THE END OF THE
TRANSITION PARADIGM
Thomas Carothers
In the last quarter of the twentieth century, trends in seven different
regions converged to change the political landscape of the world: 1) the
fall of right-wing authoritarian regimes in Southern Europe in the mid-
1970s; 2) the replacement of military dictatorships by elected civilian
governments across Latin America from the late 1970s through the late
1980s; 3) the decline of authoritarian rule in parts of East and South
Asia starting in the mid-1980s; 4) the collapse of communist regimes in
Eastern Europe at the end of the 1980s; 5) the breakup of the Soviet
Union and the establishment of 15 post-Soviet republics in 1991; 6) the
decline of one-party regimes in many parts of sub-Saharan Africa in the
first half of the 1990s; and 7) a weak but recognizable liberalizing trend
in some Middle Eastern countries in the 1990s.
The causes, shape, and pace of these different trends varied con-
siderably. But they shared a dominant characteristic—simultaneous
movement in at least several countries in each region away from dic-
tatorial rule toward more liberal and often more democratic governance.
And though differing in many ways, these trends influenced and to some
extent built on one another. As a result, they were considered by many
observers, especially in the West, as component parts of a larger whole,
a global democratic trend that thanks to Samuel Huntington has widely
come to be known as the “third wave” of democracy.
1
This striking tide of political change was seized upon with enthusiasm
by the U.S. government and the broader U.S. foreign policy community.
As early as the mid-1980s, President Ronald Reagan, Secretary of State
Thomas Carothers is vice president for studies at the Carnegie Endow-
ment for International Peace in Washington, D.C. He is the author of
many works on democracy promotion, including Aiding Democracy
Abroad: The Learning Curve (1999), and is the coeditor with Marina
Ottaway of Funding Virtue: Civil Society Aid and Democracy Promotion
(2000).
Carothers, Thomas. “The End of the Transition Paradigm.” Journal of Democracy 13:1 (2002).
© The Johns Hopkins University Press and the National Endowment for Democracy. Reprinted
with the permission of The Johns Hopkins University Press.
Journal of Democracy6
George Shultz, and other high-level U.S. officials were referring regularly
to “the worldwide democratic revolution.” During the 1980s, an active
array of governmental, quasi-governmental, and nongovernmental
organizations devoted to promoting democracy abroad sprang into being.
This new democracy-promotion community had a pressing need for an
analytic framework to conceptualize and respond to the ongoing political
events. Confronted with the initial parts of the third wave—democ-
ratization in Southern Europe, Latin America, and a few countries in
Asia (especially the Philippines)—the U.S. democracy community rapid-
ly embraced an analytic model of democratic transition. It was derived
principally from their own interpretation of the patterns of democratic
change taking place, but also to a lesser extent from the early works of
the emergent academic field of “transitology,” above all the seminal
work of Guillermo O’Donnell and Philippe Schmitter.
2
As the third wave spread to Eastern Europe, the Soviet Union, sub-
Saharan Africa, and elsewhere in the 1990s, democracy promoters
extended this model as a universal paradigm for understanding democ-
ratization. It became ubiquitous in U.S. policy circles as a way of talking
about, thinking about, and designing interventions in processes of
political change around the world. And it stayed remarkably constant
despite many variations in those patterns of political change and a stream
of increasingly diverse scholarly views about the course and nature of
democratic transitions.
3
The transition paradigm has been somewhat useful during a time of
momentous and often surprising political upheaval in the world. But it
is increasingly clear that reality is no longer conforming to the model.
Many countries that policy makers and aid practitioners persist in calling
“transitional” are not in transition to democracy, and of the democratic
transitions that are under way, more than a few are not following the
model. Sticking with the paradigm beyond its useful life is retarding
evolution in the field of democratic assistance and is leading policy
makers astray in other ways. It is time to recognize that the transition
paradigm has outlived its usefulness and to look for a better lens.
Core Assumptions
Five core assumptions define the transition paradigm. The first, which
is an umbrella for all the others, is that any country moving away from
dictatorial rule can be considered a country in transition toward
democracy. Especially in the first half of the 1990s, when political change
accelerated in many regions, numerous policy makers and aid prac-
titioners reflexively labeled any formerly authoritarian country that was
attempting some political liberalization as a “transitional country.” The
set of “transitional countries” swelled dramatically, and nearly 100
countries (approximately 20 in Latin America, 25 in Eastern Europe
Thomas Carothers 7
and the former Soviet Union, 30 in sub-Saharan Africa, 10 in Asia, and
5 in the Middle East) were thrown into the conceptual pot of the transition
paradigm. Once so labeled, their political life was automatically analyzed
in terms of their movement toward or away from democracy, and they
were held up to the implicit expectations of the paradigm, as detailed
below. To cite just one especially astonishing example, the U.S. Agency
for International Development (USAID) continues to describe the
Democratic Republic of Congo (Kinshasa), a strife-wracked country
undergoing a turgid, often opaque, and rarely very democratic process
of political change, as a country in “transition to a democratic, free market
society.”
4
The second assumption is that democratization tends to unfold in a
set sequence of stages. First there occurs the opening, a period of
democratic ferment and political liberalization in which cracks appear
in the ruling dictatorial regime, with the most prominent fault line being
that between hardliners and softliners. There follows the breakthrough—
the collapse of the regime and the rapid emergence of a new, democratic
system, with the coming to power of a new government through national
elections and the establishment of a democratic institutional structure,
often through the promulgation of a new constitution. After the transition
comes consolidation, a slow but purposeful process in which democratic
forms are transformed into democratic substance through the reform of
state institutions, the regularization of elections, the strengthening of
civil society, and the overall habituation of the society to the new
democratic “rules of the game.”
5
Democracy activists admit that it is not inevitable that transitional
countries will move steadily on this assumed path from opening and
breakthrough to consolidation. Transitional countries, they say, can and
do go backward or stagnate as well as move forward along the path. Yet
even the deviations from the assumed sequence that they are willing to
acknowledge are defined in terms of the path itself. The options are all
cast in terms of the speed and direction with which countries move on
the path, not in terms of movement that does not conform with the path
at all. And at least in the peak years of the third wave, many democracy
enthusiasts clearly believed that, while the success of the dozens of new
transitions was not assured, democratization was in some important sense
a natural process, one that was likely to flourish once the initial break-
through occurred. No small amount of democratic teleology is implicit
in the transition paradigm, no matter how much its adherents have denied
it.
6
Related to the idea of a core sequence of democratization is the third
assumption—the belief in the determinative importance of elections.
Democracy promoters have not been guilty—as critics often charge—
of believing that elections equal democracy. For years they have
advocated and pursued a much broader range of assistance programs
Journal of Democracy8
than just elections-focused efforts. Nevertheless, they have tended to
hold very high expectations for what the establishment of regular,
genuine elections will do for democratization. Not only will elections
give new postdictatorial governments democratic legitimacy, they
believe, but the elections will serve to broaden and deepen political
participation and the democratic accountability of the state to its citizens.
In other words, it has been assumed that in attempted transitions to
democracy, elections will be not just a foundation stone but a key
generator over time of further democratic reforms.
A fourth assumption is that the underlying conditions in transitional
countries—their economic level, political history, institutional legacies,
ethnic make-up, sociocultural traditions, or other “structural” features—
will not be major factors in either the onset or the outcome of the
transition process. A remarkable characteristic of the early period of the
third wave was that democracy seemed to be breaking out in the most
unlikely and unexpected places, whether Mongolia, Albania, or
Mauritania. All that seemed to be necessary for democratization was a
decision by a country’s political elites to move toward democracy and
an ability on the part of those elites to fend off the contrary actions of
remaining antidemocratic forces.
The dynamism and remarkable scope of the third wave buried old,
deterministic, and often culturally noxious assumptions about democracy,
such as that only countries with an American-style middle class or a
heritage of Protestant individualism could become democratic. For policy
makers and aid practitioners this new outlook was a break from the long-
standing Cold War mindset that most countries in the developing world
were “not ready for democracy,” a mindset that dovetailed with U.S.
policies of propping up anticommunist dictators around the world. Some
of the early works in transitology also reflected the “no preconditions”
view of democratization, a shift within the academic literature that had
begun in 1970 with Dankwart Rustow’s seminal article, “Transitions to
Democracy: Toward a Dynamic Model.”
7
For both the scholarly and
policy communities, the new “no preconditions” outlook was a
gratifyingly optimistic, even liberating view that translated easily across
borders as the encouraging message that, when it comes to democracy,
“anyone can do it.”
Fifth, the transition paradigm rests on the assumption that the
democratic transitions making up the third wave are being built on
coherent, functioning states. The process of democratization is assumed
to include some redesign of state institutions—such as the creation of
new electoral institutions, parliamentary reform, and judicial reform—
but as a modification of already functioning states.
8
As they arrived at
their frameworks for understanding democratization, democracy aid
practitioners did not give significant attention to the challenge of a society
trying to democratize while it is grappling with the reality of building a
Thomas Carothers 9
state from scratch or coping with an existent but largely nonfunctional
state. This did not appear to be an issue in Southern Europe or Latin
America, the two regions that served as the experiential basis for the
formation of the transition paradigm. To the extent that democracy
promoters did consider the possibility of state-building as part of the
transition process, they assumed that democracy-building and state-
building would be mutually reinforcing endeavors or even two sides of
the same coin.
Into the Gray Zone
We turn then from the underlying assumptions of the paradigm to the
record of experience. Efforts to assess the progress of the third wave are
sometimes rejected as premature. Democracy is not built in a day,
democracy activists assert, and it is too early to reach judgments about
the results of the dozens of democratic transitions launched in the last
two decades. Although it is certainly true that the current political
situations of the “transitional countries” are not set in stone, enough
time has elapsed to shed significant light on how the transition paradigm
is holding up.
Of the nearly 100 countries considered as “transitional” in recent
years, only a relatively small number—probably fewer than 20—are
clearly en route to becoming successful, well-functioning democracies
or at least have made some democratic progress and still enjoy a positive
dynamic of democratization.
9
The leaders of the group are found
primarily in Central Europe and the Baltic region—Poland, Hungary,
the Czech Republic, Estonia, and Slovenia—though there are a few in
South America and East Asia, notably Chile, Uruguay, and Taiwan.
Those that have made somewhat less progress but appear to be still
advancing include Slovakia, Romania, Bulgaria, Mexico, Brazil, Ghana,
the Philippines, and South Korea.
By far the majority of third-wave countries have not achieved
relatively well-functioning democracy or do not seem to be deepening
or advancing whatever democratic progress they have made. In a small
number of countries, initial political openings have clearly failed and
authoritarian regimes have resolidified, as in Uzbekistan, Turkmenistan,
Belarus, and Togo. Most of the “transitional countries,” however, are
neither dictatorial nor clearly headed toward democracy. They have
entered a political gray zone.
10
They have some attributes of democratic
political life, including at least limited political space for opposition
parties and independent civil society, as well as regular elections and
democratic constitutions. Yet they suffer from serious democratic
deficits, often including poor representation of citizens’ interests, low
levels of political participation beyond voting, frequent abuse of the
law by government officials, elections of uncertain legitimacy, very low
Journal of Democracy10
levels of public confidence in state institutions, and persistently poor
institutional performance by the state.
As the number of countries falling in between outright dictatorship
and well-established liberal democracy has swollen, political analysts
have proffered an array of “qualified democracy” terms to characterize
them, including semi-democracy, formal democracy, electoral
democracy, façade democracy, pseudo-democracy, weak democracy,
partial democracy, illiberal democracy, and virtual democracy.
11
Some
of these terms, such as “façade democracy” and “pseudo-democracy,”
apply only to a fairly specific subset of gray-zone cases. Other terms,
such as “weak democracy” and “partial democracy,” are intended to have
much broader applicability. Useful though these terms can be, especially
when rooted in probing analysis such as O’Donnell’s work on “delegative
democracy,” they share a significant liability: By describing countries
in the gray zone as types of democracies, analysts are in effect trying to
apply the transition paradigm to the very countries whose political
evolution is calling that paradigm into question.
12
Most of the “qualified
democracy” terms are used to characterize countries as being stuck
somewhere on the assumed democratization sequence, usually at the start
of the consolidation phase.
The diversity of political patterns within the gray zone is vast. Many
possible subtypes or subcategories could potentially be posited, and much
work remains to be done to assess the nature of gray-zone politics. As a
first analytic step, two broad political syndromes can be seen to be
common in the gray zone. They are not rigidly delineated political-system
types but rather political patterns that have become regular and somewhat
entrenched. Though they have some characteristics in common, they
differ in crucial ways and basically are mutually exclusive.
The first syndrome is feckless pluralism. Countries whose political
life is marked by feckless pluralism tend to have significant amounts of
political freedom, regular elections, and alternation of power between
genuinely different political groupings. Despite these positive features,
however, democracy remains shallow and troubled. Political partici-
pation, though broad at election time, extends little beyond voting.
Political elites from all the major parties or groupings are widely per-
ceived as corrupt, self-interested, and ineffective. The alternation of
power seems only to trade the country’s problems back and forth from
one hapless side to the other. Political elites from all the major parties
are widely perceived as corrupt, self-interested, dishonest, and not serious
about working for their country. The public is seriously disaffected from
politics, and while it may still cling to a belief in the ideal of democracy,
it is extremely unhappy about the political life of the country. Overall,
politics is widely seen as a stale, corrupt, elite-dominated domain that
delivers little good to the country and commands equally little respect.
And the state remains persistently weak. Economic policy is often poorly
Thomas Carothers 11
conceived and executed, and economic performance is frequently bad
or even calamitous. Social and political reforms are similarly tenuous,
and successive governments are unable to make headway on most of
the major problems facing the country, from crime and corruption to
health, education, and public welfare generally.
Feckless pluralism is most common in Latin America, a region where
most countries entered their attempted democratic transitions with
diverse political parties already in place yet also with a deep legacy of
persistently poor performance of state institutions. Nicaragua, Ecuador,
Guatemala, Panama, Honduras, and Bolivia all fall into this category,
as did Venezuela in the decade prior to the election of Hugo Chávez.
Argentina and Brazil hover uneasily at its edge. In the postcommunist
world, Moldova, Bosnia, Albania, and Ukraine have at least some
significant signs of the syndrome, with Romania and Bulgaria teetering
on its edge. Nepal is a clear example in Asia; Bangladesh, Mongolia,
and Thailand may also qualify. In sub-Saharan Africa, a few states, such
as Madagascar, Guinea-Bissau, and Sierra Leone, may be cases of
feckless pluralism, though alternation of power remains rare generally
in that region.
There are many variations of feckless pluralism. In some cases, the
parties that alternate power between them are divided by paralyzing
acrimony and devote their time out of power to preventing the other
party from accomplishing anything at all, as in Bangladesh. In other
cases, the main competing groups end up colluding, formally or
informally, rendering the alternation of power unhelpful in a different
manner, as happened in Nicaragua in the late 1990s. In some countries
afflicted with feckless pluralism, the political competition is between
deeply entrenched parties that essentially operate as patronage networks
and seem never to renovate themselves, as in Argentina or Nepal. In
others, the alternation of power occurs between constantly shifting
political groupings, short-lived parties led by charismatic individuals
or temporary alliances in search of a political identity, as in Guatemala
or Ukraine. These varied cases nonetheless share a common condition
that seems at the root of feckless pluralism—the whole class of political
elites, though plural and competitive, are profoundly cut off from the
citizenry, rendering political life an ultimately hollow, unproductive
exercise.
Dominant-Power Politics
The most common other political syndrome in the gray zone is
dominant-power politics. Countries with this syndrome have limited but
still real political space, some political contestation by opposition groups,
and at least most of the basic institutional forms of democracy. Yet one
political grouping—whether it is a movement, a party, an extended
Journal of Democracy12
family, or a single leader—dominates the system in such a way that
there appears to be little prospect of alternation of power in the
foreseeable future.
Unlike in countries beset with feckless pluralism, a key political
problem in dominant-power countries is the blurring of the line between
the state and the ruling party (or ruling political forces). The state’s
main assets—that is to say, the state as a source of money, jobs, public
information (via state media), and police power—are gradually put in
the direct service of the ruling party. Whereas in feckless pluralism
judiciaries are often somewhat independent, the judiciary in dominant-
power countries is typically cowed, as part of the one-sided grip on
power. And while elections in feckless-pluralist countries are often quite
free and fair, the typical pattern in dominant-power countries is one of
dubious but not outright fraudulent elections in which the ruling group
tries to put on a good-enough electoral show to gain the approval of the
international community while quietly tilting the electoral playing field
far enough in its own favor to ensure victory.
As in feckless-pluralist systems, the citizens of dominant-power
systems tend to be disaffected from politics and cut off from significant
political participation beyond voting. Since there is no alternation of
power, however, they are less apt to evince the “a pox on all your houses”
political outlook pervasive in feckless-pluralist systems. Yet those
opposition political parties that do exist generally are hard put to gain
much public credibility due to their perennial status as outsiders to the
main halls of power. Whatever energies and hopes for effective
opposition to the regime remain often reside in civil society groups,
usually a loose collection of advocacy NGOs and independent media
(often funded by Western donors) that skirmish with the government on
human rights, the environment, corruption, and other issues of public
interest.
The state tends to be as weak and poorly performing in dominant-
power countries as in feckless-pluralist countries, though the problem
is often a bureaucracy decaying under the stagnancy of de facto one-
party rule rather than the disorganized, unstable nature of state
management (such as the constant turnover of ministers) typical of
feckless pluralism. The long hold on power by one political group usually
produces large-scale corruption and crony capitalism. Due to the
existence of some political openness in these systems, the leaders do
often feel some pressure from the public about corruption and other
abuses of state power. They even may periodically declare their intention
to root out corruption and strengthen the rule of law. But their deep-
seated intolerance for anything more than limited opposition and the
basic political configuration over which they preside breed the very
problems they publicly commit themselves to tackling.
Dominant-power systems are prevalent in three regions. In sub-
Thomas Carothers 13
Saharan Africa, the widely hailed wave of democratization that washed
over the region in the early 1990s has ended up producing many
dominant-power systems. In some cases, one-party states liberalized yet
ended up permitting only very limited processes of political opening, as
in Cameroon, Burkina Faso, Equatorial Guinea, Tanzania, Gabon, Kenya,
and Mauritania. In a few cases, old regimes were defeated or collapsed,
yet the new regimes have ended up in dominant-party structures, as in
Zambia in the 1990s, or the forces previously shunted aside have
reclaimed power, as in Congo (Brazzaville).
Dominant-power systems are found in the former Soviet Union as
well. Armenia, Azerbaijan, Georgia, Kyrgyzstan, and Kazakhstan fall
in this category. The other Central Asian republics and Belarus are better
understood as out-and-out authoritarian systems. The liberalization trend
that arose in the Middle East in the mid-1980s and has unfolded in fits
and starts ever since has moved some countries out of the authoritarian
camp into the dominant-power category. These include Morocco, Jordan,
Algeria, Egypt, Iran, and Yemen. Dominant-power systems are scarce
outside of these three regions. In Asia, Malaysia and Cambodia count
as examples. In Latin America, Paraguay may be one case, and Venezuela
is likely headed toward becoming a second.
Dominant-power systems vary in their degree of freedom and their
political direction. Some have very limited political space and are close
to being dictatorships. Others allow much more freedom, albeit still with
limits. A few “transitional countries,” including the important cases of
South Africa and Russia, fall just to the side of this syndrome. They
have a fair amount of political freedom and have held competitive
elections of some legitimacy (though sharp debate on that issue exists
with regard to Russia). Yet they are ruled by political forces that appear
to have a long-term hold on power (if one considers the shift from Yeltsin
to Putin more as a political transfer than an alternation of power), and it
is hard to imagine any of the existing opposition parties coming to power
for many years to come. If they maintain real political freedom and open
competition for power, they may join the ranks of cases, such as Italy
and Japan (prior to the 1990s) and Botswana, of longtime democratic
rule by one party. Yet due to the tenuousness of their new democratic
institutions, they face the danger of slippage toward the dominant-power
syndrome.
As political syndromes, both feckless pluralism and dominant-power
politics have some stability. Once in them, countries do not move out of
them easily. Feckless pluralism achieves its own dysfunctional
equilibrium—the passing of power back and forth between competing
elites who are largely isolated from the citizenry but willing to play by
widely accepted rules. Dominant-power politics also often achieves a
kind of stasis, with the ruling group able to keep political opposition on
the ropes while permitting enough political openness to alleviate pressure
Journal of Democracy14
from the public. They are by no means permanent political configu-
rations; no political configuration lasts forever. Countries can and do
move out of them—either from one to the other or out of either toward
liberal democracy or dictatorship. For a time in the 1990s, Ukraine
seemed stuck in dominant-power politics but may be shifting to some-
thing more like feckless pluralism. Senegal was previously a clear case
of dominant-power politics but, with the opposition victory in the 2000
elections, may be moving toward either liberal democracy or feckless
pluralism.
Although many countries in the gray zone have ended up as examples
of either feckless pluralism or dominant-power politics, not all have. A
small number of “transitional countries” have moved away from
authoritarian rule only in the last several years, and their political trajec-
tory is as yet unclear. Indonesia, Nigeria, Serbia, and Croatia are four
prominent examples of this type. Some countries that experienced
political openings in the 1980s or 1990s have been so wracked by civil
conflict that their political systems are too unstable or incoherent to pin
down easily, though they are definitely not on a path of democratization.
The Democratic Republic of Congo, Liberia, Sierra Leone, and Somalia
all represent this situation.
The Crash of Assumptions
Taken together, the political trajectories of most third-wave countries
call into serious doubt the transition paradigm. This is apparent if we
revisit the major assumptions underlying the paradigm in light of the
above analysis.
First, the almost automatic assumption of democracy promoters during
the peak years of the third wave that any country moving away from
dictatorship was “in transition to democracy” has often been inaccurate
and misleading. Some of those countries have hardly democratized at
all. Many have taken on a smattering of democratic features but show
few signs of democratizing much further and are certainly not following
any predictable democratization script. The most common political
patterns to date among the “transitional countries”—feckless pluralism
and dominant-power politics—include elements of democracy but should
be understood as alternative directions, not way stations to liberal
democracy. The persistence in official U.S. democracy-promotion circles
of using transitional language to characterize countries that in no way
conform to any democratization paradigm borders in some case on the
surreal—including not just the case of Congo cited above but many
others, such as Moldova (“Moldova’s democratic transition continues
to progress steadily”), Zambia (“Zambia is . . . moving steadily toward
. . . the creation of a viable multiparty democracy”), Cambodia (“policy
successes in Cambodia towards democracy and improved governance
Thomas Carothers 15
within the past 18 months are numerous”), and Guinea (“Guinea has
made significant strides toward building a democratic society”).
13
The
continued use of the transition paradigm constitutes a dangerous habit
of trying to impose a simplistic and often incorrect conceptual order on
an empirical tableau of considerable complexity.
Second, not only is the general label and concept of “transitional
country” unhelpful, but the assumed sequence of stages of democrati-
zation is defied by the record of experience. Some of the most encour-
aging cases of democratization in recent years—such as Taiwan, South
Korea, and Mexico—did not go through the paradigmatic process of
democratic breakthrough followed rapidly by national elections and a
new democratic institutional framework. Their political evolutions were
defined by an almost opposite phenomenon—extremely gradual, incre-
mental processes of liberalization with an organized political opposition
(not softliners in the regime) pushing for change across successive
elections and finally winning. And in many of the countries that did go
through some version of what appeared to be a democratic breakthrough,
the assumed sequence of changes—first settling constitutive issues then
working through second-order reforms—has not held. Constitutive issues
have reemerged at unpredictable times, upending what are supposed to
be later stages of transition, as in the recent political crises in Ecuador,
the Central African Republic, and Chad.
Moreover, the various assumed component processes of consoli-
dation—political party development, civil society strengthening, judicial
reform, and media development—almost never conform to the techno-
cratic ideal of rational sequences on which the indicator frameworks
and strategic objectives of democracy promoters are built. Instead they
are chaotic processes of change that go backwards and sideways as much
as forward, and do not do so in any regular manner.
The third assumption of the transition paradigm—the notion that
achieving regular, genuine elections will not only confer democratic
legitimacy on new governments but continuously deepen political
participation and democratic accountability—has often come up short.
In many “transitional countries,” reasonably regular, genuine elections
are held but political participation beyond voting remains shallow and
governmental accountability is weak. The wide gulf between political
elites and citizens in many of these countries turns out to be rooted in
structural conditions, such as the concentration of wealth or certain
sociocultural traditions, that elections themselves do not overcome. It is
also striking how often electoral competition does little to stimulate the
renovation or development of political parties in many gray-zone
countries. Such profound pathologies as highly personalistic parties,
transient and shifting parties, or stagnant patronage-based politics appear
to be able to coexist for sustained periods with at least somewhat legiti-
mate processes of political pluralism and competition.
Journal of Democracy16
These disappointments certainly do not mean that elections are
pointless in such countries or that the international community should
not continue to push for free and fair elections. But greatly reduced
expectations are in order as to what elections will accomplish as genera-
tors of deep-reaching democratic change. Nepal is a telling example in
this regard. Since 1990, Nepal has held many multiparty elections and
experienced frequent alternation of power. Yet the Nepalese public
remains highly disaffected from the political system and there is little
real sense of democratic accountability.
Fourth, ever since “preconditions for democracy” were enthusi-
astically banished in the heady early days of the third wave, a contrary
reality—the fact that various structural conditions clearly weigh heavily
in shaping political outcomes—has been working its way back in. Look-
ing at the more successful recent cases of democratization, for example,
which tend to be found in Central Europe, the Southern Cone, or East
Asia, it is clear that relative economic wealth, as well as past experience
with political pluralism, contributes to the chances for democratic suc-
cess. And looking comparatively within regions, whether in the former
communist world or sub-Saharan Africa, it is evident that the specific
institutional legacies from predecessor regimes strongly affect the
outcomes of attempted transitions.
During the 1990s, a number of scholars began challenging the “no
preconditions” line, with analyses of the roles that economic wealth,
institutional legacies, social class, and other structural factors play in
attempted democratic transitions.
14
Yet it has been hard for the democracy-
promotion community to take this work on board. Democracy promo-
ters are strongly wedded to their focus on political processes and
institutions. They have been concerned that trying to blend that focus
with economic or sociocultural perspectives might lead to the dilution or
reduction of democracy assistance. And having set up as organizations
with an exclusively political perspective, it is hard for democracy-
promotion groups to include other kinds of expertise or approaches.
Fifth, state-building has been a much larger and more problematic
issue than originally envisaged in the transition paradigm. Contrary to
the early assumptions of democracy-aid practitioners, many third-wave
countries have faced fundamental state-building challenges. Approxi-
mately 20 countries in the former Soviet Union and former Yugoslavia
have had to build national state institutions where none existed before.
Throughout much of sub-Saharan Africa, the liberalizing political wave
of the 1990s ran squarely into the sobering reality of devastatingly weak
states. In many parts of Latin America, the Middle East, and Asia,
political change was carried out in the context of stable state structures,
but the erratic performance of those states complicated every step.
Where state-building from scratch had to be carried out, the core
impulses and interests of powerholders—such as locking in access to
Thomas Carothers 17
power and resources as quickly as possible—ran directly contrary to
what democracy-building would have required. In countries with existing
but extremely weak states, the democracy-building efforts funded by
donors usually neglected the issue of state-building. With their frequent
emphasis on diffusing power and weakening the relative power of the
executive branch—by strengthening the legislative and judicial branches
of government, encouraging decentralization, and building civil
society—they were more about the redistribution of state power than
about state-building. The programs that democracy promoters have
directed at governance have tended to be minor technocratic efforts,
such as training ministerial staff or aiding cabinet offices, rather than
major efforts at bolstering state capacity.
Letting Go
It is time for the democracy-promotion community to discard the
transition paradigm. Analyzing the record of experience in the many
countries that democracy activists have been labeling “transitional
countries,” it is evident that it is no longer appropriate to assume:
• that most of these countries are actually in a transition to democracy;
• that countries moving away from authoritarianism tend to follow a
three-part process of democratization consisting of opening, break-
through, and consolidation;
• that the establishment of regular, genuine elections will not only
give new governments democratic legitimacy but foster a longer term
deepening of democratic participation and accountability;
• that a country’s chances for successfully democratizing depend
primarily on the political intentions and actions of its political elites
without significant influence from underlying economic, social, and
institutional conditions and legacies;
• that state-building is a secondary challenge to democracy-building
and largely compatible with it.
It is hard to let go of the transitional paradigm, both for the conceptual
order and for the hopeful vision it provides. Giving it up constitutes a
major break, but not a total one. It does not mean denying that important
democratic reforms have occurred in many countries in the past two
decades. It does not mean that countries in the gray zone are doomed
never to achieve well-functioning liberal democracy. It does not mean
that free and fair elections in “transitional countries” are futile or not
worth supporting. It does not mean that the United States and other
international actors should abandon efforts to promote democracy in
the world (if anything, it implies that, given how difficult democratization
is, efforts to promote it should be redoubled).
It does mean, however, that democracy promoters should approach
their work with some very different assumptions. They should start by
Journal of Democracy18
assuming that what is often thought of as an uneasy, precarious middle
ground between full-fledged democracy and outright dictatorship is
actually the most common political condition today of countries in the
developing world and the postcom-
munist world. It is not an exceptional
category to be defined only in terms
of its not being one thing or the
other; it is a state of normality for
many societies, for better or worse.
The seemingly continual surprise and
disappointment that Western political
analysts express over the very fre-
quent falling short of democracy in
“transitional countries” should be
replaced with realistic expectations
about the likely patterns of political
life in these countries.
Aid practitioners and policy makers
looking at politics in a country that
has recently moved away from
authoritarianism should not start by asking, “How is its democratic
transition going?” They should instead formulate a more open-ended
query, “What is happening politically?” Insisting on the former approach
leads to optimistic assumptions that often shunt the analysis down a
blind alley. To take one example, during the 1990s, Western policy
makers habitually analyzed Georgia’s post-1991 political evolution as
a democratic transition, highlighting the many formal achievements, and
holding up a basically positive image of the country. Then suddenly, at
the end of the decade, the essential hollowness of Georgia’s “democratic
transition” became too apparent to ignore, and Georgia is now suddenly
talked about as a country in serious risk of state failure or deep socio-
political crisis.
15
A whole generation of democracy aid is based on the transition
paradigm, above all the typical emphasis on an institutional “checklist”
as a basis for creating programs, and the creation of nearly standard
portfolios of aid projects consisting of the same diffuse set of efforts all
over—some judicial reform, parliamentary strengthening, civil society
assistance, media work, political party development, civic education,
and electoral programs. Much of the democracy aid based on this
paradigm is exhausted. Where the paradigm fits well—in the small
number of clearly successful transitions—the aid is not much needed.
Where democracy aid is needed most, in many of the gray-zone countries,
the paradigm fits poorly.
Democracy promoters need to focus in on the key political patterns
of each country in which they intervene, rather than trying to do a little
What is often thought of
as an uneasy, precarious
middle ground between
full-fledged democracy
and outright dictatorship
is actually the most
common political
condition today of
countries in the
developing world and the
postcommunist world.
Thomas Carothers 19
of everything according to a template of ideal institutional forms. Where
feckless pluralism reigns, this means giving concentrated attention to
two interrelated issues: how to improve the variety and quality of the
main political actors in the society and how to begin to bridge the gulf
between the citizenry and the formal political system. Much greater
attention to political party development should be a major part of the
response, with special attention to encouraging new entrants into the
political party scene, changing the rules and incentive systems that shape
the current party structures, and fostering strong connections between
parties and civil society groups (rather than encouraging civil society
groups to stay away from partisan politics).
In dominant-power systems, democracy promoters should devote
significant attention to the challenge of helping to encourage the growth
of alternative centers of power. Merely helping finance the proliferation
of nongovernmental organizations is an inadequate approach to this
challenge. Again, political party development must be a top agenda item,
especially through measures aimed at changing the way political parties
are financed. It should include efforts to examine how the over-
concentration of economic power (a standard feature of dominant-power
systems) can be reduced as well as measures that call attention to and
work against the blurring of the line between the ruling party and the
state.
In other types of gray-zone countries, democracy promoters will need
to settle on other approaches. The message for all gray-zone countries,
however, is the same—falling back on a smorgasbord of democracy
programs based on the vague assumption that they all contribute to some
assumed process of consolidation is not good enough. Democracy aid
must proceed from a penetrating analysis of the particular core syndrome
that defines the political life of the country in question, and how aid
interventions can change that syndrome.
Moving beyond the transition paradigm also means getting serious
about bridging the longstanding divide between aid programs directed
at democracy-building and those focused on social and economic
development. USAID has initiated some work on this topic but has only
scratched the surface of what could become a major synthesis of disparate
domains in the aid world. One example of a topic that merits the
combined attention of economic aid providers and democracy promoters
is privatization programs. These programs have major implications for
how power is distributed in a society, how ruling political forces can
entrench themselves, and how the public participates in major policy
decisions. Democracy promoters need to take a serious interest in these
reform efforts and learn to make a credible case to economists that they
should have a place at the table when such programs are being planned.
The same is true for any number of areas of socioeconomic reform that
tend to be a major focus of economic aid providers and that have
Journal of Democracy20
potentially significant effects on the underlying sociopolitical domain,
including pension reform, labor law reform, antitrust policy, banking
reform, and tax reform. The onus is on democracy-aid providers to
develop a broader conception of democracy work and to show that they
have something to contribute on the main stage of the development-
assistance world.
These are only provisional ideas. Many other “next generation”
challenges remain to be identified. The core point, however, is plain:
The transition paradigm was a product of a certain time—the heady early
days of the third wave—and that time has now passed. It is necessary
for democracy activists to move on to new frameworks, new debates,
and perhaps eventually a new paradigm of political change—one suited
to the landscape of today, not the lingering hopes of an earlier era.
NOTES
The author would like to thank Jeffrey Krutz for his research assistance relating to this
article and Daniel Brumberg, Charles King, Michael McFaul, Marina Ottaway, Chris
Sabatini, and Michael Shifter for their comments on the first draft.
1. Samuel P. Huntington, The Third Wave: Democratization in the Late Twentieth
Century (Norman: University of Oklahoma Press, 1991).
2. Guillermo O’Donnell and Philippe C. Schmitter, Transitions from Authoritarian
Rule: Tentative Conclusions About Uncertain Democracies (Baltimore: Johns Hopkins
University Press, 1986).
3. Ruth Collier argues that a similar transition paradigm has prevailed in the scholarly
writing on democratization. “The ‘transitions literature,’ as this current work has come
to be known, has as its best representative the founding essay by O’Donnell and Schmitter
(1986), which established a framework that is implicitly or explicitly followed in most
other contributions.” Ruth Berins Collier, Paths Toward Democracy: The Working Class
and Elites in Western Europe and South America (Cambridge: Cambridge University
Press, 1999), 5.
4. “Building Democracy in the Democratic Republic of Congo,” www.usaid.gov/
democracy/afr/congo.html. Here and elsewhere in this article, I cite USAID documents
because they are the most readily available practitioners’ statements of guidelines and
political assessments, but I believe that my analysis applies equally well to most other
democracy-promotion organizations in the United States and abroad.
5. The conception of democratization as a predictable, sequential process of
incremental steps is vividly exemplified in USAID’s “managing for results” assessment
system. See Handbook of Democracy and Governance Program Indicators (Washington,
D.C.: USAID, August 1998).
6. Guillermo O’Donnell argues that the concept of democratic consolidation has
teleological qualities, in “Illusions About Consolidation,” Journal of Democracy 7 (April
1996): 34–51. A response to O’Donnell’s charge is found in Richard Gunther, P. Niki-
foros Diamandouros, and Hans-Jürgen Puhle, “O’Donnell’s ‘Illusions’: A Rejoinder,”
Journal of Democracy 7 (October 1996): 151–59.
7. See, for example, Giuseppe Di Palma, To Craft Democracies: An Essay on
Democratic Transitions (Berkeley: University of California Press, 1991). Dankwart
Thomas Carothers 21
Rustow’s article “Transitions to Democracy: Toward a Comparative Model,” originally
appeared in Comparative Politics 2 (April 1970): 337–63.
8. USAID’s current listing of the types of governance programs in its democracy-
assistance portfolio, for example, contains no examples of work on fundamental state-
building. See “Agency Objectives: Governance,” www.usaid.gov/democracy/gov.html.
9. An insightful account of the state of the third wave is found in Larry Diamond,
“Is the Third Wave Over?” Journal of Democracy 7 (July 1996): 20–37.
10. Larry Diamond uses the term “twilight zone” to refer to a sizeable but smaller
set of countries—electoral democracies that are in a zone of “persistence without
legitimation or institutionalization,” in Developing Democracy: Toward Consolidation
(Baltimore: Johns Hopkins University Press, 1999), 22.
11. David Collier and Steven Levitsky, “Democracy with Adjectives: Conceptual
Innovation in Comparative Research,” World Politics 49 (April 1997): 430–51.
12. Guillermo O’Donnell, “Delegative Democracy,” Journal of Democracy 5
(January 1994): 55–69.
13. These quotes are all taken from the country descriptions in the democracy-
building section of the USAID website, www.usaid.gov/democracy.html.
14. See, for example, Michael Bratton and Nicolas van de Walle, Democratic
Experiments in Africa: Regime Transitions in Comparative Perspective (Cambridge:
Cambridge University Press, 1997); Valerie Bunce, Subversive Institutions: The Design
and Destruction of Socialism and the State (Cambridge: Cambridge University Press,
1999); Ruth Collier, Paths Toward Democracy; Dietrich Rueschmeyer, Evelyne Huber
Stephens, and John D. Stephens, Capitalist Development and Democracy (Chicago:
Chicago University Press, 1992); Adam Przeworksi, Democracy and the Market:
Political and Economic Reforms in Latin America and Eastern Europe (Cambridge:
Cambridge University Press, 1991); and Adam Przeworksi and Fernando Limongi,
“Political Regimes and Economic Growth,” Journal of Economic Perspectives 7 (Summer
1993): 51–69.
15. See Charles King, “Potemkin Democracy,” The National Interest 64 (Summer
2001): 93–104.
