life, liberty, and the pur-
suit of happiness, that to secure these rights govern-
ments are instituted among a people, deriving their
just powers from the consent of the governed, then we
are obliged to support the goal of political equality.
06dahl.121_140 11/27/01 4:40 PM Page 139
Political equality requires democratic political in-

stitutions.
The supposed conflict between liberty and politi-
cal equality is spurious, first, because an inherent part
of democratic political institutions is a substantial body
of fundamental rights, liberties, and opportunities; and,
second, because a people committed to democracy and
its political institutions will almost certainly expand
the sphere of fundamental rights, liberties, and oppor-
tunities well beyond those strictly necessary for de-
mocracy and political equality.
Among a people committed to democracy and po-
litical equality, a constitution should serve those ends
by helping to maintain political institutions that foster
political equality among citizens and all the necessary
rights, liberties, and opportunities that are essential to
the existence of political equality and a democratic
government.
06dahl.121_140 11/27/01 4:40 PM Page 140
07dahl.141_158 11/27/01 4:41 PM Page 141
chapter 7
Some Reflections on the Prospects

for a More Democratic Constitution
I
N A1987 SURVEY THAT REVEALED STRONG SUPPORT
among Americans for the Constitution on the
whole, the results of one question stand out. When
respondents were asked, How good a job has [the sys-
tem of government established by the Constitution]
done in treating all people equally? Fifty-one percent
answered that it had done a bad job, 8 percent ex-
pressed no opinion, and a minority of 41 percent said
that it had done a good job.1
If we want to enjoy a system of government that
performs better in treating all people equallyat least
in their roles as democratic citizenswhat might we
do? As I said at the beginning of Chapter 1, my aim in
these essays is not to offer a set of specific proposals
for changes in the Constitution but rather to encour-
07dahl.141_158 11/27/01 4:41 PM Page 142
142 p ro s p e c t s f o r a m o r e d e m o c r at i c c o n st i t u t i o n
age a change in the way we think about our Constitu-

tion. How then might we begin to think realistically
about changing it? What possibilities would actually be
open to a body of framers in the early years of the
twenty-first century? What limits on the range of pos-
sibilities should they expect to confront?
The Limited Role of a Constitution
To begin with, tomorrows constitutional reformers

would be wise to recognize that no matter what a consti-
tution prescribes on paper, it can achieve only a limited
range of goals. For example, as I pointed out in Chapter
5, no constitution can ensure democracy in a country
where the conditions favorable to democracy are ab-
sent. To preserve and improve these favorable condi-
tions would accomplish far more in achieving a more
democratic order than any changes in the constitution.
The Framers of 1787 were well aware of their lim-
its. One of the striking features of the constitution
they wrote is its admirable brevity. Depending on the
style in which it is printed, the written text runs fifteen
to twenty pages, plus another five to seven pages of
amendments. What enables the written Constitution
to achieve this brevity is its almost exclusive focus on
just three matters: structures, powers, and rights.
Most of the Constitution is devoted to the first two
matters, structures and the powers allocated to these
structures. Provisions about the third matter, rights, are
07dahl.141_158 11/27/01 4:41 PM Page 143
p ro s p e c t s f o r a m o r e d e m o c r at i c c o n st i t u t i o n 143
found mainly in the Bill of Rights and in later amend-

ments. An important feature of these constitutional
rights is that they are guaranteed almost entirely by
imposing constitutional limits on the government. The
Constitution tacitly assumes that citizens themselves
will somehow possess the opportunities and resources
necessary in order for them to act on their rights. Ill
come back to this assumption in a moment.
Constitutional Structures
Let me say a few words about structures. I am inclined

to believe that three, and possibly four, structural ele-
ments in our constitutional system are not, realistically
speaking, open to change in the foreseeable future.
Federalism. One element is our federal system.
Just as the Framers knew that they could not abolish
the states, constitutional reformers today should prob-
ably assume that the states will remainand in my
judgment they should remainas fundamental units
in a federal government that are endowed with signifi-
cant powers. As has been true for two centuries, the
question of how power is shared between the federal
government and the states will persist as a subject of
endless dispute. But I do not believe that constitu-
tional framers today would or should attempt to dis-
solve the existing states.
Presidentialism. A second limit on constitutional
reform, I believe, is our presidential system. We may
07dahl.141_158 11/27/01 4:41 PM Page 144
be able to alter it a bit around the edges with amend-

ments or changed practices, but the option of a parlia-
mentary system is, I would guess, simply not attractive
to most Americans. We have so deeply invested our-
selves in the mythical aspects of the presidency that
short of some constitutional breakdown, which I nei-
ther foresee nor, certainly, wish for, we wont seriously
consider changing it. For better or worse, we Ameri-
cans are stuck with a presidential system.
Inequality in representation. The other feature
that I fear constitutional reformers would be unable
change is the gross inequality in representation result-
ing from the fixed allocation of two senators to each
state without regard to population. Let me remind you
again that Section 3 of Article I reads: The Senate of
the United States shall be composed of two Senators
from each State for six Years, and each Senator shall
have one Vote. A constitutional amendment to change
this provision faces two formidable obstacles. The first
makes it most unlikely that any constitutional amend-
ment to change the composition of the Senate will be
adopted. The second makes it virtually impossible.
First, under Article V of the Constitution, you may re-
call, amendments can be proposed only with a two-
thirds vote in both houses or by a convention sup-
ported by two-thirds of the state legislatures; and they
can be adopted only after ratification by the legisla-
tures or conventions in three-fourths of the states. I
cant help thinking that at least thirteen of the least
07dahl.141_158 11/27/01 4:41 PM Page 145
populated states would exercise their veto to prevent

the adoption of any amendment that would reduce
their influence in the Senate. But in the highly im-
probable event that such an amendment might some-
how make its way past this formidable obstacle, the
second barrier promises to be totally impregnable.
No State, without its Consent, Article V of the Con-
stitution concludes, shall be deprived of its equal Suf-
frage in the Senate.
In effect, those fifteen words end all possibility of
amending the constitution in order to reduce the un-
equal representation of citizens in the Senate. So we
are destined, it seems, to be saddled indefinitely with
a greater degree of unequal representation in the
upper house than exists in any of the other established
democracies.
If I am correct about these three fixed elements
federalism, presidentialism, and unequal representa-
tion in the Senatethey in turn seem likely to impose
other limits on what todays framers might accomplish.
The electoral college. For example, what can we
do about changing the electoral college? In Chapter 4 I
showed how the inequality of representation in the Sen-
ate reproduces itself, though somewhat weakened, in
the electoral college. It plays out yet again in diminish-
ing the chances that the Constitution can be amended
to replace the electoral college with a system of popu-
lar election. Consequently, I suggested, the relative
desirability of three possible solutions to the problem
07dahl.141_158 11/27/01 4:41 PM Page 146
of unequal representation in the electoral college is in-

versely related to the probability of their enactment.
Consensual, majoritarian, or neither? I have al-
ready expressed some sympathy toward a consensual
system as an alternative to a majoritarian system. The
United States, I suggested, is neither one nor the other.
It is a hybrid that just might have the vices of both and
the virtues of neither.
That our political leaders manage much of the
time to avoid complete deadlock and make the system
workmore or lessis testimony, I think, to their ex-
ceptional political skills, which tend to be vastly under-
valued in the media and among ordinary citizens. I
have also suggested that the necessary wheeling and
dealing, the inevitable behind-the-scenes compro-
mises, and the unavoidable gap between public rheto-
ric and the mutual concessions among insiders, result
in a political system so opaque and so at odds with
general conceptions of public virtue that it weakens
both civic understanding and citizens confidence in
our political institutions.
If we are unlikely to change either our presidential
system or our severely unequal representation in the
Senate, then it will also be very difficult to bring about
some constitutional options that otherwise could be,
and should be, seriously considered. In particular, I
am not entirely sure that we can redesign our present
hybrid so that it facilitates either greater consensus or
stronger majoritarianism.
07dahl.141_158 11/27/01 4:41 PM Page 147
Neither alternative can exist without an appropri-

ate political culture. Yet, a system designed to be con-
sensual would harbor a serious danger if it were not
strongly implanted in a political culture that fosters
agreement. Lacking the appropriate political culture, a
constitutional design intended for a consensual system
would enable a minority to veto any changes from the
status quo that threaten its privileges, as the Southern
states did before the Civil War. Or a regionally privi-
leged minority could extort concessions from the ma-
jority by threatening to use its veto, as the Southern
states did when they compelled the rest of the country
to abandon efforts to protect the civil rights of African-
Americans after the Civil War.
I am inclined to believe that our political culture
unlike, say, that of Sweden, Switzerland, or Holland
would prevent a consensual design from realizing its
potential advantages.
The corresponding fear about a strictly majoritarian
system is that it might fail to provide the leaders of a
majority with adequate incentives to seek greater con-
sensus before they invoke their power as a majority.
Im not raising again here the issue of liberty versus po-
litical equality that I discussed earlier. Even if the lead-
ers of a majority were to maintain the fullest respect
for the democratic rights of minorities, they might see
slight reason to explore options that would achieve a
wider range of agreement and support than they need
to push through a law or policy by majority vote.
07dahl.141_158 11/27/01 4:41 PM Page 148
The ideal solution, it seems to me, would be a po-

litical system that provides strong incentives for politi-
cal leaders to search for the broadest feasible agree-
ment before adopting a law or policy and yet allows
the decision to be made, if need be, by majority vote
always, of course, within the limits set by the need to
preserve fundamental democratic rights. No majority
should have the right, moral or constitutional, to fore-
close decisions by future majorities.
Here again, I fear that this ideal solution may not
be open to us so long as severely unequal representa-
tion in the Senate allows some geographical minorities
to block decisions by representatives elected by a ma-
jority of their fellow citizens. Because of this minority
veto, the search for consensus could easily turn into
what might be harshly described as extortion and black-
mail by a minority of Senators.
Although this barrier to majority rule may be im-
possible to change, rules of the Senate that further
compound the power of privileged minorities might
be more amenable to change, as, in my view, they def-
initely should be. For example, it is doubtful that our
futile and counterproductive policy toward Cuba would
have remained so long in effect were it not for the
ability of a handful of U.S. Senators to extort from Con-
gress and the president the policies they favor, in re-
turn for their support on other issues.
So we arrive at this unhappy conclusion: The Fram-
ers of 1787 appear to have limited todays framers to a
system that is neither consensual nor majoritarian but
07dahl.141_158 11/27/01 4:41 PM Page 149
is a hybrid that possesses the vices of both and the

virtues of neither.
Hidden costs and uncertainties. The difficulty of
transforming a long-established political culture into
one appropriate to a new and different constitutional
structureone more consensual, say, or more majori-
tarianillustrates yet another problem. Major consti-
tutional change involves large hidden costs and a great
uncertainty. The hidden costs arise because of the
need to abandon familiar habits, practices, beliefs, and
understandings that exist among the political elites
and that are embedded in the popular culture as well.
Creating an appropriate political culture may be al-
most as far beyond the capacities of constitutional
framers today as it was for the Framers in 1787. What
is more, even if our knowledge about the likely conse-
quences of alternative constitutional structures is im-
measurably better than that of the Framers, after two
centuries of experience with different democratic con-
stitutions predicting the outcome of major changes
remains fraught with considerable uncertainty.
Powers
Are the constitutional powers of the states, the federal
government, and the three main branches of the fed-
eral government appropriate to our democratic needs
and values today? An attempt to answer this daunting
question would so far exceed my limits here that I shall
simply call attention to its relevance and importance.
07dahl.141_158 11/27/01 4:41 PM Page 150
150
Rights
As with powers, the subject of rights is so vast that I

can do no more than sketch a view that seems to me
useful for appraising fundamental rights in the context
of a democratic countrys constitution.
It is a standard view in jurisprudence that rights
imply duties: in order for a right to be effectively exer-
cised, government officials and others must assume
the duty of protecting the right against persons who
seek to violate it. A less-common assumption, but one
present by implication, is that rights also imply oppor-
tunities: your right to vote is meaningless if you dont
actually have an opportunity to vote. So too with free-
dom of expression. What would a right to free speech
mean to you if you didnt have any opportunity to
speak freely?
Now I want to add a fourth element to our discus-
sion. To rights, duties, and opportunities I would add
resources.2 Suppose, for example, that voting booths
were placed by officials in locations far from the
homes of many citizens and were open for only one
hour in mid-morning: most citizens would lack both
opportunities and resources necessary to cast a vote.
They would be outraged, and so would you and I.
To clarify further what I have in mind when I refer
to resources, Im going to invite you to contemplate a
fictional scenario.
Say that we are all citizens in a New England town
with a traditional town meeting. As usual, a modest
07dahl.141_158 11/27/01 4:41 PM Page 151
proportion of the citizens eligible to attend have actu-

ally turned out, lets say four or five hundred.
After calling the meeting to order, the moderator
announces:
We have established the following rules for this

evenings discussion. After a motion has been properly
made and seconded, in order to ensure free speech
under rules fair to everyone here, each of you who
wishes to do so will be allowed to speak on the motion.
However, to enable as many as possible to speak, no one
will be allowed to speak for more than two minutes.
Perfectly fair so far, you might say. But now our

moderator goes on:
After everyone who wishes to speak for two minutes

has had the floor, each and every one of you is free to
speak further, but under one condition. Each additional
minute will be auctioned off to the highest bidder.
The ensuing uproar from the assembled citizens

would probably drive the moderator and the board of
selectman away from the town halland perhaps out
of town.
Yet isnt this in effect what the Supreme Court
decided in the famous case of Buckley v. Valeo? In
a seven-to-one vote, the court held that the First
Amendmentguarantee of freedom of expression was
impermissibly infringed by the limits placed by the
Federal Election Campaign Act on the amounts that
candidates for federal office or their supporters might
07dahl.141_158 11/27/01 4:41 PM Page 152
spend to promote their election.3 Well, weve had time

to see the appalling consequences.
What went wrong? The justices failed to view cam-
paign expenditures and contributions in the context of
a democratic system that derives its legitimacy from
the principles of political equality that I described ear-
lier. In order to exercise the fundamental rights to
which citizens in a democratic order are entitled
to vote, speak, publish, protest, assemble, organize,
among otherscitizens must also possess the minimal
resources that are necessary in order to take advantage
of the opportunities and to exercise their rights.
The problem of matching resources to democratic
rights admits of no easy answer. Nor can the problem
be solved merely by constitutional prescription. But
surely a constitution is deeply flawed if the highest
court in the land can interpret it to impose an im-
moveable barrier to the achievement of a satisfactory
degree of political equality among its citizens.
A Democratic Role for the Supreme Court
In earlier chapters I alluded to a problem that our

best legal and constitutional scholars have disputed at
length yet remains with us still. In American constitu-
tional circles it sometimes travels under the name of
the counter-majoritarian difficulty.
Ill put the difficulty this way. We cannot simulta-
neously lodge the authority to make laws and policies
07dahl.141_158 11/27/01 4:41 PM Page 153
exclusively in the hands of elected officials who are, at

least in principle, accountable to citizens through elec-
tions and at the same time give the judicial branch the
authority, in effect, to make crucial public policies.
That dilemma presents us with a difficult choice. Many
Americans will resist making it. But if we were ever to
undertake a discussion about the adequacy of our con-
stitution when we assess it against democratic stan-
dards, this problem, which so far been discussed
mainly among legal scholars, would have to be opened
up to public debate and discussion.
There is, I believe, an important place in a demo-
cratic country for a supreme court with the authority
to review the constitutionality of legislative and ad-
ministrative enactments. For one thing, a federal sys-
tem needs a high court empowered to decide if and
when state authorities have exceeded their proper
bounds. But a supreme court should also have the au-
thority to overturn federal laws and administrative de-
crees that seriously impinge on any of the funda-
mental rights that are necessary to the existence of a
democratic political system: rights to express ones
views freely, to assemble, to vote, to form and to par-
ticipate in political organizations, and so on.4
When the court acts within this sphere of funda-
mental democratic rights, the legitimacy of its actions
and its place in the democratic system of government
can hardly be challenged. But the more it moves out-
side this realma vast realm in itselfthe more du-
bious its authority becomes. For then it becomes an
07dahl.141_158 11/27/01 4:41 PM Page 154
unelected legislative body. In the guise of interpreting

the Constitutionor, even more questionable, divin-
ing the obscure and often unknowable intentions of
the Framersthe high court enacts important laws
and policies that are the proper province of elected
officials.
Even within the realm of fundamental democratic
rights the decisions of the court will arouse contro-
versy. And controversy is all the more likely because
our understanding of democratic rights will surely
continue to evolve.
Is Significant Change Possible?
My reflections lead me to a measured pessimism about

the prospects for greater democratization of the Ameri-
can Constitution. Changes described in this chapter that
would be desirable from a democratic point of view
seem to me to have very little chance of coming about
in the indefinite future. Although my judgments about
likelihoods are necessarily subjective, I believe that
most others familiar with American political life would
concur.
The likelihood of reducing the extreme inequality
of representation in the Senate is virtually zero. The
chances of altering our constitutional system to make
it either more clearly consensual or more definitely
majoritarian are also quite low. The likelihood is very
low that the Supreme Court will refrain from legislat-
07dahl.141_158 11/27/01 4:41 PM Page 155
ing public policies, often highly partisan ones, and in-

stead focus its power of judicial review strictly on the
protection of fundamental democratic rights and is-
sues of federalism. The combination of chief executive
and monarchy in the American presidency is not likely
to change. Finally, the probability that democratic
changes in the electoral college will occur appear to be
inversely related to their desirability, with the most de-
sirable having the lowest probability of occurring.
There is at least a modest chance that some states
might require their electoral votes to be allocated in
proportion to the popular votes. But a constitutional
amendment that makes the number of a states elec-
tors proportionate to its population stands little chance
of adoption. And the inequality in representation in
the Senate makes a constitutional amendment provid-
ing for direct popular election of the president virtu-
ally impossible.
Those who regard the Constitution as a sacred
icon will, no doubt, take comfort from this conclusion:
my pessimism is the mirror image of their optimism.
Yet the historic, if fitful, American impulse toward
democracy and political equality has not come to an
end. And so, we face a challenge. Given the present
limits I have described, how might we advance toward
a fuller achievement of democratic processes, rights,
liberties, opportunities, and resources?
Let me suggest two very general strategies.
First, it is timelong-past timeto invigorate
and greatly widen the critical examination of the Con-
07dahl.141_158 11/27/01 4:41 PM Page 156
stitution and its shortcomings. Public discussion that

penetrates beyond the Constitution as a national icon
is virtually nonexistent. Even when in-depth analysis
does occurmainly among constitutional scholars in
schools of law and departments of political science and
historythe Constitution as a whole is rarely tested
against democratic standards or against the perform-
ance of constitutional systems in other advanced dem-
ocratic countries.
I can envision the possibilityhere a degree of
optimism breaks throughof a gradually expanding
discussion that begins in scholarly circles, moves out-
ward to the media and intellectuals more generally,
and after some years begins to engage a wider public.
I cannot say what the outcome might be. But surely it
would heighten understanding of the relevance of
democratic ideas to the constitution of a democratic
country, and specifically it would heighten understand-
ing of the shortcomings of the existing constitution
viewed from that perspective and of the possibilities of
change.
Meanwhile, however, we need a second strategy,
one designed to achieve greater political equality within
the limits of the present American Constitution. A major
objective of such a strategy would be to reduce the
vast inequalities in the existing distribution of political
resources. The characteristics of the Constitution that
I have described in this book will, of course, stand as
obstacles to the success of such a strategy, for they arm
those who possess the greatest resources with strong
07dahl.141_158 11/27/01 4:41 PM Page 157
defensesopportunities to veto changesagainst all

efforts to reduce their privileged positions.
I cannot foresee how successful either strategy
may prove to be. But the belief of most Americans
that a democratic government, warts and all, is better
in the end than any feasible alternative to it is justified
on fundamental principles of human equality that can-
not be tightly bounded. Our understanding of the im-
plications of those principles will therefore continue
to evolve indefinitely. So, too, will the implications of
those principles for our democratic political system,
and its Constitution, under which we Americans freely
choose to live.
07dahl.141_158 11/27/01 4:41 PM Page 158
08dahlAppA.159_162 11/27/01 4:41 PM Page 159
appendix a
On the Terms
Democracy and Republic
T
HE VIEW THAT THE FRAMERS INTENDED TO
create a republic, not a democracy, probably
has its origins in comments by Madison in
Federalist No. 10. Although there as elsewhere he also
used the expression popular government as a kind of
generic term, he distinguished further between a
pure democracy, by which I mean a society consisting
of a small number of persons, who assemble and ad-
minister the government in person, and a republic,
by which I mean a government in which the scheme of
representation takes place. The two great points of
difference between a democracy and a republic are:
first, the delegation of the government, in the latter,
to a small number of citizens elected by the rest;
secondly, the greater number of citizens, and greater
08dahlAppA.159_162 11/27/01 4:41 PM Page 160
160 o n t h e t e r m s d e m o c r ac y a n d r e p u b l i c
sphere of the country, over which the latter may be ex-

tended.1
Here Madison was making the common distinction
that political scientists and others would later differen-
tiate as direct democracy and representative de-
mocracy. For it was as evident to the Framers as it is
to us that given the size of a nation composed of the
thirteen existing states, with more to come, the
people could not possibly assemble directly to enact
laws, as they did at the time in New England town
meetings and had done two millennia earlier in Greece,
where the term democracy was invented. It was per-
fectly obvious to the Framers, then, that in such a
large country, a republican government would have to
be a representative government, where national laws
would be enacted by a representative legislative body
consisting of members chosen directly or indirectly by
the people.
Madison was probably also influenced by a long
tradition of republicanism that in both theory and
practice leaned somewhat more toward aristocracy, lim-
ited suffrage, concern for property rights, and fear of
the populace than toward a broadly based popular gov-
ernment more dependent on the will of the people.
It is also true, however, that during the eighteenth
century the terms democracy and republic were
used rather interchangeably in both common and
philosophical usage.2 Madison, in fact, was well aware
of the difficulty of defining republic. In Federalist
No. 39, he posed the question What, then, are the
08dahlAppA.159_162 11/27/01 4:41 PM Page 161
o n t h e t e r m s d e m o c r ac y a n d r e p u b l i c 161
distinctive characters (sic) of the republican form? In

response he pointed to the enormous range of mean-
ings given to the word republic. Were an answer to
this question to be sought . . . in the application of the
term by political writers, to the constitutions of differ-
ent states, no satisfactory one could ever be found.
Holland, in which no particle of the supreme authority
is derived from the people, has passed almost univer-
sally under the denomination of a republic. The same
title has been bestowed on Venice, where absolute
power over the great body of the people is exercised,
in the most absolute manner, by a small body of here-
ditary nobles.
In view of this ambiguity, Madison proposed that
we may define a republic to be . . . a government
which derives all its powers directly or indirectly from
the great body of the people, and is administered by
persons holding their offices during pleasure, or for a
limited period, or during good behavior.3 By defining
a republic as a government which derives all its pow-
ers directly or indirectly from the great body of the
people, Madison now seems to be contradicting the
distinction he had drawn earlier in Federalist No. 10.
We might read his struggle with definitions as a fur-
ther illustration of the prevailing confusion over the
two terms.
If further evidence were needed of the ambiguity
of terminology, we could turn to a highly influential
writer whose work was well known to Madison and
many of his contemporaries. In The Spirit of the Laws
08dahlAppA.159_162 11/27/01 4:41 PM Page 162
162 o n t h e t e r m s d e m o c r ac y a n d r e p u b l i c
(1748) Montesquieu had distinguished three kinds of

governments: republican, monarchic, and despotic.
Republican governments were of two kinds: When, in
a republic, the people as a body have the sovereign
power, it is a Democracy. When the sovereign power is
in the hands of a part of the people, it is called an Aris-
tocracy.4 But Montesquieu also insisted that It is in
the nature of a republic that it has only a small terri-
tory: without that it could scarcely exist.5
Although the Framers differed among themselves
as to how democratic they wanted their republic to be,6
for obvious reasons they were of one mind about the
need for a representative government. But as events
soon showed, they could not fully determine just how
democratic that representative government would be-
comeunder the leadership of, among others, James
Madison.
09dahlAppB.163_172 11/27/01 4:41 PM Page 163
Appendix B
Tables and Figures

09dahlAppB.163_172 11/27/01 4:41 PM Page 164
164 ta b l e a n d f i g u r e s
TABLE 1. Countries Steadily Democratic

Since at Least 1950
1. Austria
2. Australia
3. Belgium
4. Canada
5. Costa Rica
6. Denmark
7. Finland
8. France
9. Germany
10. Iceland
11. Ireland
12. Israel
13. Italy
14. Japan
15. Luxembourg
16. Netherlands
17. New Zealand
18. Norway
19. Sweden
20. Switzerland
21. United Kingdom
22. United States
Note: The countries are broken down as follows: European: Austria, Belgium,
Denmark, Finland, France, Germany, Iceland, Ireland, Italy, Luxembourg,
Netherlands, Norway, Sweden, Switzerland, United Kingdom (15); English-
speaking: Australia, New Zealand, United Kingdom, United States (4); Latin
American: Costa Rica; other: Israel, Japan. Although India gained independ-
ence in 1947, adopted a democratic constitution, and has, except for one inter-
val, maintained its democratic institutions in the face of extraordinary chal-
lenges of poverty and diversity, I have omitted it from the list for two reasons.
First, continuity was interrupted from 1975 to 1977 when the prime minister,
Indira Gandhi, staged a coup dtat, declared a state of emergency, suspended
civil rights, and imprisoned thousands of opponents. Second, because India is
one of the poorest countries in the world, comparisons with the wealthy coun-
tries in Table 5 would make little sense.
09dahlAppB.163_172 11/27/01 4:41 PM Page 165
ta b l e s a n d f i g u r e s 165
TABLE 2. How the U.S. Constitutional-Political

System Compares with the
Other 21 Older Democracies
Among the other 21 countries,
those that are similar
in this respect
Characteristics ofxxx
the U.S. systemxxx (n) Countries
Federalism
1. Strong federalism 5 Australia, Canada,

Germany, Switzerland,
Belgium (since 1993)
2. Strongly bicameral 3 All federal: Australia,
legislature Germany, Switzerland
3. Significant unequal 4 All federal: Australia,
representation in the Canada, Germany,
upper house Switzerland
Non-Federal Features
4. Strong judicial review of 2 Canada, Germany

national legislation
5. Electoral system: plurality 2 Britain, Canada
elections in single member
districts (FPTP)
6. Strong two-party system: 3 Australia2, New
third parties weak1 Zealand,3 Costa Rica
7. Presidential system: single None
popularly elected chief
executive with important
constitutional powers
1 Untilthe 1997 election, held under the new PR system, when the two major
parties received only 61% of the vote. The rest was split almost entirely among
three minor parties.
2 Where the total vote cast for third party candidates is ordinarily under 10% of
the total.
3 Until2000, counting the Liberal National (earlier Liberal-Country) coalition
as one party.
09dahlAppB.163_172 11/27/01 4:41 PM Page 166
TABLE 3. The Electoral Systems of the Advanced

Democracies
Plurality in
SingleMember
PR: PR: Districts
List System Variants (FPTP) Variant
1. Austria 1. Australia 1. Canada France

(AV)
2. Belgium 2. Germany 2. United
(MMP) Kingdom (2 rounds)
3. Costa Rica 3. Ireland 3. United
(STV) States
4. Denmark 4. Italy
(MMP)
5. Finland 5. Japan
(semi PR)
6. Iceland 6. New
Zealand
(MMP)
7. Israel
8. Luxembourg
9. Netherlands
10. Norway
11. Sweden
12. Switzerland
Source: Andrew Reynolds and Ben Reilly, The International IDEA Handbook of
Electoral System Design (1997)
Note: AV = Alternative Vote; MMP = Multimember Proportional; STV = Single
Transferable Vote.
09dahlAppB.163_172 11/27/01 4:41 PM Page 167
TABLE 4. Proportionality vs. Majoritarianism:

20 Democratic Countries
Predominantly Predominantly
Proportional Mixed Majoritarian
Austria Ireland Australia
Belgium Japan Canada
Denmark Spain France
Finland USA Greece
Germany New Zealand (to 1993)
Netherlands United Kingdom
New Zealand (1993)
Norway
Sweden
Switzerland
Source: Powell, Elections as Instruments of Democracy (2000), p. 41. I have
changed his table to show the adoption of proportionality in New Zealand in
1993.
Note: The list is slightly different from Table 1. It includes Spain and Greece,
democratized since 1950, and it does not include Costa Rica, Iceland, Israel,
or Luxembourg.
09dahlAppB.163_172 11/27/01 4:41 PM Page 168
TABLE 5. Measures of U.S. Performance among

Democratic Countries
Percentage
of Countries
Performing
U.S. No. of Better than
Variable Rank Countries U.S.1
U.S. in BEST third
Economic 5th best (tied) of 18 24

growth,
198095
U.S. in MIDDLE third
Womens cabinet 8th most of 22 33

representa-
tion, 199395
Popular support 10th most of 22 43
for executive,
194596 2
Budget deficits, 8th highest (tied) 16 47
197095 of
Unemployment, 8th highest (tied) 18 59
197195 3 of
Family policy, 12th best of 18 65
197682
Inflation (CPI), 12th highest of 18 65
197095
09dahlAppB.163_172 11/27/01 4:41 PM Page 169
U.S. in WORST third
Womens 18th lowest 22 81

parliamentary (tied) of
representation,
197195
Rich-poor ratio, 4th greatest of 18 82
198193
Energy 19th lowest of 22 86
efficiency,
199094
Welfare state 17th lowest of 18 94
index, 1980
Social 17th lowest of 18 94
expenditures,
1992
Voter turnout, 21st lowest of 22 95
197196
Incarceration 1st highest of 18 100
rate, 199295
Foreign aid, 19th lowest of 19 100
199295
Source: I am indebted to Arend Lijphart for his permission to use the data set
he developed for Patterns of Democracy (New Haven: Yale University Press,
1999) where the variables and their sources are described. I also owe thanks to
Jennifer Smith for her contributions to the construction of the table.
1 Thepercentages are those of countries scoring better than the United States,
and not tied. Variables for which the higher the ranking, the worse a countrys
performance are in italics.
2 The average percentage of the voters who gave their votes to the party or
parties that formed the cabinet, or in presidential systems, the percentage of
voters who voted for the winning presidential candidate, weighted by the time
each cabinet or president was in office. Lijphart 1999, p. 290.
3
Unstandardized.
09dahlAppB.163_172
11/27/01
4:41 PM
Page 170
FIGURE 1.
Presidential Elections Won with Less Than 50% of the Popular Vote
09dahlAppB.163_172
11/27/01
4:41 PM
Page 171
Residents per elector (thousands)

States
FIGURE 2.
Unequal Representation in the Electoral College
09dahlAppB.163_172 11/27/01 4:41 PM Page 172
10dahlNotes.173_190 11/27/01 4:42 PM Page 173
Notes
CHAPTER 1. Introduction
1. Although in three statesDelaware, New Jersey, and
Georgiathe vote was unanimous, in the rest it was divided,
sometimes closely after sharp debate. For example, in Massa-
chusetts the delegates split 187 to 168; in New Hampshire, 57
to 46; and in Virginia, the state from which several of the prin-
cipal authors of the Constitution came, the supporters of the
Constitution won by a single vote: 80 to 79.
2. In the ten states where the Convention vote was not
unanimous, a total of 1540 delegates voted on the Constitu-
tion, 964 for, and 576 against.
CHAPTER 2. What the Framers Couldnt Know

1. I quote here from the Journal of the Convention, in
Max Farrand, ed., The Records of the Federal Convention of
1787, 3 vols. (New Haven: Yale University Press, 1966), 1: 1.
The three volumes edited by Farrand were reprinted in 1987
with a fourth volume, a Supplement edited by James H. Hutson
174 n ot e s to pag e s 7 1 2
(New Haven: Yale University Press, 1987). My references to

the records of the Convention are drawn from these four vol-
umes and are cited hereafter as Records.
2. William Miller, The Business of May Next: James Madi-
son and the Founding (Charlottesville: University Press of Vir-
ginia, 1992), 41ff. Lance Banning writes that Madison had
come to Philadelphia the best prepared of all who gathered for
the Federal Convention. . . . He first prepared elaborate re-
search notes on the histories and structures of other ancient
and modern confederacies. 115.
3. Records, 1: 425.
4. Madison took extensive notes during the Convention,
which he later edited and also collated with the very brief
Journal of the Convention, published in 1819. His notes were
published posthumously in 1840. These form a part of the se-
ries described in note 1, above. I have maintained Madisons
punctuation and spelling.
5. His speech on June 18 as reported by Madison is in
Records, 1: 282ff. Hamilton said that he had no scruple in de-
claring . . . that the British Govt. was the best in the world. . . .
As to the Executive, it seemed to be admitted that no good one
could be established on Republican principle. . . . The English
model was the only good one on this subject. 288, 299.
6. In order to serve as a check on the popular chamber of
the national legislature, the second chamber must have great
personal property, it must have the aristocratic spirit; it must
love to lord it thro pride. . . . The aristocratic body, should be
as independent & firm as the democratic. . . . To make it inde-
pendent, it should be for life. Ibid., 1: 512. With his usual ad-
miration for the British system, in his first speech to the Con-
vention, Hamilton opined that the House of Lords is a most
noble institution. Having nothing to hope for by a change, and
sufficient interest by means of their property, in being faithful
to the National interest, they form a permanent barrier agst.
every pernicious innovation. 1: 288 (June 18).
7. A few delegates favored the idea of somehow getting
rid of the states and consolidating power in the national gov-
ernment. George Read of Delaware disliked the idea of guar-
antying territory. It abetted the idea of distinct States wch.
n ot e s to pag e s 1 4 19 175
would be a perpetual source of discord. There can be [no]

cure for this evil but in doing away with States altogether and
uniting them all into [one] great Society. Records, 1: 202
(June 11). He had made a similar proposal a few days earlier,
on June 6. 13637. In his maiden address cited above, Hamil-
ton proposed that the Governor or president of each state
shall be appointed by the General Government and shall have
a negative upon the laws about to be passed in the State of
which he is Governor or President 293.
8. Records, 1: 466.
9. Records, 1: 49293. I have put these remarks in the
first person. In Madisons published notes they are recorded in
the third person.
10. The advocates of equal representation in the House
were defeated on June 29 by a vote of six states for and four
states against, with one state (Maryland) divided. Their pro-
posal for equality in the Senate was stalemated by a tie vote on
July 2 (five to five, with Georgia divided) and finally carried on
July 7 with six voting yes, three voting no, and two states
(Massachusetts and Georgia) divided. Records, 1: 549.
11. Article I, Section 9. For an excellent account of the
only full public debate over the slavery issue, see Joseph J.
Ellis, Founding Brothers: The Revolutionary Generation (New
York: Alfred A. Knopf, 2000), 81119. The debate took place
in the House of Representatives in March 1790 in response to
petitions from Quakers in New York and Philadelphia calling
for the federal government to put an immediate end to the Af-
rican slave trade. (81).
12. Article IV, Section 2.
13. Article I, Sections 2, 3.
14. For a magisterial study of the evolution of American cit-
izenship, see Rogers Smith, Civic Ideals: Conflicting Visions of
Citizenship in U.S. History (New Haven: Yale University Press,
1997). On the constitutions omission of citizenship for women,
Native Americans, and African Americans, see 13034.
15. Article I, Section 3.
16. By the same electorate as that for the most numerous
branch of the state legislature. (Article I, Section 2).
17. Records, 2: 83.
18. For evidence that the Supreme Court sometimes plays

such a role, see my Decision-Making in a Democracy: The
Supreme Court as a National Policy-Maker, Journal of Public
Law 6, no. 2, 27995.
19. It is only fair to point out that given the political oppo-
sition to any increase in federal powers, the Framers may well
have gone as far as they could go. Their major opponents, the
Anti-Federalists, who saw the constitution as a threat to popu-
lar government at the state level, objected that the powers of
Congress to regulate interstate commerce were excessive.
Richard L. Perry, ed., The Sources of Our Liberties: Documen-
tary Origins of Individual Liberties in the United States Con-
stitution and Bill of Rights (New York: American Bar Associa-
tion, 1959), 240.
20. For an account of the development in the American
colonies of ideas and practices concerning popular govern-
ment, see Edmund S. Morgan, Inventing the People: The Rise
of Popular Sovereignty in England and America (New York:
W. W. Norton, 1988), esp. Chs. 8 and 9, 174 233.
21. Gordon S. Wood, The Radicalism of the American Rev-
olution (New York: Alfred A. Knopf, 1992), 230.
22. The numbers are a matter of uncertainty. In some
colonies the suffrage may have grown more restricted during
the colonial period. What is also unclear is just how many
people could and did vote. This issue is a source of controversy
among historians, some of whom conclude that colonial Amer-
ica was a land of middle class democracy in which 80 or 80
percent of all adult white males were enfranchised, while
others depict a far more oligarchic and exclusive political
order. In fact, enfranchisement varied greatly by location.
There certainly were communities, particularly newly settled
communities where land was inexpensive, in which 70 or 80
percent of all white men were enfranchised. Yet there were
also locales . . . where the percentages were far lower, closer to
40 or 50 percent. Levels of enfranchisement seem to have
been higher in New England and in the South (especially Vir-
ginia and the Carolinas) than they were in the mid-Atlantic
colonies (especially New York, Pennsylvania, and Maryland;
not surprisingly, they also tended to be higher in newer settle-
n ot e s to pag e s 2 2 2 6 177
ments than in more developed areas. On the whole, the fran-

chise was far more widespread than it was in England; yet as
the revolution approached, the rate of property ownership was
falling, and the proportion of adult white males who were eli-
gible to vote was probably less than 60 percent. Alexander
Keyssar, The Right to Vote: The Contested History of Democ-
racy in the United States (New York: Basic Books, 2000), 7.
23. On the constitutions exclusion of women, Native
Americans, and African Americans, see Keyssar, 13034.
24. Democracy in America, trans. Henry Reeve (New York:
Schocken, 1961), 1: lxvii.
25. Although Jefferson and his followers often referred to
their political group as Republican, their party seems to have
taken the name Democratic Republican as early as 1796, and
it retained the name through the election of 1828. In 1820,
Monroe ran as a Democratic Republican and Adams as an In-
dependent Democratic Republican. In 1824 all four candi-
datesAdams, Jackson, Crawford, and Clayran as factions
of the Democratic-Republican party. In 1828, Jackson ran as a
Democratic Republican, Adams as a National-Republican. In
1832, Jackson ran as the candidate of the Democratic Party
and Clay as a National-Republican. Congressional Quarterly,
Presidential Elections Since 1789, 2nd ed. (Washington, D.C.:
Congressional Quarterly, 1979), 1927.
26. Joyce Appleby, Inheriting the Revolution: The First
Generation of Americans (Cambridge, Mass.: Harvard Univer-
sity Press, 2000), 65.
27. When land offices opened on the frontier, land sales
soared. In 1800 some 67,000 acres passed into private hands;
497,939 acres did so in 1801. By 1815 annual sales hit one and
a half million dollars, more than doubling four years later.
Appleby, Inheriting the Revolution, 64. As Gordon Wood re-
marks in his review, Tens of thousands of ordinary folk pulled
up stakes in the East and moved westward, occupying more
territory in a single generation than had been occupied in the
150 years of colonial history. Early American Get-up-and-
Go, New York Review, June 29, 2000, 50.
28. For a rather unsympathetic portrait, see Stanley Elkins
and Eric McKitrick, The Age of Federalism: The Early Ameri-
178 n ot e s to pag e s 2 7 4 3
can Republicanism, 17881900 (New York: Oxford University

Press, 1993), 706ff.
29. Dubious as one might beand I am profoundly doubt-
fulabout the contemporary relevance of the Second Amend-
ment securing the right of the people to keep and bear Arms,
I have no doubt that contemporaries saw it as important to
maintaining their liberty from a potentially dangerous central
government.
30. Democracy and the Amendments to the Constitution
(Lexington, Mass.: Lexington Books, 1978), 166.
31. The Federalist (New York: Modern Library, n.d.), 53ff.
32. The often murky maneuvers preceding the compromise
are described in Bernard A. Weisberger, America Afire: Jeffer-
son, Adams, and the Revolutionary Election of 1800 (New York:
William Morrow, 2000), 258 77.
33. Elkins and McKitrick, 263 et seq.
34. Quoted, ibid., 267.
35. Records, 3: 45255. Italics added. Spelling and punctu-
ation as in the original.
36. The Forging of American Federalism: Selected Writings
of James Madison, Saul K. Padover, ed. (New York: Harper
Torchbooks, 1953), letter to Thomas Ritchie, 1825, 46.
37. Marvin Meyers, ed., The Mind of the Founder: Sources
of the Political Thought of James Madison (New York: Bobbs-
Merrill, 1973), 520.
38. Ibid., 523, 525, 530.
CHAPTER 3. The Constitution as a Model:

An American Illusion
1. In a 1997 survey, 34% strongly agreed and 33% some-
what agreed with the statement The U.S. Constitution is used
as a model by many countries. Only 18% somewhat or
strongly disagreed. (Nationwide telephone survey of 1,000
adult U.S. Citizens conducted for the National Constitution
Center, September, 1997.) To the statement I am proud of
the U.S. Constitution, 71% strongly agreed and 18% some-
what agreed. In 1999, 85% said the Constitution was a major
reason for Americas success in the twentieth century. (Survey
of 1,546 adults for the Pew Research Center by the Princeton

Survey Research Associates.)
2. Although India gained independence in 1947, adopted a
democratic constitution, and has, except for one interval, main-
tained its democratic institutions in the face of extraordinary
challenges of poverty and diversity, I have omitted it from the
list for two reasons. First, continuity was interrupted from 1975
to 1977 when the prime minister, Indira Gandhi, staged a coup
detat, declared a state of emergency, suspended civil rights,
and imprisoned thousands of opponents. Second, because
India is one of the poorest countries in the world, comparisons
with the wealthy democratic countries would make little sense.
3. For a summary of the constitutional differences among
twenty-two older democracies, see Appendix B, Table 2.
4. Plus six half-cantons.
5. Robert Hazell and David Sinclair, The British Consti-
tution: Labours Constitutional Revolution, Annual Review of
Political Science, 3 (Palo Alto: Annual Reviews, 2000), 379
400, 393.
6. For example, in the Prussian elections of 1858, 4.8% of
the inhabitants were entitled to one-third of the seats, 13.4%
to another third, and 81.8% to the remaining third. Thus
members of the wealthiest third in effect possessed 17 times as
many votes as members of the bottom third. Bernhard Vogel
and Rainer-Olaf Schultze, Deutschland, in Die Wahl Der
Parlamente, Dolf Sternberger and Bernard Vogel, eds. (Berlin:
Walter De Gruyter, 1969), 189 411, Tabelle A 4, p. 348.
7. Lest you think me biased against Nevada, the Rocky
Mountain states, or small states in general: I have the greatest
affection for Alaska, where I grew up in the days when it was
still a territory, and for the Rocky Mountain states, where I like
to spend some time every summer. And at just over 3 million
people, Connecticut gives me a wholly undeserved voting ad-
vantage of nine to one over my sons in California.
8. Alfred Stepan, Toward a New Comparative Analysis of
Democracy and Federalism: Demos Constraining and Demos
Enabling Federations, paper for the meeting of the Interna-
tional Political Science Association, Seoul, Aug. 1722, 1997.
9. For a comprehensive description, analysis, and critique
of unequal representation in the Senate, see Francis E. Lee

and Bruce I. Oppenheimer, Sizing Up the Senate: The Un-
equal Consequences of Unequal Representation (Chicago: Uni-
versity Chicago Press, 1999).
10. Stepan, supra n. 8.
11. More precisely, a governmental unit of a State de-
fined as a territorial system with a government that success-
fully upholds a claim to the exclusive regulation of the legiti-
mate use of physical force in enforcing its rules within a given
territorial area.
12. For Mason, see Records, 1: 483; for Madison, see
447 48.
13. Barry R. Weingast, Political Stability and Civil War: In-
stitutions, Commitment, and American Democracy, in Robert
H. Bates, Avner Greif, Margaret Levi, Jean-Laurent Rosenthal,
and Barry R. Weingast, Analytic Narratives (Princeton: Prince-
ton University Press, 1988), 14893, 166, and Table 4.3, 168.
14. Article II, Section 4 provides: The times, places, and
manner of holding elections for Senators and Representatives,
shall be prescribed in each State by the Legislature thereof;
but the Congress may at any time by Law make or alter such
regulations, except as to the place of choosing Senators. Ar-
ticle II, Section 1 provides: Each state shall appoint, in such
manner as the legislature therefore may direct, a number of
Electors.
15. Robert Hazell and David Sinclair, The British Consti-
tution: Labours Constitutional Revolution, in Nelson W.
Polsby, ed., Annual Review of Political Science, vol. 3 (Palo
Alto: Annual Reviews, 2000), 379 400, 38285, 391.
16. Lani Guanier, No Two Seats: The Elusive Quest for
Political Equality, Virginia Law Review 77 (1991).
17. Maurice Duverger, Political Parties: Their Organiza-
tion and Activity in the Modern State (New York: John Wiley,
1954), 217.
18. In an appraisal of Duvergers propositions in 1958, John
Grumm observed that it may be more accurate to conclude
that proportional representation is a result rather than a cause
of the party system in a given country. Theories of Electoral
Systems, Midwest Journal of Political Science 2 (1958): 357

76, 375.
19. Arend Lijphart, Patterns of Democracy, Government
Forms and Performance in Thirty-Six Countries (New Haven:
Yale University Press, 1999) uses ten variables to distinguish
majoritarian from consensus democracies. Table 14.1, p. 245.
G. Bingham Powell, Elections as Instruments of Democracy,
Majoritarian and Proportional Visions (New Haven: Yale Uni-
versity Press, 2000) uses the terms of his title: majoritarian
and proportional. See pp. 20ff and the classification of
twenty democratic countries on p. 41.
20. Records, 1: 288, 299.
21. Madisons notes, published in 1840 after his death. See
above, note 4.
22. The only delegate recorded by Madison as speaking fa-
vorably about the British monarchy was Hamilton. See note 3
above. Ironically, the Federalist Papers defending the provi-
sions of the Constitution on the executiveNos. 6777
were by Hamilton.
23. In what follows I have drawn freely from my Pluralist
Democracy in the United States (Chicago: Rand McNally,
1967), 85ff.
24. A convenient source for some of the relevant discus-
sions is Richard J. Ellis, ed., Founding the American Presi-
dency (Lanham, Md.: Rowman and Littlefield, 1999), Ch. 3,
6396.
25. For a critical view, see my The Myth of the Presiden-
tial Mandate, Political Science Quarterly 105, no. 3 (Fall 1990):
35572.
CHAPTER 4. Electing the President

1. For an excellent and much fuller account of the elec-
toral college than brevity permits me here, see Lawrence Lon-
gley and Neal R. Peirce, The Electoral College Primer (New
Haven: Yale University Press, 1999). See also Robert M. Hard-
away, The Electoral College and the Constitution: The Case for
Preserving Federalism (Westport, CT: Praeger, 1994).
2. Records, 2: 501.
3. James Wilsons Final Summation and Rebuttal, De-
cember 11, 1787, in The Debate on the Constitution, Bernard
Bailyn, ed. , 2 vols, Vol. 1, 849.
4. Records, 2: 497.
5. Records, 2: 522.
6. The Federalist (New York: Modern Library, n.d.), 441.
Although the other members of the New York delegation had
withdrawn from the Convention in June, Hamilton stayed on,
though he rarely intervened in the debates and seems to have
little influence on the outcome.
7. Ibid., 443.
8. My interpretation of anxious.
9. Records, 2: 500. A fear of cabal had been frequently
expressed throughout the earlier discussions.
10. Article II provided that The Person having the great-
est Number of [electoral] Votes shall be the President, and
after the choice of the President, the Person having the great-
est Number of Votes of the Electors shall be the Vice Presi-
dent.
11. How could the framers have made such an elementary
and colossal blunder? No less puzzling, why did the opponents
of the Constitution, who were generally so eager to seize on
the documents weaknesses, never once expose this weakness
in the mechanics of the electoral college? The short answer is
that neither supporters nor opponents of the Constitution an-
ticipated the formation of organized national political parties.
Richard J. Ellis, ed., Founding the American Presidency (Lan-
ham, Md.: Rowman and Littlefield, 1999), 114.
12. It also provided that if no candidate received a majority
of electoral votes, the House would choose among the top
three (not five as in the original article).
13. The classic account is by C. Vann Woodward, Reunion
and Reaction: The Compromise of 1877 and the End of Recon-
struction (Boston: Little, Brown: 1951).
14. Congressional Quarterly, Presidential Elections Since
1789, 2nd ed. (Washington, D.C.: Congressional Quarterly,
1979), 11.
15. Ellis, 118.
n ot e s to pag e s 86 87 183
16. The figures may change slightly when the House seats
are reapportioned to conform to the 2000 census figures.
17. Ellis, 118.
18. Ibid., 119.
19. For a more extended examination of possibilities for re-
form, see Lawrence D. Longley and Alan G. Braun, The Poli-
tics of Electoral College Reform, foreword by U.S. Senator
Birch Bayh, 2nd ed. (New Haven, Yale University Press, 1975).
20. The need for a second election could be avoided by
means of an electoral system various called Instant Run-off,
Alternative Vote (AV), or Preferential Voting. [U]nder AV
electors rank the candidates in the order of their choice by
marking 1 for their favorite candidate, 2 for their second-
choice, 3 for their third choice, and so on. . . . [A] candidate
who has won an absolute majority of votes (50% plus one) is
immediately elected. However, if no candidate has an absolute
majority, under AV the candidate with the lowest number of
preferences is eliminated from the count, and his or her . . .
second preferences . . . are then assigned to the remaining
candidates in the order as marked on the ballot. This process is
repeated until one candidate has an absolute majority, and is
declared duly elected. Andrew Reynolds and Ben Reilly, The
International IDEA Handbook of Electoral System Design
(Stockholm: International IDEA, 1997), 38. This system is
used in Australia for electing members of parliament in single-
member districts. Since 1922 an analogous systemthe Single
Transferable Vote (STV)has been used in the Republic of
Ireland for the election of members of parliament. However,
unlike the presidential elections in the U.S. and parliamentary
elections in Australia, in Ireland the members of parliament
are elected in districts returning three, four, or five members.
The STV system produces a high degree of proportionality be-
tween the size of a partys votes and the number of its M.P.s.
(Ibid., 85ff ).
21. A Gallup survey in 1968 asked respondents, Would
you approve or disapprove of an amendment to the Constitu-
tion which would do away with the Electoral College and base
the election of the President on the total vote throughout the
nation? Eighty-one per cent approved, 12 percent disap-
proved, and seven percent had no opinion. Longley and Braun,

154. In a 1992 survey, respondents were presented with the
statement that If Perot runs, there is a chance that no presi-
dential candidate will get enough electoral votes to win. If that
happens, the Constitution gives the House of Representatives
the power to decide who will be the next President. Do you
think that is a fair way to choose the President, or should the
Constitution be changed. Only 31% felt it was a fair way,
while 61% answered that the Constitution should be changed.
CBS News-New York Times national telephone survey of
1,346 adults, July, 1992. Another survey in 1992 indicated that
Americans were far from agreed on how their Representative
in Congress should vote if none of the candidates received an
Electoral College majority. The respondents split their choices
among the options they were offered. Their Representative
should vote for the candidate:
With the most popular votes nationwide 29%
Who carries your state 16%
Who carries your congressional district 14%
Who would make the best president 33%
Dont know, no answer 7%
Gallup national telephone survey of 1,006 adults, August 1995.

22. Longley and Braun, 154.
23. Shlomo Slonin, The Electoral College at Philadelphia:
The Evolution of an Ad Hoc Congress for the Selection of a
President, Journal of American History 73 (June 1986): 35
58, cited in Ellis, 110.
24. Longley and Braun, 169.
CHAPTER 5. How Well Does the Constitutional

System Perform?
1. I have drawn feely here from my Thinking About De-
mocratic Constitutions: Conclusions from Democratic Ex-
perience, in Ian Shapiro and Russell Hardin, eds., NOMOS
XXXVIII, Political Order (New York: New York University
Press, 1996).
n ot e s to pag e s 96 108 185
2. The survival of basic democratic institutions should not

be confused with the stability or turnover of parliamentary
governments. Parliamentary systems display a huge variation
in the durability of cabinet coalitionsgovernmentsrang-
ing among our twenty two democratic countries from the rela-
tively high turnover of cabinets in Italy to their much higher
stability in Norway and Britain. Even in Italy, however, an out-
going cabinet may return with essentially the same members
and parties represented in the government.
3. Mathew Soberg Shugart and John M. Carey, Presidents
and Assemblies: Constitutional Design and Electoral Dynamics
(Cambridge: Cambridge University Press, 1992), 41.
4. Support for this view and also more skeptical views may
be found in Juan Linz and Arturo Valenzuela, eds., The Failure
of Presidential Democracy: Comparative Perspectives, vol. 1
(Baltimore: Johns Hopkins University Press, 1994).
5. Shugart and Carey, 42.
6. Aili Piano and Arch Puddington, The 2000 Freedom
House Survey, Journal of Democracy 12 (January 2001):
8792.
7. Freedom House, Press Freedom Survey: Press Free-
dom World Wide (January 1, 1999).
8. Here I follow the usage in G. Bingham Powell, Jr.,
Elections as Instruments of Democracy (New Haven: Yale Uni-
versity Press, 2000).
9. An excellent analysis supported by extensive analysis
and data is provided by Powell, Elections.
10. See ibid., Chs. 4 and 6.
11. Ibid., 129, 130, 197.
12. Arend Lijphart, Democracies, Patterns of Majoritarian
and Consensus Government in Twenty-One Countries (New
Haven: Yale University Press, 1984), and Patterns of Democ-
racy, Government Forms and Performance in Thirty-Six Coun-
tries (New Haven: Yale University Press, 1999).
13. Hans Daalder, The Netherlands: Opposition in a Seg-
mented Society, in Robert A. Dahl, ed., Political Oppositions
in Western Democracies (New Haven: Yale University Press,
1966), 188236; Arend Lijphart, The Politics of Accommoda-
186 n ot e s to pag e s 10 9 1 18
tion: Pluralism and Democracy in the Netherlands, 2nd rev.

ed. (Berkeley: University of California Press, 1975), 104ff.
14. Leif Leiwin, Majoritarian and Consensus Democracy:
The Swedish Experience, Scandinavian Political Studies 21,
no. 3 (1988): 195206.
15. Christopher J. Anderson and Christine A. Guillory,
Political Institutions and Satisfaction with Democracy: A
Cross-National Analysis of Consensus and Majoritarian Sys-
tems, American Political Science Review 91, (March 199):
6681.
16. Ibid., fig. 4, p. 77.
17. Ibid., 78.
18. David R. Mayhew, Divided We Govern: Party Control,
Lawmaking, and Investigations, 19461990 (New Haven: Yale
University Press, 1991), 1.
19. Ibid., 76.
20. John J. Coleman, Unified Government, Divided Gov-
ernment, and Party Responsiveness, American Political Sci-
ence Review 93 (December 1999): 82136.
21. Jeffrey K. Tulis, The Rhetorical Presidency (Princeton:
Princeton University Press, 1987), 87ff. Gil Troy, Candidates
Take the Stump, Then and Now, letter, New York Times, Jan-
uary 17, 1988.
22. I have examined these at greater length in The Myth
of the Presidential Mandate, Political Science Quarterly 105
(Fall 1990): 35572.
23. Fred I. Greenstein, The Benevolent Leader: Chil-
drens Images of Political Authority, American Political Sci-
ence Review 54 (December 1960): 934 43. The views of
American school children about the president differ from
those of British and French school children about their chief
executives. Children and Politics in Britain, France, and the
United States: Six Examples, Fred I. Greenstein and Sidney
Tarrow, Youth and Society 2 (1970): 11128.
24. Edward Tufte and I undertook such an inquiry in Size
and Democracy (Stanford: Stanford University Press, 1975)
but regrettably the topic seems not to have attracted much
subsequent investigation.
25. Dahl and Tufte, Size and Democracy, 95ff.
n ot e s to pag e s 1 2 2 1 2 7 187
26. Though on a Purchasing Power Parity Basis, the differ-

ence diminishes to less than four times. U.S. Census Bureau,
Statistical Abstract of the United States, The National Data
Book, 1999 (Washington, D.C.: U.S. Government Printing Of-
fice, 1999), 841, Table 1362.
27. For more details, see Table 5.1, Appendix.
28. Lijphart (1999), supra, 3012.
29. Juan Linz and Alfred Stepan provide impressive evi-
dence that by providing more points at which privileged mi-
norities can veto the enactment of federal polices, the Ameri-
can federal system is the worst performer on social policies
among all of the OECD countries. A brief preliminary version
of their findings is Inequality Inducing and Inequality Reduc-
ing Federalism: With Special Reference to the Classic Out-
lierthe USA, paper prepared for the XVIII World Con-
gress of the International Political Science Association, August
15, 2000, Quebec City, Canada.
CHAPTER 6. Why Not a More Democratic Constitution?

1. Michael Schudson, The Good Citizen, A History of
American Civic Life (Cambridge, Mass.: Harvard University
Press, 1998), 202.
2. Constitutional Knowledge Survey, National Constitu-
tional Center, September, 1997, question 2.
3. Gallup Organization, 1999.
4. In the following sections I have drawn freely on my
essay The Future of Political Equality, in Keith Dowding,
James Hughes, and Helen Margetts, eds., Challenges to Democ-
racy (Hampshire, U.K.: Palgrave, 2001).
5. For a full account of failures to provide equal citizenship
among Americans, see Rogers M. Smiths masterly work, Civic
Ideals: Conflicting Visions of Citizenship in U S. History (New
Haven: Yale University Press, 1997).
6. Probably because he was heavily in debt, he freed only
five slaves on his death. Annette Gordon-Reed, Thomas Jeffer-
son and Sally Hemings: An American Controversy (Charlottes-
ville: University of Virginia Press, 1997), 38. Although his rea-
sons for freeing these five are unclear, all were related to his
188 n ot e s to pag e s 1 2 9 13 1
mistress, Sally Hemings, and two were probably sons by her.

Although the issue of paternity is disputed, Gordon-Reed pro-
vides strong circumstantial evidence. For her Summary of the
Evidence, see 210ff and see also Appendix B, The Memoirs
of Madison Hemings, 245ff. DNA tests provide additional cir-
cumstantial, though not conclusive, evidence. See Dinitia
Smith and Nicholas Wade, DNA Test Finds Evidence of Jef-
ferson Child by Slave, New York Times, November 1, 1998.
7. New York: Schocken Books, 1961, Henry Reeve, trans.,
vol. 1, p. lxxxi.
8. I have drawn these estimates from Adrian Karatnycky,
The 1999 Freedom House Survey: A Century of Progress,
Journal of Democracy 11, no. 1 (January 2000): 187200;
Robert A. Dahl, Democracy and Its Critics (New Haven: Yale
University Press, 1989): Table 17.2, p. 240; and Tatu Van-
hanen, The Emergence of Democracy: A Comparative Study of
119 States, 18501879 (Helsinki: Finnish Academy of Sci-
ences and Letters, 1984), Table 22, p. 120.
9. As one example, he writes, that among the untouch-
ables of India there is persuasive evidence that the Hindu doc-
trines that would legitimize caste domination are negated,
reinterpreted, or ignored. Scheduled castes are much less
likely than Brahmins to believe that the doctrine of karma ex-
plains their present condition; instead they attribute their sta-
tus to their poverty and to an original, mythical act of injus-
tice. Domination and the Arts of Resistance (New Haven: Yale
University Press, 1990), 117.
10. The late Joseph Hamburger showed that to secure the
expansion of the suffrage (and ultimately the passage of the
Reform Act of 1832), James Mill, though opposed to violence
as a means, deliberately set out to create a fear of revolution
among members of the oligarchy. Since Mill wished to
achieve fundamental reforms without violence, it became nec-
essary to devise means by which an oligarchy would be led to
grant concessions out of self-interest. . . . [T]here were only
two alternatives: [The people] can only obtain any consider-
able amelioration in their government by resistance, by apply-
ing physical force to their rulers, or at least, by threats so likely
n ot e s to pag e s 13 1 15 3 189
to be followed by performance, as may frighten their rulers

into compliance. Since the use of physical force was to be
avoided, Mill built his hopes on the second alternative. . . .
Mill was proposing that revolution be threatened. He assumed
that the threat would be sufficient and that it would not be
necessary to carry it out. James Mill and the Art of Revolution
(New Haven: Yale University Press, 1963), 2324.
11. I provide a fuller account in Democracy and Its Critics,
84ff. There and elsewhere I have drawn on Stanley I. Benn,
Egalitarianism and the Equal Consideration of Interests, in
J. R. Pennock and J. W. Chapman, Equality (Nomos IX) (New
York: Atherton, 1967), 6178.
12. This assumption is more fully developed in Democracy
and Its Critics, 105ff, and restated in briefer form in On
Democracy (New Haven: Yale University Press, 1998), 74ff.
13. For an excellent analysis, see Amartya Sen, Inequality
Rexamined (Cambridge, Mass.: Harvard University Press,
1992). Libertarians, he writes, must think it important that
people should have liberty. Given this, questions would imme-
diately arise regarding who, how much, how distributed, how
equal. Thus the issue of equality immediately arises as a sup-
plement to the assertion of the importance of liberty. (22)
14. These sentences are a close paraphrase of his state-
ments in 1: 298, 304, and 2: 38081.
CHAPTER 7. Some Reflections on the Prospects

for a More Democratic Constitution
1. CBS News /New York Times telephone poll of 1,254
adults, May 1987, question 53.
2. See also Amartya Sen, Inequality Reexamined, 3637
and passim. Ronald Dworkin, What Is Equality? Part 2:
Equality of Resources, Philosophy and Public Affairs 10 (1981).
3. Buckley v. Valeo, 424 U.S. 1 (1976).
4. For an extended argument consistent with my brief
comments, see John Hart Ely, Democracy and Distrust: A
Theory of Judicial Review (Cambridge, Mass.: Harvard Uni-
versity Press, 1980).
190 n ot e s to pag e s 16 0 16 2
APPENDIX A
1. The Federalist (New York: Modern Library, n.d.), 59.
2. Willi Paul Adams, The First American Constitutions:
Republican Ideology and the Making of State Constitutions in
the Revolutionary Era (Chapel Hill: University of North Car-
olina Press, 1980), 106ff.
3. The Federalist No. 39, in The Federalist, op. cit., 242ff.
4. Montesquieu, De lEsprit des Lois, Tome I (Paris: Edi-
tions Garnier Frres, 1961), Bk 2, Ch. 2, p. 12.
5. Ibid., Bk. 8, Ch. 16, p. 131. One might expect the rea-
son for this conclusion to lie in the difficulty of assembling the
people in a large territory. But in direct contradiction to Madi-
sons later argument in Federalist No.10 that the danger of fac-
tionalism would be reduced by increasing the size of the polit-
ical unit, Montesquieu contended that in a large republic the
common good would suffer. In a little (republic), the common
good is better felt, better known, and closer to each citizen.
6. For further discussion, see my Pluralist Democracy in
the United States (Chicago: Rand McNally, 1967), 34ff.
11dahlindex.191_198 11/27/01 4:42 PM Page 191
Index
Absolute majority, 56 Article II, 65

Adams, John, 45, 78 Article V, 144145
Adams, John Quincy, 69 Assumption of intrinsic equality,
Adams, Samuel, 45 131 132
Advanced democratic countries, Australia, 46, 98
4172, 91119, 164169 Austria, 50
African Americans: basic human
rights, 53; civil rights, 125, 147; Belgium, 4445, 98
discrimination, 28; voting rights, Bicameralism, 4546
16, 28, 128, 130. See also Slavery; Bill of Rights, 27, 30, 32, 51, 143
Suffrage Brazil, 49, 50
Alien and Sedition Acts, 26 Britain: bicameralism, 4546; civil
Amendments, 3234, 155. See liberties in, 98; constitution of,
also Electoral college; specific 65; constitutional system, 12,
amendment 7071; first-past-the-post
American Revolution, 95 system, 58; franchise, 127128;
Anti-slavery measures, 53 majoritarian system, 108109;
Argentina, 49, 50 parliamentary system, 7071;
Aristocracy, 1112 third party in, 103
Article I, Section 3, 144 Buckley v. Valeo, 151 152
192 index
Bundesrat, 105 change, 149; historical develop-

Burke, Edmund, 47 ments and, 910; inequality of
Burr, Aaron, 31, 78 representation in Senate,
144145, 154; judicial review,
California, 4950, 81 5455; limited role of,
Campaign speeches, 112 142 143; as model, 41 72;
Canada, 58, 116 models for, 9; as national icon,
Castes, reduction of discrimination 121123, 155; political practices
against, 130 and institutions, 2931; possi-
Chief executives, 6272. See also bility of significant change,
Head of state; President; Presi- 154157; practical limitations,
dential systems 910; presidential system,
Chile, 135, 137 143144; protections in, 51;
Civil Rights Act of 1964, 128, 130 public discussion of, 155156;
Civil Rights Acts, 130 rights, 142 143, 150 152;
Civil War, 95 shortcomings of, 1520; states
Cleveland, Grover, 69 and, 12; structures, 143149;
Committee on Detail, 66, 74 Supreme Courts role, 152155;
Compromises, 1213 survey results, 108109, 141;
Confederal system, 12 undemocratic elements in,
Congressional power, 1920; judi- 1520, 3839; utility and legiti-
cial review of actions, 5455 macy of, 39. See also Bill of
Connecticut, 27 Rights; Constitutional systems;
Connecticut Compromise, 18, 48 Political equality; specific
Consensual systems, 103109, amendment
117118, 146149, 154 Constitutional Convention,
Constitution: advanced democratic 18; delegates, 12, 45, 24;
countries, 4172, 91119, Founding Fathers and, 45;
164169; amendments, 2628; preparation for, 45; voting
brevity, 142; commitment to coalitions, 15
self-government and, 20; com- Constitutional systems: accounta-
promises and, 1213; consen- bility, 101102, 115; advanced
sual, majoritarian, or neither, democratic nations, 4172,
146149, 154; democratic 91119, 164169; British, 12,
changes to, 2631; emergent 7071; consensual systems,
democratic beliefs, 2026; 103 109, 117 118; defined, 41;
favorable conditions, 9496, democratic effectiveness,
142; federalism, 143; fundamen- 116119, 168169; democratic
tal questions and, 16; future fairness, 99103; diversity and,
democratic trends, 141157; 116; divided government,
hidden cost and uncertainties of 110111; electoral arrange-
index 193
ments, 99103; electoral system, 12; judicial review, 55;

systems, 3536, 5561, 166; oldest, 41. See also Constitution;
encouraging consensus, Constitutional systems
103109; evaluating perform- Democracy: breakdown of, 9596,
ance, 91119; favorable condi- 134135; citizen survey of satis-
tions, 9496, 99, 142; federal faction, 108109; creation of, 6;
or unitary, 4345; historical meaning of, 10; ordinary citizens
development, 10; hybrid systems, and, 25; rights, liberties, and
59, 62, 110115, 117119, opportunities, 136138; unfold-
146149, 167; maintaining ing of ideas and institutions, 10;
democratic stability, 9397; versus republic, 159162
majoritarian systems, 100103, Democratic Party, 24, 111
117118, 167; mixed system, Democratic-Republican Party,
6263; Netherlands, 104105, 24, 33
106, 107, 108109; parliamen- Democratic republics, 56, 10,
tary system, 63, 65; party 22 26
systems, 6162; performance Democratization, result of, 10
of, 91119; powers in, 149; Denmark, 45, 71, 116
presidential system, 6272, 96, Direct popular election: of presi-
111115, 143144; propor- dent, 8688, 155; of senators, 28
tional vs. majoritarian, 167; Discrimination, 28; Civil Rights
proportionality, 100109; Acts, 128, 130; in India, 130. See
protecting fundamental rights, also African Americans
9799; relative affluence Duvergers Law, 61
and, 116; size and, 116; state
authority, 51; Sweden, 106107; Elections: direct election of presi-
Switzerland, 105106. See also dent, 8688; direct popular,
Constitution 86 88, 155; of 1800, 37, 68,
Constitutionality of laws, 1819 77 78; of 1876, 7980; elec-
Continental Congress, 124 toral arrangements, 99103;
Costa Rica, 43, 62, 98, 116 Framers design for, 1617; of
Cuba, 148 president, 1617, 7389; of
senators, 28. See also Electoral
Declaration of Independence, college; Presidential elections
45, 21, 2223, 123, 124125 Electoral college, 7389; altering
Declaration of Rights, Virginia, 27, or abolishing, 8388, 179;
32 changes in, 8689, 145146,
Delegates, 12, 45, 24 155; choosing electors, 8283,
Democracies, 12; advanced demo- 8687, 155; constitutional
cratic countries, 4172, amendment, 8688, 155;
91119, 164169; federal creation of, 7376; democratic
194 index
Electoral college (continued) Federalism, 33, 4445, 143

changes in, 155; direct election Federalist No. 10, 29, 33, 34, 159,
of president, 8688, 155; district 161
system, 82, 87; election of 1800, Federalist No. 39, 160 161
31, 68, 7778; election of 1876, Federalist No. 68, 76
7980; election of 2000, 31, 68, Federalist Papers, 6364
73, 7981; entitlement to extra Federalists and Federalist Party,
protection, 8485; failure of, 24 26, 29, 30
7779; Framers design, 16; Fifteenth Amendment, 28
future of, 8689; improbable Finland, 63
outcome, 3031; inherent First Amendment, 26
defects, 7982; lacking majority First-past-the-post system,
of popular votes, 80; losing 56 62, 100, 166. See also Politi-
though preferred by majority, cal parties; Two-party systems
81; majority of popular votes, Founding Fathers, 45
80; need for extra protection, Fourteenth Amendment, 28
8586; popular vote vs. elec- Framers: awareness of limitations,
toral vote, 7981; reasons for, 142; defined, 45; Founding
7677; reform of, 8788, Fathers distinguished, 45;
145146; remedial defect, limits of opportunities, 1115;
8283; Senate and, 8788; practical limitations on, 910;
small states, 8486; swing what they couldnt know,
states, 83; unequal representa- 7 39. See also Constitutional
tion of voters, 8182, 171; Convention
winner-take-all-system, 8283; France, 26, 58, 63, 71, 98
winning with minority of popu- Franchise. See Elections; Suffrage;
lar votes, 8081, 170 Voting
Electoral systems, 3536, 5561, Freedom House, 98
166 French Revolution, 95
Equal representation in Senate, Fugitive Slave laws, 16
1315, 1718. See also Fundamental questions, 16,
Unequal representation 50 54
Equal Rights Amendment, 28 Fundamental rights, 51, 9799,
136 138, 153 154, 155
Factions, 2930, 33
Federal courts, 1819 Georgia, 27, 81
Federal Election Campaign Act, Germany, 46, 98, 137
151152 Gore, Al, 81
Federal systems, 12, 4350, 49, Government: limits on, 143;
51, 54. See also Constitutional majority rule, 3637; popular
systems rule, 2425; proto-republican
index 195
phase, 2122; republican, 11; 3233; representative of popu-

requirements of self govern- lar will, 69; role in two-party
ment, 2122. See also Constitu- politics, 2930; slavery and,
tional systems 125; supporter of democratic
Greece, 160 republic, 2325
Gross National Product (GNP), 116 Judicial legislation, 19, 20, 153155
Judicial power, 1819
Hamilton, Alexander, 11, 1315, Judicial review of national legisla-
64, 76, 77 tion, 5455
Hayes, Rutherford, 80
Head of state, 71, 112. See also Lnder, 43
President; Presidential systems Landowners, 25, 3536
House of Commons, 59, 128 Legislation, judicial review of,
House of Lords, 17, 4546, 128 54 55
House of Representatives, majori- Lijphart, Arend, 103104,
ties in, 110 117 118
Hybrid systems, 59, 62, 110115, Lincoln, Abraham, 69
117119, 146149, 167
Madison, James, 5, 78; choosing
Iceland, 116 chief executive, 17, 66; creation
Incarceration, 117, 169 of political party by, 24; demo-
Income inequality, 92 cratic revolution and, 3137;
Income taxes, 28 drafting amendments, 27; early
Independent Commission on the democrat, 10; Federalist Papers
Voting System, 59 and, 33, 64; intention to create
India, 130 republic versus democracy, 56,
Inequality of representation in 159162; judicial power and,
Senate, 144145, 154. See also 19; life of, 31 37; minority
Senate interest protection and, 51;
Israel, 43, 98, 104 opposed to Federalist policies,
Italy, 98 3233; political development of,
3137; principle architect of
Jackson, Andrew, 24, 69, 112 Constitution, 5; representative
Japan, 43, 71 of popular will and, 69; role in
Jay, John, 64, 78 two-party politics, 2930;
Jefferson, Thomas: creation of Senate representation and,
political party by, 24; democracy 1315; separation of powers
and, 10; election of 1800, 31, 78; and, 65; small state protection
foreseeing democratic republic, and, 52; supporter of demo-
25; as Founding Father, 45; cratic republic, 2324; unequal
opposed to Federalist policies, representation and, 84
196 index
Majoritarian systems, 62, 100109, Partisan politics, 7879

117118, 146149, 154, 167 Party systems, 6162, 167. See
Majority: absolute majority, 56; in also Constitutional systems;
House of Representatives, 110; Multiparty systems; Political
relative majority, 56, 100; three parties; Two-party systems
majorities, 110. See also Majority Peaceful democratic revolution,
rule 8 10, 20
Majority rule, 3637, 5054, 101. Plurality, 56, 100
See also Majoritarian systems Political equality, 123139;
Mandate, presidential, 6970, 113 greater, 128130; important
Mason, George, 19, 27, 32 democratic value, 4; justifiable
Massachusetts, 82 goal, 130132; realistic goal,
Mayhew, David, 110111 123128; strategy for future,
Mill, John Stuart, 60 156157; threaten liberty,
Minorities: balancing rights and 132 138
interests in representation, Political parties, 2930, 3334.
5054; eliminating discrimina- See also Constitutional systems;
tion against, 28. See also African Multiparty systems; Two-party
Americans systems
Minority parties, 100 Poll tax, 28
Monarchies, 11, 65, 7071 Popular rule, 2425
Monroe, James, 69 President: American hybrid system
Morris, Gouverneur, 11, 15, 17, 77 and, 111115; campaign
Multiparty systems, 5759, 100, speeches, 112; change in role of,
102. See also Constitutional 155; creation of, 6668; election
systems of, 16 17, 73 89; failure of
Framers design, 6872;
Native Americans, 16, 128 Framers design for, 6468;
Nebraska, 45 judicial review of actions,
Netherlands, 98, 104105, 106, 54 55; mandate, 69 70, 113;
107, 108109 myth of the presidential
New England town meetings, mandate, 6970; role of,
150151, 160 111 115. See also Electoral
New Hampshire, 12 college; Presidential elections;
New York, 4950 Presidential systems
New Zealand, 127 Presidential elections: of 1800, 31,
Nineteenth Amendment, 28 68, 77; of 1876, 79 80; of 2000,
Norway, 45, 71, 98, 116 31, 68, 73, 79 81
Presidential systems, 6272, 96,
Parliamentary Commission, 46 143 144. See also Electoral
Parliamentary systems, 63, 6568, college; President
7071, 96 Prime ministers, 63, 7071
index 197
Proportional systems, 5759, 154; direct elections of senators,

100109, 166, 167. See also 28; election of senators, 28;
Constitutional systems; Multi- electoral college and, 8788;
party systems; Political parties equal representation in, 1315,
Proto-republican phase, 2122 1718; inequality of representa-
Prussia, 48 tion in, 144145, 154; majorities
Pseudo-democratization, 70 in, 110; representation in,
1314; unequal representation,
Reconstruction, 53 48 50, 144 145, 148
Regulatory powers, 19 Separation of powers, 65
Relative majority, 56, 100 Seventeenth Amendment, 28
Republican Party, 29 Sixteenth Amendment, 28
Republican phase, 22 Slavery, 13, 15 16, 27 28, 53,
Republicanism, 160 124 125
Republics: aristocratic, 56; citi- Small states, 5254, 8486
zen body of, 25; creation of, 5; South Carolina, 82
democratic, 56, 10, 2226; Southern veto, 53
form of government, 11, 20; Soviet Union, 41
proto-republican phase, 2122; Spain, 71
republican phase, 22; Roman, 8; States, 4344; judicial review of
selecting chief executive, actions, 54; role of, 12; small,
6465; Third Republic in 52 54, 84 86. See also Federal
France, 71; U.S. transition to systems
democratic republic, 2226; Suffrage, 16, 3436; African
Venetian, 8; versus democracy, Americans, 28; Britain,
159162 127128; New Zealand, 127;
Revolution, peaceful democratic, unequal representation and,
810, 20 4750; universal, 127; women,
Rhode Island, 12 16, 28, 127, 128. See also Elec-
Rights: fundamental, 51, 9799, tions; Voting
136138, 153155; minority Supreme Court, 55, 152155
interests, 5054 Sweden, 12, 45, 71, 116
Roosevelt, Franklin D., 69 Swing states, 83
Roosevelt, Theodore, 69 Switzerland, 44, 46, 50, 98,
Russia, 41, 50 105 106, 116
Scotland, 59 Taxation, 19, 28, 48

Second Continental Congress, 124 Third parties, 83, 100, 103
Self-government requirements, Third Republic (France), 71
2122 Thirteenth Amendment, 28
Senate: choosing of senators, 17; Tocqueville, Alexis de, 4, 23,
constitutional amendment and, 126 127, 133 135
198 index
Tories, 30 tional system, 91119; presi-

Town meetings, 150151, 160 dential system, 6272; ranking,
Twelfth Amendment, 31, 78 117, 168 169; size, 116;
Twenty-Fourth Amendment, 28 unequal representation, 4849
Twenty-Sixth Amendment, 28 Uruguay, 135, 137
Two-party systems, 29, 5658,
6162, 100103, 110. See also Veto, 69; judicial, 19; Southern, 53
Multiparty systems Vice president, election of, 31, 78
Virginia, 27, 32
Unequal representation: balancing Virginia Plan, 1819, 32, 68
rights and interests, 5054; Virtual representation, 47
second chamber, 4654; Senate, Vote of confidence, 70
4850, 144145, 148; slavery, Voting, 28, 150; African Americans,
53. See also Equal representa- 16, 28, 128, 130; age, 28; India,
tion in Senate 130; poll tax, 28; property tax
Unitary systems, 12, 4345 classes, 48. See also Elections;
United States: bicameralism, 46; Suffrage
comparison with advanced
democratic countries, 4172, Wales, 59
91119; democratic effective- Whigs, 30
ness, 168169; diversity, 116; White males: land owners, 25;
first-past-the-post system, voters, 79
5859; fundamental rights, 98; Wilson, James, 1315, 52, 54,
hybrid system, 59, 62, 110115, 73 74, 75 76, 84
117119, 146149, 167; incar- Wilson, Woodrow, 69, 112
ceration in, 117; majoritarian Women: discrimination, 28;
system, 62, 167; monarchy and, suffrage, 16, 28, 127, 128
71; performance of constitu- Wyoming, 50, 81

Dahl Robert How Democrati Is The American Constitution

Uploaded by

Copyright:

Available Formats

Dahl Robert How Democrati Is The American Constitution

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dahl Robert How Democrati Is The American Constitution

Uploaded by

Copyright:

Available Formats

00dahlFM.

i_x 11/27/01 4:37 PM Page i

The Castle Lectures in Ethics, Politics, and Economics

Yale University Press / New Haven & London

Published with assistance from the foundation established in memory of

Copyright 2001 by Yale University.

Designed by James J. Johnson and set in New Caledonia and

Library of Congress Cataloging-in-Publication Data

Dahl, Robert Alan, 1915

Parts of this book were given as a series

The Castle Lectures, endowed by Mr.

THE INVITATION TO DELIVER THE CASTLE LECTURES AT

For her invaluable research assistance I am in-

states sent delegates, in fact Rhode Island refused to

Our citizen might now fall back on another line of

important democratic value is political equality, wont

BEFORE TURNING TO THESE QUESTIONS, I NEED TO DIS-

delegates to the Convention had also signed the Dec-

ship of James Madison, among others, Americans would

What the Framers Couldnt Know

phia,1 Madison studied the best sources as carefully

1787. But reliable knowledge about constitutions

Leaping into the Unknown

Among the important aspects of an unforeseeable fu-

Second, partly in response to that continuing revo-

What the Framers Couldnt Do

The Framers were not only limited by, so to speak,

that the delegates could not agree on just who these

members of the convention held different views on

vidual men which ought we to respect most, the rights

The large states dare not dissolve the Confederation. If

To which Rufus King of Massachusetts replied:

I cannot sit down, without taking some notice of the lan-

Faced with the refusal of the small states to ac-

cepted compromise of principle as the price of a con-

Undemocratic Elements in the

It was within these limits, then, that the Framers con-

on slavery not only denied Congress the effective

dom and virtue who would choose the chief executive

of the famousor from a democratic point of view, in-

number of the National Judiciary, ought to compose a

matters,19 unless the constitution could be altered by

The Framers Constitution Meets

We may tend to think of the American republic and its

historical experience would reveal, in countries where

bother to read todaythe authors mainly protested

before most Americans would be inclined to agree

During the three decades before Tocqueville ar-

the views of the Framers and the Federalists is sym-

and to property rights. Contrary to these pessimistic

in 1798 by the Federalists, who were alarmed not only

Democratic Changes to the

The fate of the Alien and Sedition Acts symbolizes a

The constitutional system that has emerged is no

As I have mentioned, the most profound violation of

Thirteenth, Fourteenth, and Fifteenth Amendments

The constitution of the Framers was changed not only

Whatever the best answers to these hard ques-

From the time the final results were known in late