January 2007, Volume 18, Number 1 $11.

00

How Democracies Emerge

Thomas Carothers Sheri Berman

Revolution Reconsidered
Charles H. Fairbanks, Jr.

The Perpetual Crises of Democracy

Guillermo O’Donnell

Benjamin Reilly on the Asia-Pacific

Gideon Maltz on Presidential Term Limits
Thomas B. Pepinsky on Malaysia
Axel Hadenius & Jan Teorell on Paths from Authoritarianism
Gideon Rahat on Candidate Selection

The Mexican Standoff

Andreas Schedler Luis Estrada & Alejandro Poiré
Jorge G. Casta~neda & Marco A. Morales
 How Democracies Emerge

THE “SEQUENCING” FALLACY

Thomas Carothers

Thomas Carothers is vice-president for international politics and gov-

ernance and director of the Democracy and Rule of Law Project at the
Carnegie Endowment for International Peace. His most recent book is
Confronting the Weakest Link: Aiding Political Parties in New Democ-
racies (2006).

I n the second half of the 1990s, a counterreaction emerged to the heady

enthusiasm about democracy and democracy promotion that flourished
during the peak years of democracy’s “third wave” in the late 1980s and
early 1990s. Believing that the global democratic wave had been over-
sold, several policy experts and scholars produced a series of influen-
tial articles articulating a pessimistic, cautionary view. Fareed Zakaria,
alarmed by what he saw as a dangerous rash of newly elected leaders
restricting rights and abusing power from Peru and Argentina to the
Philippines and Kazakhstan, warned that rapid democratization was
producing a plague of “illiberal democracy.”1 Troubled by violent con-
flicts breaking out in former Yugoslavia, the former Soviet Union, and
elsewhere, Edward Mansfield and Jack Snyder argued that democratiz-
ing states are in fact more conflict-prone than stable autocracies.2 Dis-
turbed by the specter of ethnic conflict in different parts of Asia, Amy
Chua asserted that the simultaneous pursuit of democracy and market
reform in countries with “market-dominant minorities” leads to ethnic
conflict and antimarket backlashes.3
Although their specific areas of focus and analytic frameworks var-
ied, these different accounts coalesced around a central argument that
appealed to what might be called the need for democratic sequencing.
In this view, it is a mistake to assume that democratization—especially
open national elections—is always a good idea. When tried in countries
poorly prepared for it, democratization can and often does result in bad
outcomes—illiberal leaders or extremists in power, virulent national-
ism, ethnic and other types of civil conflict, and interstate wars. To

Journal of Democracy Volume 18, Number 1 January 2007

© 2007 National Endowment for Democracy and The Johns Hopkins University Press
Thomas Carothers 13

prevent such results, certain preconditions, above all, the rule of law
and a well-functioning state, should be in place before a society democ-
ratizes. Therefore, the United States, and the West generally, should
rethink their approach and commitment to democracy promotion. In
some countries, staying with an existing autocratic regime is a better
alternative. Where outside actors do want to promote positive political
change in a nondemocratic society, they should concentrate first on
helping it to achieve the rule of law and a well-functioning state. Only
much farther down the road, when those preconditions are established,
should outsiders push for elections and the other associated elements of
what sequentialists refer to, warily, as “mass political participation” 4 or
“mass plebiscites.”5
Democratic sequentialism is one part of a wider body of skeptical
thinking about democracy’s global prospects that gained popularity in
the 1990s in reaction to third-wave enthusiasm. This wider body, which
might be called “democratic pessimism,” is represented most vividly in
the much-discussed writings of journalist Robert Kaplan.6 Sequential-
ism has found a vital place in this more generalized pessimism thanks to
its concrete policy implications and intuitive appeal: Pursuing a se-
quential path promises to rationalize and defang democratic change by
putting the potentially volatile, unpredictable actions of newly em-
powered masses and emergent elected leaders into a sturdy cage built of
laws and institutions.
Sequentialism met a warm welcome from various parts of the interna-
tional policy community. Traditional realists, who had been uneasily
eyeing the ascendancy of a prodemocratic policy outlook in the 1980s
and 1990s, were happy to have an additional set of arguments for
downplaying democracy promotion and maintaining cordial relations
with friendly autocracies. Traditional developmentalists, still loyal to
old-school modernization theory’s notion that development must pre-
cede democracy and feeling upset by democracy promotion’s sudden
rise to prominence, were delighted to have a new set of contrarian allies.
Finally, powerholders in some nondemocratic countries eagerly em-
braced sequentialism in order to argue that their reticence about open-
ing up political competition actually reflected a deeper commitment to
democracy in the long term.
Sequentialism has continued to gain attention and adherents in this
decade. The initial articles by Zakaria, Mansfield and Snyder, and Chua
have all returned in expanded form as successful books.7 Democracy’s
continued trials and tribulations, from the backsliding in the former
Soviet Union to the swelling populist currents in Latin America, have
provided plentiful grist for pessimism regarding this form of govern-
ment. The troubling results of the Bush administration’s loudly pro-
claimed drive for democracy in the Middle East have been especially
rich fodder. Sequentialists interpret the searing, dispiriting experience
14 Journal of Democracy

of attempted democracy-building in post-Saddam Iraq as telling evi-

dence of the consequences of an extreme application of a “no-precondi-
tions” outlook on democratization. Similarly, the gains of Islamists in
recent elections in Egypt, Iraq, Lebanon, and Palestine appear to some
observers to be further indications of the dangers of moving quickly
toward elections in countries with little democratic history.
Democratic sequentialism raises important questions for the Western
democracy-promotion community. Taking the measure of this line of
thought is crucial, especially in the context of a deepening debate in
the United States and worldwide over the legitimacy of democracy pro-
motion and even the value of democracy overall. Some aspects of demo-
cratic pessimism may be useful correctives to what undoubtedly have
been excesses associated with the democracy-promotion cause, such as
the view of democratization as an international cure-all that will solve
terrorism and many other scourges. But the idea of sequencing is prob-
lematic, as are the policy recommendations that flow from it. A more
useful alternative for taking into account the many complications and
risks of democratization and democracy promotion is gradualism, which
aims at building democracy slowly in certain contexts, but not avoid-
ing it or putting it off indefinitely.

Don’t Wait for the Rule of Law

The idea of sequencing rests on a mistaken two-part premise: that a
significant number of autocrats can and will act as generators of rule-of-
law development and state-building, and that democratizing countries
are inherently ill suited for these tasks. Autocracy, however, whether
liberal or nonliberal, is inherently in tension with both rule-of-law de-
velopment and state-building. New or struggling democracies certainly
encounter many difficulties in building the rule of law and the state, but
they do not face any such inherent contradiction in doing so and enjoy
at least a few advantages as well.
With regard to rule-of-law development, the presumption of sequen-
tialists appears to be that autocrats—at least the liberal ones who figure
so large in sequentialist accounts—will sedulously pursue reform ei-
ther because they are enlightened and thus instinctively committed to
reform, or because they feel impelled to do so by the imperatives of
economic development. That is to say, in order to advance economi-
cally—which they presumably want their countries to do to keep citi-
zens happy—they will try to strengthen the rule of law on the assumption
that more efficient courts, clear and predictable laws, and other such
legally related reform elements are necessary for economic progress.
The reality, however, is otherwise. Despite the persistent hopes of
the sequentialists, few autocrats are in fact enlightened. That is to say,
few keep repression to a minimum, respect law, and are committed to a
Thomas Carothers 15

serious long-term path of gradual political and governance reforms.

Bluntly stated, for every Lee Kuan Yew of Singapore there have been
dozens or even hundreds of rapacious, repressive autocrats posing as
reformers, leaders for whom the rule of law represents a straitjacket to be
avoided at all costs.
As for the notion that a desire for economic development leads natu-
rally to rule-of-law reform, this mechanism is weaker than it may first
appear. Some autocratic regimes are simply uninterested in economic
development and the welfare of their citizens. These regimes care only
about holding onto power and are willing to use unrelenting, harsh
repression to do so. Many autocrats have some interest in developing
their country but they subordinate it to other interests that not only
compromise their socioeconomic policies but undermine any serious
pursuit of the rule of law. Thus, despite some commitment to socioeco-
nomic progress, such leaders may also be fixated on enriching them-
selves, protecting certain privileged groups or sectors in the society,
and undercutting potential political rivals. These other interests usu-
ally require deforming the rule of law in significant, even systematic
ways. The terrible socioeconomic conditions and weak rule of law ap-
parent in so many developing countries are, in many cases, a legacy of
decades of misrule by autocratic regimes that claimed a deep commit-
ment to developmental goals but in fact gave greater priority to nar-
rower, self-interested, and countervailing concerns.
A small number of nondemocratic governments are sufficiently com-
mitted to economic development to put other interests on a lower tier
for the sake of pressing forward with hard, costly reform policies. The
developmentally successful East Asian states, past and present, that
sequentialists usually cite as models belong to this small set. These
governments may well find that their economic drive does create a need
for at least some rule-of-law reforms, such as updating commercial leg-
islation and creating functional commercial courts. Yet it remains true
that substantial economic progress can be achieved without major
progress toward the rule of law. The cases of China and Vietnam are
telling in this regard, and they weigh heavily in the calculations of
policy elites in nondemocratic countries around the world as they con-
template their developmental options. Both countries are remarkable
economic success stories, particularly in terms of lifting large numbers
of people out of poverty, yet they have produced sustained growth
without adopting substantial parts of the overall rule-of-law agenda.
In other words, to the extent that some autocratic governments do
feel pushed toward rule-of-law reform by the desire for successful eco-
nomic development, it is only in the direction of a highly truncated
version of the rule of law. This version usually emphasizes elements
directly relating to the commercial domain. Missing from it will be
cardinal rule-of-law features such as civil and political liberties, the
16 Journal of Democracy

subordination of political power to the law, and independence of the

judiciary. Some Western observers hope that reforms in the commercial-
law domain will over time engender reforms in the more political areas
of the rule of law, with commercial law reform functioning as a sort of
Trojan horse for broader rule-of-law change. Appealing though this idea
may be, it is based more on hope than on experience. The story of rule-
of-law development in countries with well-developed rule-of-law sys-
tems is not one in which the commercial domain drove progress in other
areas. And careful analyses of whether the spreading of reforms across
domains is actually happening in countries such as China have been
reaching skeptical conclusions.8
Some autocratic governments trumpet their embrace of rule-of-law
reform for reasons other than those of economic development. This em-
brace usually takes the form of anticorruption campaigns or crime-re-
duction policies. Again, however, these programs are typically quite
limited in scope and avoid essential components of genuine rule-of-law
reform. The anticorruption campaigns, for example, usually consist of
some technocratic legal changes, public education about the evils of
corruption, the establishment of an anticorruption commission, and per-
haps a few selective prosecutions of corrupt officials. The campaigns
rarely entail a serious effort to produce deep-reaching systemic change,
starting from the top and working down, by applying the law impartially
and fully to all public officials and eliminating political constraints on
and manipulation of the principal law-enforcement agencies. Such cam-
paigns are not natural precursors to democratization. On the contrary,
they typically form part of a defensive strategy designed to mollify pub-
lic discontent and undercut rivals, all with the aim of enhancing rather
than reducing the durability of the autocratic government’s grip on power.

Autocracy versus the Rule of Law

The core point is that rule-of-law development and autocracy (liberal
or otherwise) go poorly together. Key elements of the rule of law directly
threaten autocratic rule. An independent judiciary will be a source of
power and authority beyond the executive’s reach. Impartial adjudica-
tion, fair and equal treatment of all persons before the law, respect for
political and civil rights—all these essential components of the rule of
law restrict or remove the tools that autocrats typically employ to control
political life and stay in power. Autocrats tend to block or at least trun-
cate the rule of law. Thus the idea that rule-of-law development under
autocracy is a natural precursor to democracy gets the story backwards. It
is the lack of democracy—that is, the persistence of autocracy in many
countries—that is a fundamental obstacle to rule-of-law development.
Egypt is a telling case in point. During the 1990s, the U.S. government
pushed the government of President Hosni Mubarak to carry out rule-of-
Thomas Carothers 17

law reforms, especially judicial reform, to facilitate Egypt’s attempted

economic reforms and to help lay the foundation for political reform and
democratization further down the road. The U.S. Agency for International
Development mounted one of its largest-ever rule-of-law assistance pro-
grams to help the Egyptian government move ahead in this domain.
Despite these efforts, however, rule-of-law reform made little progress
in those years and remains blocked today. The Egyptian government
continues to exert political control over the adjudication of sensitive
cases and to oppose efforts by judges to exert greater independence.
The Egyptian security forces continue to operate in an opaque and
frequently abusive manner. The government maintains the 25-year-old
state of emergency, limiting citizens’ basic rights and preserving wide
discretionary legal authority for the government.9 Whether it is Western
or Egyptian officials who voice it, the argument that Egypt must lay
down a rule-of-law foundation before it can raise the edifice of democ-
ratization verges on the self-serving. It seems to provide a way for the
West to justify its continued tolerance of a lack of democracy in Egypt,
and for the Mubarak government to excuse its own nondemocratic be-
havior. Democratization in Egypt does not await the rule of law. Rather
it is the lack of genuine democratization there—thanks to the Mubarak
government’s paralyzing grip—that is preventing rule-of-law develop-
ment. Only if there is a serious political opening, including free and fair
competition for the presidency, will there be any real hope for deep-
reaching reforms to build the rule of law.
Russia provides another relevant example. When President Vladimir
Putin came to power in 2000, rule-of-law reform was one of his declared
priorities. Many Russians, as well as Western officials, applauded this
emphasis, believing that Russia had erred in the 1990s by proceeding too
rapidly with democratization without having first created a proper rule-
of-law base. Yet Putin’s rule-of-law drive has foundered. The exercise of
law in Russia today is marred by serious, systematic deficiencies that stem
from Putin’s approach to wielding political power: the politicized use of
statutes to punish political opponents, the continued growth of a shadowy
and unaccountable security elite with its hands on the state apparatus, and
politically based favoritism and interference in adjudication.
Putin’s backing away from democracy feeds these rule-of-law prob-
lems. Reduced media freedom, diminished tolerance for political oppo-
sition, and the undermining of electoral mechanisms that might establish
basic political accountability undercut the capacity of Russian society
to call attention to and push back against rule-of-law abuses and short-
comings. As in Egypt, prescribing more rule of law in Russia as the basis
for eventual democratization gets it backwards: More democratization
is vital to strengthening the rule of law.
Highlighting the intrinsic tensions between autocracy and the rule
of law does not mean that rule-of-law development will be simple or
18 Journal of Democracy

inevitable under democracy. Democratizing countries struggle hard with

rule-of-law development and fall short in countless ways. Various as-
pects of democratization can directly contribute to rule-of-law prob-
lems. The introduction or reintroduction of elections into a formerly
closed society can heighten political patronage in the judicial domain.
The relaxation of authoritarian controls can fuel a generalized weaken-
ing of law enforcement, leading to a surge in crime. Political parties
grasping for power in new democracies exhibit a distressing tendency
to become cesspools of corrupt financing and influence.
At the same time, other aspects of democratization facilitate rule-of-
law development. As Rebecca Bill Chavez shows in her illuminating
study of the state of law in two Argentine provinces, alternation in power
contributes significantly to rule-of-law progress by helping to break up
stagnant concentrations of power.10 The lines of accountability between
citizens and the state that elections help to create, however imperfectly,
fortify efforts to hold public officials to the law, a central part of rule-of-
law development. Despite frequent failings in criminal-law enforcement,
even very weak democratizing governments usually greatly reduce gross
abuses of human rights compared to their authoritarian predecessors.
Most importantly, unlike with autocracy, no inherent conflict exists
between democracy and the rule of law. On the contrary, the two con-
cepts are inextricably intertwined. They share crucial components such
as respect for political and civil rights and the subordination of power-
holders to law. Many elements of democratization bring progress to-
ward the rule of law and vice-versa. The close association around the
world of established rule of law and consolidated democracy is not a
coincidence. As Marc F. Plattner has argued, attempting to draw a line
or posit a sequential path between rule-oriented liberalism and liberal
democracy is futile—the history of the established Western democra-
cies is in many places a story of their simultaneous development and
mutual reinforcement.11

The State-Building Challenge

Similar arguments apply to state-building, the other precondition for
democracy that sequentialists frequently prescribe. It is true that many
democracy promoters have tended to underestimate the importance of a
capable state. They have taken too long to move away from the assump-
tion, which took hold during the peak years of the third wave, that the
core challenge for democracy promotion in countries moving away from
authoritarianism is diminishing or counterbalancing an overly strong
state.12 In many authoritarian societies, the state is heavy-handed and
repressive but weak in terms of core capacity, for reasons discussed
below. Thus in most postauthoritarian contexts, strengthening state ca-
pacity should usually be a priority alongside democratization.
Thomas Carothers 19

In certain situations, democratization does need to wait for state-build-

ing. Where a state has completely collapsed or failed under the lash of
civil conflict or other accumulated or acute calamities, moving rapidly
toward open political competition and elections makes no sense. The
state will need to have at least minimal functional capacity as well as
something resembling a monopoly of force before such a country can
pull itself onto the path of sustainable, pluralistic political development.
Yet the sequentialists are misguided in propounding the idea that
democratization must wait until there exists not merely an adequate but
a well-functioning state—one with capable, impartial institutions and a
solid capacity to develop, legislate, and implement effective policies.
This mistaken idea rests on the seductive but illusory notion that auto-
crats are natural state builders, harking to the image of firm, sober
strongmen imbued with a state-building ethos, ` a la Mustafa Kemal
Atatürk or Otto von Bismarck. Yet as with rule-of-law development, for
every effective state-building strongman of the past century, there have
been dozens more autocrats whose self-serving, erratic leadership has
resulted in tremendously debilitated states rife with patronage, corrup-
tion, and incompetence.
As Joel Migdal observes in his definitive work on state-building in
developing countries, autocrats are in fact poorly suited for the second
phase of state-building—that is, the one that goes beyond asserting a
monopoly on force and creating a basic state framework in order to de-
velop an effective state bureaucracy that can carry out the many func-
tions demanded of a modern state. 13 It is this second phase that
sequentialists emphasize when they talk about the need for “a devel-
oped infrastructure of impartial administration” prior to democratiza-
tion.14 Like rule-of-law development, with which it is linked, this second
phase of state-building is in fundamental tension with autocratic rule.
Autocrats habitually misuse the state—as a source of corrupt income for
themselves and their cronies, as a tool to persecute and suppress critics
or rivals, and as a parking place for favored friends and groups. Conven-
tional state-building goals such as efficiency and impartiality only ren-
der a state apparatus less malleable for the autocrats’ purposes. An
effective state enjoys significant political autonomy, as well as its own
legitimacy and authority. In other words, such a state is independent in
important ways from the political levers of power and therefore naturally
threatening to an autocratic leader. This is not to say that autocratic
governments are incapable of state-building, only that it is seriously
misguided to view them as natural or likely masters of the process.
Fledging democratic governments also struggle with state-building.
As with rule-of-law development, various aspects of democratization
can complicate the task. New political parties looking for sources of
strength may abuse the state in their push to build patronage networks.
A period of weak, shifting coalition governments can set back institu-
20 Journal of Democracy

tion-building and sustained policy implementation. At the same time,

however, some parts of a democratization process can contribute to
state-building. The establishment of public accountability through vot-
ing may create incentives for good state performance. Opening up a
closed media will allow greater public scrutiny of poorly performing
areas of state function. Creating space for independent civil society
permits advocacy groups to monitor and critique state performance and
work together with the state to offer new policy ideas.
State-building is not necessarily successful in new democracies. Yet
unlike with autocracies, there is no basic underlying tension between
an effective state and a successful democratic government. It is no coin-
cidence that almost all of the most effective states in the world today are
in democratic countries. As with the rule of law, state-building beyond
the initial stage is best pursued at the same time as democratization,
with an effort to find points of complementarity and mutual reinforce-
ment. Doing this will involve many complications, of course, but the
alternative—betting that an autocratic government will avoid lapsing
into the dysfunctional habits that have undermined the state perfor-
mance of so many autocracies in the past—is usually worse.

Overstating the Role of Outsiders

A further serious problem with the sequentialist outlook is its im-
plicit notion that democratization in many countries can be put off
until a better day if the United States and other Western powers would
just stop pushing it so hard. The writings of the sequentialists give the
impression that a substantial part of the global wave of democracy is
merely electoral froth, caused by the West promiscuously pushing for
elections all over the world. This view misconstrues the third wave, in
particular by underestimating the normative shift in favor of elections
that has occurred around the world. It also overestimates the consis-
tency and power of Western support for elections and democracy in
developing and postcommunist countries.
The third wave of democracy has certainly turned out to be less sweep-
ing and successful than many had hoped. Yet it has reflected a genuine,
often powerful impulse toward democracy on the part of people in vastly
disparate regions and countries. The many plunges into democratiza-
tion, often including rapid movement to elections, have usually oc-
curred because citizens of those societies wanted it that way. This has
been true where citizens rose up to drive dictators out—people mobi-
lized not just against the existing regime but for their own political
empowerment, for the right to choose their leaders themselves. It has
also been true in cases where dictatorial regimes have simply collapsed
due to the death of the leader, as in Nigeria in 1998 or Togo in 2005.
Citizens in such contexts have responded to the sudden political dis-
Thomas Carothers 21

juncture with a strong, quickly awakened desire for elections, deriving

not from outsiders’ urgings but rather from their own sense that “now
it’s our turn.”
The global spread of the idea that being elected is the root of govern-
mental legitimacy has been integral to the third wave. Yet the legiti-
macy that elections confer has often been much shakier than democracy
promoters (and those elected) have hoped, and elections in many new
democracies have done less than expected to create well-functioning
mechanisms of democratic accountability. A handful of nondemocratic
governments still maintain significant legitimacy without elections,
whether through cultural norms concerning inherited rule, good eco-
nomic performance (or at least the reasonably competent distribution of
economic largesse from oil revenues), or the mantle of national struggle
against foreign enemies. Weak and problematic though elections often
are, they now form a crucial step in the process of attaining political
legitimacy throughout most of the world. The laborious, often convo-
luted efforts of many semiauthoritarian leaders in the former Soviet
Union, Africa, the Middle East, and elsewhere to carry out at least
vaguely plausible elections, rather than just to rule without them, are
evidence of this fact. And frustrating though the quest for democratic
accountability through elections has been in many places, the chances
for building accountability under autocracy are usually worse. It may
be true that in many countries democracy can barely live with elections,
but in no country can it live without them.
The idea that there has been a rash Western drive for democracy at all
costs would likely provoke a hearty, cynical laugh from people in most
parts of the world if it were proposed to them. Certainly the United
States, and to a lesser extent other Western powers, have often talked
grandly in the past several decades about their commitment to global
democracy. But underneath the rhetoric is a long record of a very mixed
policy reality. Where democratic change in a particular country or re-
gion aligns with Western economic or security interests, it receives
support. In many places, however, the United States and Europe have
been and continue to be quite happy to support or get along with auto-
cratic governments for a host of reasons.
Once democratization clearly starts somewhere, a collection of West-
ern governments and other organizations usually steps in to back it. Much
less frequently, however, is the West ahead of the political curve, push-
ing hard for change where autocracy is stable. The case of Indonesia in
the final phase of President Suharto’s rule is a valuable example to recall.
As Suharto began to lose his grip, the United States and other Western
powers were hesitant to push hard for democracy, loyal up to the very end
to a leader who had been good to them, and fearful of what might come
after him. It was only after Indonesians drove him out that the West came
in with substantial diplomatic and material support for democracy.
22 Journal of Democracy

The role of outside actors in most attempted democratic transitions is

relatively limited. A wide range of governments, international organi-
zations, aid groups, and transnational nongovernmental organizations
provide what is usually nonconfrontational democracy assistance. They
may cajole, beseech, or pressure officials in the country in question to
move forward with democratic reforms. Western democracy promoters
have supported hundreds of elections in new or struggling democracies
through help with election administration, domestic and international
election observing, and political-party training. Many of these democ-
racy programs and policies are useful, but they overwhelmingly play
only a supporting and not a leading role. Large though it has grown, the
democracy-promotion community does not in most cases drive or even
shape political transitions in the world. Instead, it backs them, trying to
help domestic actors achieve what they have already decided they want
for themselves.
This is usually true even when the external role is loud and seemingly
large. It is a mistake, for example, to attribute the holding of various
elections in the Arab world during the past two years to the much-publi-
cized U.S. pressure for democratization. This attribution is often asserted
as part of a critique: The Bush administration pushed Arab countries to
hold elections, the argument goes, and look what it got—a string of
Islamist victories. Palestinians themselves pushed to hold elections after
the death of Yasir Arafat in 2004, who himself had been elected eight
years before. Egypt has been holding regular parliamentary elections
since 1976. The ones that it held in 2005, which resulted in sizeable
gains by the Muslim Brotherhood, were just another in that series (al-
though President Mubarak’s decision that year to hold direct presiden-
tial elections may have resulted in part from U.S. pressure). Lebanon’s
2005 elections were one more occurrence of elections in a series going
back some years.
In a small number of cases, outside actors play a very large role in
trying to help provoke or shape a democratic transition. One example is
where the United States and some European actors pursue highly asser-
tive policies of support for opposition movements in countries where a
strongman whom the West dislikes is attempting to get himself reelected.
Western support for the movements opposing President Slobodan
Miloševiæ in Serbia in the late 1990s and President Alyaksandr Lukash-
enka in Belarus in this decade are leading examples. Such cases, how-
ever, are rare. Moreover, although in these places the West is trying to
push the political envelope, it is not trying to make elections happen
where they are not planned, but rather is trying to influence the quality
or outcome of already planned elections.
The other types of cases of an international elephant in the room of a
country’s political transition are forcible interventions or conflict-re-
lated international engagements: where the United States and some of
Thomas Carothers 23

its allies intervene militarily to oust a dictator and attempt to establish

a democratic government in his place, as in Panama, Afghanistan, and
Iraq; or where the United Nations or another multilateral organization is
involved in helping to end a civil war and stays on to take a significant
part in the postconflict political-reconstruction process, as in Cambo-
dia, Mozambique, and Bosnia in the 1990s.
The outside actors in such cases usually push for elections, often
trying to rush the country to elections as part of their own exit strategy.
Even in these cases, however, the idea of holding elections as a central
component of the political-reconstruction process does not necessarily
originate from the outside—it is something that citizens and major po-
litical actors frequently expect and want. The case of Iraq is instructive
in this regard. After the ouster of Saddam Hussein, the U.S.-led Coali-
tion Provisional Authority (CPA) tried to put off the holding of elec-
tions. Yet despite the apparently overwhelmingly powerful U.S. role in
the country, homegrown demands by Iraqis that they be allowed to vote
forced the CPA to back down and agree to relatively early elections.
In short, even if it were possible to ensure rule-of-law development
and effective state-building under autocrats, sequencing would still be
deeply problematic. Whatever might be theoretically preferable regard-
ing paths of development, people in many parts of the world want to
attain political empowerment now, not at some indefinite point in the
future. Elections—even if held hurriedly and before all conditions are
ideal—have become the most visible embodiment of their aspiration to
self-rule and the urgency with which they feel it.
Even if one holds it morally defensible to let a generation endure
more decades of autocratic repression for the sake of possible future
benefits, persuading people to defer their ambition to vote in a free
election is most often not an option. The sequentialist argument is dogged
by a paradox: On the one hand, sequentialists advocate caution and
minimalism regarding democracy promotion on the grounds that it is
beyond the West’s power to shape or control inevitably unpredictable
processes of democratization. On the other hand, the core sequentialist
argument rests squarely on the belief that the West does have power to
shape political change in other societies precisely by convincing frus-
trated, mobilizing citizens to ignore—perhaps for decades—their own
heartfelt desire to take part openly and actively in politics.

Gradualism, Not Sequencing

Sequentialists’ ideas about democratic preconditions go too far. Their
raising of the issue, however, is helpful. It highlights the need to pay
more attention to the effects that a country’s underlying economic, so-
cial, and political conditions, structures, and historical legacies will
have on the chances that a democratic transition can succeed there.
24 Journal of Democracy

With striking breadth and speed, the third wave upended the lingering
conventional wisdom from the 1960s about democratic preconditions,
which stressed the need for a middle class and other attributes of eco-
nomic development. Democracy promoters in the 1980s and 1990s em-
braced the view that democracy could succeed anywhere, and could do
so quickly. They focused on the mode of transition as a key determinant
of success. In so doing, they tended to underestimate the complications
that different underlying conditions could present and to overestimate
the power of elections alone to produce fundamental political change.
As the third wave has aged, however, and many “transitional coun-
tries” have fallen short of early hopes, the importance of underlying
conditions and structures for democratic success has become increas-
ingly evident. A large array of such factors is relevant, but the record of
recent years indicates that five are of particular importance:
• Level of economic development: In general, the wealthier a country
is, the better will be its chances of consolidating a democratic transition.
• Concentration of sources of national wealth: Countries whose
national wealth comes mainly from highly concentrated sources (such
as oil or mineral deposits) tend to experience significant difficulties
with democratization.
• Identity-based divisions: Countries where the population is di-
vided along ethnic, religious, tribal, or clan lines often have a harder
time with democratization than more homogeneous societies.
• Historical experience with political pluralism: Countries with little
record of political pluralism almost always have a harder time with
democratization than those having such experience.
• Nondemocratic neighborhoods: Countries in regions or subregions
where most or all of the countries are nondemocratic usually struggle more
with democratization than do countries in more democratic neighborhoods.
These five factors should not be thought of as preconditions, a term
that causes considerable confusion among analysts. The factors are not
prior requirements, without which democratization cannot advance sig-
nificantly. Rather they are core facilitators or nonfacilitators. We should
understand them as making democratization harder or easier, and not
either certain or impossible. When several or all of them lean decisively
in a positive direction, a country has a much better chance of succeed-
ing with democratization than when they lean in the other direction. All
of them come in gradations; they are not black-or-white conditions.
Taken together they form a continuum of likelihood of democratic suc-
cess, not a bipolar division between countries that can be democratic
and countries that cannot.
We can see that these factors are not preconditions by thinking of the
counterexamples that can be named with relation to each. The ranks of
democracies, after all, include poor countries: not only India, that pe-
rennial favorite, but also Panama, Suriname, Botswana, Lesotho, and
Thomas Carothers 25

Senegal. At least one very successful democracy, Norway, has highly

concentrated sources of national wealth (North Sea oil). Many democra-
cies, including some longstanding ones such as Switzerland, Canada,
and Belgium, contain deeply entrenched identity-based divisions. A
lack of experience with political pluralism has unquestionably hurt
many third wave democratic efforts, but at least a few countries lacking
such experience (Mongolia, Georgia, and Ukraine), have made a par-
tially successful run at democracy. Nondemocratic neighborhoods are
problematic but not a complete bar, as evidenced by India in the first
few decades of its postcolonial existence, Costa Rica in the 1960s and
1970s, and Israel for decades.
Associated with the “no-preconditions” enthusiasm of democracy
promoters has been a dominant model of transition featuring a decisive
breakthrough in which the old regime collapses and the country moves
very quickly to open national elections, followed by longer-term pro-
cesses of state reform and civil society strengthening. Only a minority
of countries in the past several decades have closely adhered to this
model, however, and they have been (as in Central Europe) generally
well positioned in terms of facilitative factors. Elsewhere, attempted
transitions often have led to different outcomes, including hybrid poli-
ties and even outright reversions to autocracy. As a result, democracy
promoters are increasingly seeking alternative approaches for countries
that face complicating conditions.
One avenue of this search leads to a more gradualistic approach to
democratization and democracy promotion. Democratic gradualism is
different from sequencing. It does not entail putting off for decades or
indefinitely the core element of democratization—the development of
fair and open processes of political competition and choice. It involves
reaching for the core element now, but doing so in iterative and cumu-
lative ways rather than all at once. Gradualism can take different forms
depending on the context.
With respect to societies emerging from violent conflicts, for example,
a valuable debate has emerged among democracy promoters in recent
years about the dangers of “premature elections.” Advocates of moving
more gradually toward elections in such contexts are not arguing for
indefinite delay pending the creation of deep structural conditions. The
idea, rather, is to put off elections for at most several years to allow in-
depth negotiations between contending political groups, so that the main
political forces can get used to dealing with one another peacefully and
agree on the rules of the game before potentially divisive elections are
held.15 As the case of South Africa’s unhurried approach to elections in
the early 1990s demonstrates, citizens eager to have their chance to vote
will be willing to wait if the delay in voting is clearly limited and if it is
tied to an inclusive process of building a national consensus.
Authoritarian governments that show some genuine interest in re-
26 Journal of Democracy

form, usually as part of a quest for economic development, often claim

that they are gradualist democratizers. Such claims are usually false.
The brand of pseudodemocratic gradualism that authoritarians tout typi-
cally overstates the political impact that economic reforms will likely
exert, and limits political reforms to very minor steps. In places such as
China and parts of the Middle East, democracy promoters are increas-
ingly seeking a middle path between unrealistic calls for sudden politi-
cal openings and excessive praise for minor reforms. This new approach
highlights the need for small but significant steps that create space and
mechanisms for true political competition and point the way to an even-
tual end of the rulers’ monopoly on power. Such steps might include
allowing independent civil society organizations that engage in politi-
cally related advocacy; permitting the establishment of political asso-
ciations or other types of protoparties independent of the ruling party;
holding local or provincial elections in which not just independent
candidates but candidates representing these political associations can
compete; and tolerating a modicum of open public space in which truly
independent media can operate and criticism of the rulers can find voice.
In semiauthoritarian countries, powerholders often abuse the concept
of gradualism by claiming that their partial political liberalization is a
necessary halfway house on the way to democracy, when in fact it is a
means to avoid altogether the kinds of far-reaching political changes that
would threaten semiauthoritarian rule. Democracy promoters who are
determined to avoid becoming enablers of such false gradualism should
focus on strategies aimed at rendering more meaningful such political
competition as does exist. A list of such strategies would feature the
creation of a genuinely independent system of electoral administration;
the legalization of excluded political parties; the reinforcement of those
institutions, such as legislatures, in which a degree of real pluralism is
already extant; the securing of permission for independent citizens’ groups
to monitor and criticize elections and other key aspects of the political
process; and the diminution via constitutional reform of those zones of
political power that remain exempt from the competitive process.16
Gradualism is not a magic bullet. It can be easily misused by those
who are insincere about democratization. It gives democracy promoters
no great new forms of influence that they can use to enhance their mostly
modest role. It does not mean that democracy will never go awry or disap-
point. And it is not universally applicable—there will still be places
where pushing for rapid movement toward elections makes sense, and
others where no signs of movement toward political pluralism are detect-
able at all. But gradualism is a vital additional approach that is being
added, and should be further developed as part of the broader evolution
of democracy promotion away from a standard transitional template. The
difference between gradualism and sequentialism may at first appear subtle
or merely semantic. In fact, however, it is fundamental. Gradualism is a
Thomas Carothers 27

different way of engaging in or pushing for democratization now, in

service of a belief in democratic possibility. Sequentialism is a method
for putting off democratization until some uncertain future time, rooted
in skepticism about democracy’s value and chances.

NOTES
1. Fareed Zakaria, “The Rise of Illiberal Democracy,” Foreign Affairs 76 (No-
vember–December 1997): 22–43.

2. Edward D. Mansfield and Jack Snyder, “Democratization and the Danger of

War,” International Security 20 (Summer 1995): 5–38.

3. Amy Chua, “Markets, Democracy, and Ethnicity: Toward a New Paradigm

for Law and Development,” Yale Law Journal 108 (October 1998): 1–108.

4. Edward D. Mansfield and Jack Snyder, Electing to Fight: Why Emerging

Democracies Go to War (Cambridge: MIT Press, 2005), 273.

5. Fareed Zakaria, The Future of Freedom: Illiberal Democracy at Home and

Abroad (New York: Norton, 2003), 15.

6. See for example, Robert Kaplan, “Was Democracy Just a Moment?” Atlantic
Monthly, December 1997, 55–80.

7. Zakaria, The Future of Freedom; Mansfield and Snyder, Electing to Fight;

and Amy Chua, World on Fire: How Exporting Free Market Democracy Breeds
Ethnic Hatred and Global Instability (New York: Doubleday, 2003).

8. Matthew Stephenson, “A Trojan Horse in China?” in Thomas Carothers, ed.,

Promoting the Rule of Law Abroad: In Search of Knowledge (Washington, D.C.:
Carnegie Endowment, 2006), ch. 8.

9. U.S. Department of State, 2005 Country Reports on Human Rights Practices

(Washington, D.C.: U.S. Department of State, 2006).

10. Rebecca Bill Chavez, “A Construction of the Rule of Law in Argentina: A

Tale of Two Provinces,” Comparative Politics 35 (July 2003): 417–37.

11. Marc F. Plattner, “From Liberalism to Illiberal Democracy,” Journal of

Democracy 10 (July 1999): 121–34.

12. Thomas Carothers, “The End of the Transition Paradigm,” Journal of De-
mocracy 13 (January 2002): 5–21.

13. Joel S. Migdal, Strong Societies and Weak States: State-Society Relations
and State Capabilities in the Third World (Princeton: Princeton University Press,
1988).

14. Mansfield and Snyder, Electing to Fight, 278.

15. See Krishna Kumar and Marina Ottaway, “General Conclusions and Priori-
ties for Policy Research,” in Krishna Kumar, ed., Postconflict Elections, Democra-
tization and International Assistance (Boulder, Colo.: Lynne Reinner, 1998), ch.
13.

16. Marina Ottaway, Democracy Challenged: The Rise of Semi-Authoritarian-

ism (Washington, D.C.: Carnegie Endowment, 2003), ch. 7.