Content uploaded by Jason Brownlee
Author content
All content in this area was uploaded by Jason Brownlee on Jul 07, 2014
Content may be subject to copyright.
HEREDITARY SUCCESSION IN
MODERN AUTOCRACIES
By JASON BROWNLEE*
I
N summer 2006 a ranking member of Egypt’s ruling National Dem-
ocratic Party speculated the party might nominate President Hosni
Mubarak’s forty-two-year-old son Gamal in the next presidential elec-
tion, then scheduled for 2011.
1
Given Egypt’s recent history—including
the overthrow of a monarchy in 1952 and subsequent rule by military
officers—succession by the civilian Gamal appears unprecedented.
2
Yet
placed in global context, hereditary succession in Egypt would seem
conventional. Beginning with North Korean president Kim Jong-il’s
installation in 1994, the sons of autocratic executives have come to
power at an average rate of one every three years. Regimes in Equato-
rial Guinea, Libya, and Yemen are poised for the same kind of transfer.
What explains the phenomenon of extending authoritarianism through
dynasticism?
Hereditary succession among republic-style autocracies was unher-
alded by earlier studies of authoritarianism and remains unexplained
in the hybrid regimes literature. Samuel Huntington claimed that in-
herited authority and plebiscitary institutions would not long coexist.
3
Another author stated more categorically that political rule through
“heredity has come to an end.”
4
Subsequently, students of autocracy
* The author thanks Catherine Boone, Daniel Chirot, Larry Diamond, Jennifer Gandhi, Barbara
Geddes, Kenneth Greene, Clement Henry, Steven Heydemann, Wendy Hunter, Ronald Krebs,
Steven Levitsky, Eric McDaniel, Patrick McDonald, Corrine McConnaughy, Samer Shehata, Dan
Slater, Benjamin Smith, Joshua Stacher, Alfred Stepan, Lucan Way, Ismail White, and three anony-
mous reviewers for incisive comments on prior drafts of this article.
1
“Egypt NDP Member Hints at Gamal for President,” Reuters News, September 19, 2006, http://
global.factiva.com (accessed September 25, 2006).
2
Muhammad Abdul Aziz and Youssef Hussein, “The President, the Son, and the Military:
Succession in Egypt,” Arab Studies Journal 9/10 (Fall 2001/Spring 2002); Samer Shehata, “Political
Succession in Egypt,” Middle East Policy 9 (September 2002); and Mary Anne Weaver, “Pharaohs-In-
Waiting,” The Atlantic (October 2003).
3
Samuel Huntington, Political Order in Changing Societies (New Haven, Conn.: Yale University
Press, 1968), 168.
4
Robbins Burling, The Passage of Power: Studies in Political Succession (New York: Academic Press,
1974), 142. One recent attempt at apprehending hereditary succession in autocratic regimes is Mu-
hammad Tharwat, Asr al-jumhuriyat al-malakiyah [The Era of Monarchical Republics] (Cairo, Egypt:
al-Dar al-Thaqafiyah lil-Nashr, 2002).
World Politics 59 (July 2007), 595–628
596 w o r l d p o l i t i c s
5
Marina Ottaway and Martha Brill Olcott, “The Challenge of Semi-Authoritarianism,” Carnegie
Paper, no.7 (October 1999); Andreas Schedler, “The Democratic Revelation,” Journal of Democracy 11
(October 2000); Ellen Lust-Okar and Amaney Jamal, “Rulers and Rules: Reassessing the Influence of
Regime Type on Electoral Law Formation,” Comparative Political Studies 35 (April 2002); Larry Jay Dia-
mond, “Thinking about Hybrid Regimes,” Journal of Democracy 13 (April 2002); and Steven Levitsky
and Lucan A. Way, “The Rise of Competitive Authoritarianism,” Journal of Democracy 13 (April 2002).
6
Azerbaijan held multiparty parliamentary elections in 1990, 1995, and 2000, as well as multican-
didate presidential elections won by Heydar Aliyev, in 1993 and 1998. After gaining independence in
1965, Singapore has held ten multiparty elections. The Togolese government introduced multiparty
elections in 1993. Hans Florian Grotz and Raoul Motika, “Azerbaijan,” in Dieter Nohlen, Florian
Grotz, and Christof Hartmann, eds., Elections in Asia and the Pacific: A Data Handbook (Oxford: Ox-
ford University Press, 2001), 1:362–64; Christoph Rieger, “Singapore,” in Dieter Nohlen, Florian
Grotz, and Christof Hartmann, eds., Elections in Asia and the Pacific: A Data Handbook (Oxford: Ox-
ford University Press, 2001), 2:254–55; International Institute for Democracy and Electoral Assist-
ance, http://www.idea.int/vt/country_view.cfm?CountryCode=SG (accessed November 6, 2006); and
Michael Bratton and Nicolas Van de Walle, Democratic Experiments in Africa: Regime Transitions in
Comparative Perspective (New York: Cambridge University Press, 1997), 196–97.
7
Lisa Wedeen, “Seeing Like a Citizen, Acting Like a State: Exemplary Events in Unified Yemen,”
Comparative Studies in History and Society 45 (October 2003), 691; Richard Snyder, “Beyond Electoral
Authoritarianism: The Spectrum of Nondemocratic Regimes,” in Andreas Schedler, Electoral Authori-
tarianism: The Dynamics of Unfree Competition (Boulder, Colo.: Lynne Rienner, 2006), 220–22; and
Lisa Anderson, “Searching Where the Light Shines: Studying Democratization in the Middle East,”
Annual Review of Political Science 9 (2006), 199.
8
See, among others, Barbara Geddes, “Authoritarian Breakdown: Empirical Test of a Game The-
oretic Argument,” (paper presented at the annual meeting of the American Political Science Associa-
tion, Atlanta, Ga., September 2–5, 1999); Paul Brooker, Non-Democratic Regimes: Theory, Government
and Politics (New York: St. Martin’s Press, 2000); Michael L. Ross, “Does Oil Hinder Democracy?”
World Politics 53 (April 2001); David Waldner, “Democracy and Dictatorship in the Post-Colonial
World,” (paper presented at the annual meeting of the American Political Science Association, Chi-
cago, September 2–5, 2004); Ellen Lust-Okar, Structuring Contestation in the Arab World: Incumbents,
Opponents, and Institutions (New York: Cambridge University Press, 2005); Marsha Pripstein Posus-
ney and Michele Penner Angrist, eds., Authoritarianism in the Middle East: Regimes and Resistance
(Boulder, Colo.: Lynne Rienner Press, 2005); Benjamin Smith, “Life of the Party: The Origins of
Regime Breakdown and Persistence under Single-Party Rule,” World Politics 57 (April 2005); Lucan
A. Way, “Authoritarian State Building and the Sources of Regime Competitiveness in the Fourth
Wave: The Cases of Belarus, Moldova, Russia, and Ukraine,” World Politics 57 ( January 2005); and
Dan Slater, “Ordering Power: Contentious Politics, State-Building and Authoritarian Durability in
Southeast Asia” (Ph.D. diss., Emory University, 2005).
9
Studies of succession have focused on the Soviet Union and Mexico, but have not established
cross-regional explanatory frameworks. Mexico’s Partido Revoluncionario Institucional (PRI, r. 1929–
2000) selected a nonincumbent presidential candidate every six years, providing an exceptionally
tended to focus on regimes’ semblance, if not their substantive like-
ness, to electoral democracies.
5
Such works illuminate autocracy’s latest
republican features but do not address the countercurrent of suppos-
edly defunct dynastic practices. Hereditary succession in Azerbaijan
in 2003, Singapore in 2004, and Togo in 2005 overshadowed nominal
multiparty competition, yet evoked little analysis.
6
Hence, while fo-
cusing on elections comparativists may have neglected areas and ac-
tors more decisive for the allocation of political power.
7
Research on
comparative authoritarianism is beginning to address the historic and
organizational sources of authoritarian resilience,
8
but there is little on-
going research about how such regimes might operate when the pres-
ent cohort of rulers departs.
9
h e r e d i t a r y s u c c e s s i o n 597
robust answer to the succession question. On the Soviet Union, see Valerie Bunce, “Leadership Suc-
cession and Policy Innovation in the Soviet Republics,” Comparative Politics 11 ( July 1979); and the
exchange between Valerie Bunce and Philip G. Roeder, “The Effects of Leadership Succession in the
Soviet Union,” American Political Science Review 80 (March 1986). On the PRI, consult Alan Knight,
“Mexico’s Elite Settlement: Conjuncture and Consequences,” in John Higley and Richard Gunther,
eds., Elites and Democratic Consolidation in Latin America and Southern Europe (New York: Cambridge
University Press, 1992); and Stephan Haggard and Robert R. Kaufman, The Political Economy of Dem-
ocratic Transitions (Princeton: Princeton University Press, 1995).
10
Gordon Tullock, Autocracy (Boston: Kluwer Academic Publishers, 1987).
11
The data set compiles relevant political and genealogical information on nonmonarchical rul-
ers in power for at least three years. In addition to supplementary sources on specific cases, the main
sources for information were the Library of Congress Country Studies series; Geddes (fn. 8); Dieter
Nohlen, Michael Krennerich, and Bernhard Thibaut, eds., Elections in Africa: A Data Handbook (Ox-
ford: Oxford University Press, 1999); Nohlen, Grotz, and Hartmann (fn. 6, vols. 1 and 2); Encyclopedia
of World Biography (Detroit, Mich.: Gale Group, 1998); and Hein Gomans, Kristian Skrede Gleditsch,
Giacomo Chiozza, Archigos: A Dataset of Leaders 1875–2004 (2006), (accessed August 18, 2006).
This article opens a comparative investigation into hereditary suc-
cession in post–World War II autocracies. I draw upon prior research
by John Herz and Gordon Tullock to test a political theory for he-
reditary succession. Tullock hypothesizes that hereditary succession
provides a system for sustaining extant power distributions among the
broader elite. The benefits of hereditary succession thereby spread be-
yond the immediate ruler and successor, ensuring continued status for
extra-familial elites.
10
I extend this argument to account for the contrast
cases of orderly nonhereditary succession. Whether elites will abet he-
reditary succession depends on the precedent for leadership selection.
Where rulers are predated by parties, surrounding elites will defer to
the party as the recognized arbiter of succession. Where rulers predate
their parties and elites lack an established precedent for the orderly
transfer of power, hereditary succession offers a focal point for reducing
uncertainty, achieving consensus, and forestalling a power vacuum. In
these instances elites will accede to the ruler’s choice of heir apparent.
I test this theory using an original data set of 258 post–World War
II autocrats who ruled for at least three years.
11
A ruler’s longevity in
office and his fortune at having eligible offspring are the precondi-
tions for comparing hereditary and nonhereditary successions. From
the full universe of autocratic rulers, I identify the relevant universe of
potential hereditary successions based on two criteria: the ruler’s sur-
vival in office to the point of preparing an orderly succession (typi-
cally through the designation of an heir apparent) and the presence of
a viable adult heir. In the resulting twenty-two cases I examine how
surrounding nonfamilial elites responded to heirs apparent, including
whether they backed the ruler’s choice or opposed it and installed an
alternative figure. Outcomes among the potential hereditary succession
cases are largely consistent with the theory developed from Tullock.
598 w o r l d p o l i t i c s
The institutional precedents structuring elite behavior match the di-
vergence between hereditary and nonhereditary succession in seven-
teen out of twenty-two cases: five out of seven hereditary successions,
twelve out of fifteen nonhereditary successions. Notably, elites over-
rode the ruler’s selection of heir apparent only in cases where the party
enjoyed a precedent for selecting from within its ranks. In sum, he-
reditary succession predominated among those cases where the ruler’s
authority predated the party’s and occurred very rarely when the ruler
was himself the product of a preexisting party.
12
Case synopses show
this pattern, including the instances when elites from long-standing
parties obstructed hereditary succession.
The study holds several implications for comparative scholarship on
autocracy and succession. Popular culturalist accounts notwithstand-
ing, hereditary succession appears to primarily depend on the political
profile of the regime in question. While many a ruler may dream of
founding a dynasty, a son’s rise hinges on the response of the broader
ruling elite. Those elites are more prone to abet hereditary succession
when they lack an orderly precedent for leadership selection and are
wary of a leadership vacuum. Consequently, contemporary autocrats
who overshadow the parties through which they rule are likely initia-
tors for future hereditary successions. Indeed, in those cases, a wave of
elite resistance to autocrats grooming their sons would be strong evi-
dence disconfirming the theory presented in this article.
s
u c c e s s i o n pr a c t i c e s i n Mo d e r n au t o c r a c i e s
Leadership survival and succession has long been considered one of
the most daunting challenges for authoritarian rulers. It seemed un-
likely that developing states would devise institutions for regular, or-
derly succession as the Mexican regime achieved in the 1930s.
13
More
often, autocrats were expected to meet a violent and abrupt end be-
fore ever inaugurating a smooth transition.
14
The subsequent record of
fallen dictatorships largely supports these insights, and the appellation
12
In Egypt, Anwar Sadat made Hosni Mubarak his vice president and expected successor in 1975.
Mubarak thus gained the status of heir apparent three years before Sadat inaugurated the National
Democratic Party and installed Mubarak as the party’s vice chairman. On Sadat’s tapping of Mubarak
and the n d p ’s founding, see John Waterbury, The Egypt of Nasser and Sadat: The Political Economy of
Two Regimes (Princeton: Princeton University Press, 1983), 47, 369–71.
13
Dankwart Rustow, “Succession in the Twentieth Century,” Journal of International Affairs 18
(1964), 108–109.
14
John H. Herz, “The Problem of Successorship in Dictatorial Régimes: A Study in Comparative
Law and Institutions,” Journal of Politics 14 (February 1952), 19.
h e r e d i t a r y s u c c e s s i o n 599
“president-for-life” is an unrealized aspiration for many despots.
15
When scholars account for the heavy attrition rate among autocrats,
hereditary succession appears to be a significant variant of a rare phe-
nomenon: the seamless passage of power from one autocratic ruler to
his preferred heir.
Hereditary succession can be distinguished from other forms of
leadership change by the following three factors:
1. Transfer of top governing authority from father to son.
2. Preparation or initiation of power transfer prior to the ruler’s death.
3. Absence of formal democratic procedures (electoral democracy) or
legal stipulation of familial rule (traditional monarchies).
A word on each of these points.
Regarding the first criterion, hereditary successions entail the trans-
fer of de facto power, not simply the formal transition of office from one
figure to the next. Although the father-ruler’s control may be obscured
by installing a figurehead placeholder—as Anastasio Somoza García
did from 1947 to 1950 in Nicaragua and Lee Kuan Yew did between
1990 and 2004 in Singapore—substantive political control remains in
his hands and it is this power which passes to the son-heir.
16
Heredi-
tary succession is also limited in this study to father-son leadership
transitions deliberately implemented by the ruler before his death, the
second conceptual restriction.
17
While there are many possible fam-
ily members who might conceivably take power, sons have been the
predominant heirs. Brothers are typically of the same generation and
fraternal successions are a less appealing course for reasons discussed
below.
18
Likewise, in post–World War II autocracies sisters and spouses
15
On the difficulties of undemocratic rulers retaining power, see Henry Bienen and Nicolas van
de Walle, Of Time and Power: Leadership Duration in the Modern World (Palo Alto, Calif.: Stanford
University Press, 1991); Geddes (fn. 8); Bruce Bueno de Mesquita, Alastair Smith, Randolph M.
Siverson, and James D. Morrow, The Logic of Political Survival (Boston, Mass.: MIT Press, 2003),
300–302.
16
I discuss these examples in greater detail below. During the transition period in Singapore, Lee
Kuan Yew and his eventual heir suppressed public discussions of hereditary succession, even charging the
International Herald Tribune with libel for an opinion column about dynastic politics in Singapore. Philip
Bowring, “The Claims About ‘Asian’ Values Don’t Usually Bear Scrutiny,” International Herald Tribune,
August 2, 1994; and “Nepotism: A Little More than Kin,” Economist, December 24, 1994. Subsequent
events supported the Tribune columnist’s argument that political authority had remained within the
Lee family. A decade after Lee Kuan Yew’s putative handover of power, one observer remarked, “Prime
Minister Goh Chok Tong still is seen by some to be functioning virtually on probation.” Raj Vasil,
Governing Singapore: Democracy and National Development (Singapore: Allen & Unwin, 2000), 137.
17
Under these strict definitions I omit the post hoc succession of Joseph Kabila in the Democratic
Republic of Congo following President Laurent Kabila’s assassination in 2001.
18
Were Raúl Castro to succeed Fidel Castro, it would mark the only case in the data set where
a long-ruling leader overlooked multiple eligible sons and inaugurated a brother-brother transfer of
power. Juan O. Tamayo, “Castro’s Secret Life: Even the CIA is Left to Wonder about the ‘Maximum
600 w o r l d p o l i t i c s
have not succeeded their brothers and husbands. While they may re-
main influential actors, they have not gained top posts in the wake of
the ruler’s passing. The closest case to spousal succession was Simone
Ovid Duvalier’s custodial rulership during the early years of Jeane-
Claude “Baby Doc” Duvalier’s presidency. In one apparent exception
Isabel Perón had been the elected vice-president prior to her husband’s
death and is thus omitted under the third and final regime type restric-
tion, which confines the study to nondemocracies.
Hereditary successions in modern autocracies occur in regimes
where rulers do not depend on regular, free, and fair elections to main-
tain their positions.
19
Regimes that meet the minimum Schumpeterian
standard for electoral democracy are omitted.
20
The theory of heredi-
tary succession presented here is not intended to address the persistent
influence of political families in such electoral democracies as India,
Indonesia, and the Philippines. Rather, this study concentrates on non-
monarchical autocracies in which fathers installed sons. Table 1 lists
the nine post–World War II cases identified by these measures (omit-
ting the single fraternal succession, Nicaragua’s 1967 transfer of power
from Luis Somoza Debayle to Anastasio Somoza Debayle upon the
former’s early death by cardiac arrest at age forty-four).
Although that number seems low in absolute terms, it is impossible
to assess the rarity of hereditary succession without also measuring
the frequency of the alternative: During the same post–World War II
period, how many orderly nonhereditary successions have occurred in
modern autocracies? How many rulers passed power to any successor
of their choosing? To answer these questions I composed an original
data set on post–World War II autocrats who ruled for at least three
years. Hundreds of autocrats have lost power during the post-War pe-
riod and the three-year threshold limits the focus to those rulers who
were at least minimally able to consolidate power.
21
Two hundred fifty-
eight met this criterion. I categorized those individuals based on seven
processes through which they left office: hereditary succession (includ-
ing fraternal succession); nonhereditary succession by the incumbent’s
designated successor, without term limits; nonhereditary succession
by the incumbent’s designated successor after a term-limited tenure;
Leader’ and His Family,” Seattle Times, October 25, 2000, http://global.factiva.com (accessed Decem-
ber 22, 2006).
19
Philippe Schmitter and Terry Lynn Karl, “What Democracy Is . . . And Is Not,” Journal of De-
mocracy (Summer 1991).
20
Joseph Alois Schumpeter, Capitalism, Socialism, and Democracy (New York: Harper & Brothers,
1947), 269.
21
Geddes (fn. 8), 1.
h e r e d i t a r y s u c c e s s i o n 601
replacement by the ruling party (involuntary removal); replacement by
a coup; replacement by an opposition victory (electoral or nonelectoral);
and replacement by foreign imposition. The distribution of these out-
comes is provided in Figure 1.
Given the political turbulence wrought by many autocracies, orderly
nonhereditary successions are uncommon. Of 258 rulers, 196 (76 per-
cent) were constrained by their parties or forcibly removed, depriving
them of the chance to groom an heir unfettered. A further sign that
events intervened upon succession options comes from data on the
age of rulers when they left power. Figure 2 shows the distribution of
leadership changes grouped by age of the autocrat at the time of his
exit. The sixty-two prepared hereditary or nonhereditary successions
are clustered among elderly autocrats—those who escaped the plots of
party rivals, domestic activists, and other would-be usurpers.
ta b l e 1
he r e d i t a r y su c c e s s i o n s i n Mo d e r n au t o c r a c i e s
(1946–2006)
Year Country Ruler Successor Notes
1956 Nicaragua Anastasio Luis Somoza Luis Somoza Debayle was
Somoza García Debayle succeeded by his brother
(r. 1936–56) (r. 1956–67) Anastasio Somoza
Debayle.
1960 Dominican Rafael Trujillo Rafael “Ramfis” Under U.S. military pressure,
Republic (r. 1930–61) Trujillo Ramfis Trujillo passed
Martinez power to then-nominal
(r. 1961) president Joaquín Balaguer.
1971 Haiti François Jean-Claude Jean-Claude Duvalier was
Duvalier Duvalier nineteen years old when
(r. 1957–71) (r. 1971–86) he assumed the presidency.
1975 Taiwan Chiang Kai-shek Chiang Ching-kuo
(r. 1949–75) (r. 1975–88)
1994 North Korea Kim Il-sung Kim Jong-il
(r. 1948–94) (r. 1994–present)
2000 Syria Hafez al-Assad Bashar al-Assad
(r. 1971–2000) (r. 2000–present
2003 Azerbaijan Heydar Aliyev Ilham Aliyev
(r. 1993–2003) (r. 2003–present)
2004 Singapore Lee Kuan Yew Lee Hsein Loong Lee Kuan Yew retained
(r. 1965–2004) (r. 2004–present) elder statesman post.
2005 Togo Gnassingbé Faure
Eyadéma Gnassingbé
(r. 1967–2005) (r. 2005–present)
so u r c e s : Consult references in case synopses in this article and data set (available from author).
602 w o r l d p o l i t i c s
Dictators who sufficiently entrenched themselves to survive in
power past age sixty cleared the biggest hurdles to orchestrated leader-
ship transitions, including hereditary handovers. Five of the nine rulers
who passed power to sons died in their sixties; the remaining four were
octogenarians at the time of succession. However, even among the
most resilient autocrats there are still more dictators who implement
nonhereditary successions. What explains the occurrence of hereditary
succession as opposed to the installation of other designees?
a p
o l i t i c a l th e o r y o f he r e d i t a r y su c c e s s i o n
Perhaps owing to its rarity amid the broader universe of autocratic leader-
ship outcomes, hereditary succession has seldom received systematic expla-
nation. The most prevalent treatment, by journalists, portrays hereditary
succession as a cultural fluke. It is commonly treated as a curiosity of Arab
politics. The New York Times editorial board claimed, for example, “While
dynasties can be found all over the world, they have been most com-
mon in Arab countries.”
22
That judgment elides the difference between
22
Editorial, “Dynastic Regimes,” New York Times, August 25, 2003); see also Charles M. Sennott,
“Arab Sons of Privilege Inherit Power and Instill Doubts,” Boston Globe, June 14, 2000; and Brian
Whitaker, “Hereditary Republics in Arab States,” Guardian Unlimited, August 28, 2001.
Party Removal
Mode of Exit
Frequency of Exit Mode
(number of occurrences)
80
70
60
50
40
30
20
10
0
Coup Foreign
Imposition
Hereditary
Succession
Opposition
Victory
Non-Hereditary
Succession (no
term limits)
Non-Hereditary
Succession
(term limits)
25
10
52
29
73
66
3
Hereditary Nonhereditary Nonhereditary Party Removal Coup Opposition Foreign
Succession Succession Succession Victory Imposition
(no term limits)
80
70
60
50
40
30
20
10
0
Mode of Exit
fi g u r e 1
le a d e r s h i p ch a n g e i n Mo d e r n au t o c r a c i e s
(1946–2006)
h e r e d i t a r y s u c c e s s i o n 603
long-standing monarchies and republic-style autocracies. Although
most traditional monarchies are found in the Arab world, the majority
of nonmonarchical hereditary successions occur in non-Arab states. A
span of successions stretching from the Caribbean to Southeast Asia
belies the notion of inherited rule as an Arab proclivity.
In contrast to journalists, political scientists regard hereditary succession
as the extension of an especially corrupt and personalistic autocracy.
23
The
practice, such scholars argue, originates not in the surrounding society but
in the regime itself. Encapsulating this view, Samer Shehata writes:
Rather than being the product of an essential Arab political culture, the phe-
nomenon is more likely specific to a particular type of authoritarianism—
centralization of power in the person of the leader, a small ruling elite, a lack
of institutionalized power centers outside the leader, a cult of personality, and
long-serving rulers who have been able to eliminate potential rivals.
24
Shehata correctly notes that hereditary successions outside the Arab
world have often been overlooked. Even so, the diversity of cases also
23
Juan J. Linz and Alfred Stepan, Problems of Democratic Transition and Consolidation: Southern Eu-
rope, South America, and Post-Communist Europe (Baltimore: Johns Hopkins University Press, 1996),
52; H. E. Chehabi and Juan J. Linz, “A Theory of Sultanism 1: A Type of Nondemocratic Rule,” in
H. E. Chehabi and Juan J. Linz, eds., Sultanistic Regimes (Baltimore: Johns Hopkins University Press,
1998), 15–17; Brooker (fn. 8), 144–47; and Tharwat (fn. 4), 117.
24
Shehata (fn. 2), 112.
Age at Time of Exit
(years)
fi g u r e 2
Mo d e s o f ex i t a n d ag e o f ou t g o i n g ru l e r
Frequency of Exit Mode
(number of occurrences)
Less than 40 40–50 50–60 60–70 Greater than 70
80
70
60
50
40
30
20
10
0
Removal from
Power (including
term limits)
Prepared
Succession-
(hereditary or
nonhereditary)
2
7
3
69
54
11
17
25
29
41
604 w o r l d p o l i t i c s
suggests Shehata’s account is incomplete. Neopatrimonialism and per-
sonality cults have accompanied several hereditary successions, but they
are not prerequisites. Table 1 includes Haiti, the Dominican Republic,
and North Korea, all notoriously corrupt and repressive regimes. The
same set also contains Taiwan and Singapore, governments known for
their strong institutions.
25
Hence, hereditary succession is not solely
driven by personalism run amok. Rather, the prevalence of hereditary
succession among otherwise dissimilar regimes signals a more basic
political process.
The distribution of outcomes in Figure 1 indicates rulers should be
just as concerned with threats from their ostensible coalition partners
(in the ruling party and the military) as with threats from the opposi-
tion. This insecurity, which is endemic to authoritarian regimes, be-
comes more acute as dictators promote fellow elites into positions of
influence, for in so doing they may engender their own downfall.
26
As
the ruler’s high-ranking associates accrue more power they become
more capable of mounting a successful challenge and potentially more
tempted to venture such a move. Herz dubs this security dilemma—
the risk that by clearly grooming a successor the ruler positions a ri-
val who may try to supplant him— the “crown-prince problem.”
27
As
Tullock describes the issue, “if [the ruler] anoints a successor, this gives
that successor both strong motives for assassinating him and reason-
able security that he will get away with it…Looked at from the stand-
point of the dictator, then, it’s dangerous to have an official successor.”
28
And there is a further complication; failure to choose a clear successor
heightens the likelihood of an unregulated power struggle in the wake
of the ruler’s death (without eliminating the potential for a coup prior
to that point).
29
The ruler and his associates thus shoulder a double
burden; they must provide for the incumbent’s security while the ruler
is alive and for the regime’s continuation after the ruler departs. One
mechanism for continuing the regime in an orderly fashion is a preex-
isting party through which elites may reach consensus about the in-
25
In their comparative study of developmental and patrimonial states, Peter Evans and James
Rauch code the Dominican Republic and Haiti among the most personalized regimes of their data set
(with scores of 2.0 and 4.0, out of a maximum of 13.50) while placing Singapore (13.50) and Taiwan
(12.0) at the top of their rankings. Peter Evans and James E. Rauch, “Bureaucracy and Growth: A
Cross-National Analysis of the Effects of ‘Weberian’ State Structures on Economic Growth,” Ameri-
can Sociological Review 64 (October 1999), 763.
26
For two influential examples see Bienen and van de Walle (fn. 15); and Bueno de Mesquita et
al. (fn. 15).
27
Herz (fn. 14), 30.
28
Tullock (fn. 10), 151, 153.
29
Herz (fn. 14), 37.
h e r e d i t a r y s u c c e s s i o n 605
cumbent’s replacement. Such parties are not always available, however,
and the crown-prince problem is not easily solved.
In his book-length treatment of autocratic rule, Tullock hypoth-
esizes that hereditary succession provides a method for regime stability
during and beyond the ruler’s lifetime. By grooming his son for suc-
cession the dictator may resolve the crown-prince problem, affording
mutual security to incumbent and appointee while dispelling the sur-
rounding elite’s apprehension of a power vacuum. Rulers thus prefer
sons over alternative figures more inclined to hasten the succession
through assassination or coup attempts.
“[The] son is wise to simply wait for his father to die. He knows this, his father
knows that he knows this and concern about assassination by son is less in a
hereditary successional arrangement than if the designated successor is a high
ranking official of the existing regime.”
30
Following this same logic, brothers and other coevals of the ruler’s
family are less attractive as successors for they are more likely to be
usurpers. The generational gap and genealogical line between a ruler
and his son allow both individuals to enjoy security. By contrast, a
brother or other relative close in age to the ruler is unlikely to patiently
await the incumbent’s natural passing. In fact, a ruler who promotes
a familial coeval may exacerbate the crown-prince problem.
31
Among
the universe of 258 rulers in the data set, so-called lateral successions
occurred only in the Somoza regime (at a time when neither of the
brothers had an adult son for passing power vertically to the next gen-
eration).
32
In addition to being an attractive option for incumbent autocrats, he-
reditary succession appeals to most of the surrounding political elites—
the very pool of potential heirs not designated to fill the ruler’s post.
With an established pattern for preserving the regime after a ruler’s
death, other incumbent officeholders share an incentive to accept the
ruler’s appointee rather than vie for power in an unbridled struggle.
33
30
Tullock (fn. 10), 163.
31
See Bueno de Mesquita et al. (fn. 15), 17, for historical illustrations of this problem.
32
The threat from nonfilial relatives shows in at least three cases where familial coevals attempted
or successfully executed coups. In the Dominican Republic, Rafael Trujillo’s brother Petán plotted to
seize power during the mid-1930s and was repeatedly posted abroad as a result. In Equatorial Guinea
in 1979, current president Teodoro Obiang Nguema took power by ousting his uncle, Francisco
Macías Nguema (eighteen years his senior). In Syria in 1983, Hafez al-Assad survived a coup attempt
by his brother Rifaat, who was subsequently exiled. Robert D. Crassweller, Trujillo: The Life and Times
of a Caribbean Dictator (New York: Macmillan, 1966), 140–41; Daniel Chirot, Modern Tyrants: The
Power and Prevalence of Evil in Our Age (New York: Free Press, 1994), 363–64; and Patrick Seale, Asad:
The Struggle for the Middle East (Berkeley, Calif.: University of California Press, 1988), 432–35.
33
Tullock (fn. 10), 151–57.
606 w o r l d p o l i t i c s
Tullock thus reasons there are few institutional arrangements through
which autocrats favor nonhereditary succession and he forecasts that
“hereditary monarchs” will gradually predominate among nondemoc-
racies.
34
Recent work by Michael Herb buttresses Tullock’s theory and
illustrates the elite behaviors that converge on hereditary succession.
Refuting earlier arguments that monarchies could not manage mod-
ernizing reforms, Herb found the Persian Gulf monarchies persisted
through a form of power-sharing that placated would-be successors
and prevented fissures in the ruling elite. In these regimes, even mem-
bers of the royal family not tapped as heir apparent were invested in
the regime’s survival because they held top cabinet posts including the
prime ministership, the ministry of defense, the ministry of foreign
affairs, and the ministry of the interior. Thus ensconced, when succes-
sion neared, they would band together rather than polarize and risk an
internecine feud.
35
These works outline a political explanation for hereditary succession
in modern autocracies. Because elites prefer maintaining their status
to pursuing a potentially disastrous power grab, they commonly seek
a nonsultanistic push for hereditary succession—one that extends be-
yond the immediate goals of the ruler’s family. This strategy is rational
and arithmetically prudent. Only one individual can be the top ruler
while many more than that can benefit from the regime or, alterna-
tively, suffer from its downfall. When given the choice between low-
intensity conflicts over policy issues and high-intensity struggles for
the top posts in the regime, elites tend to avoid the second kind of
conflict.
36
Clearly, the ruler, son, and first lady may see a common good
in preserving the family’s influence after the ruler’s departure. This as-
piration is ultimately enabled by the son’s role as a second-best choice,
after themselves, for each of the surrounding extra-familial elites. For
party apparatchiks and other top regime supporters, hereditary succes-
sion is inherently exclusionary; it limits the rulership to the dictator’s
lineage. However, it may be much less exclusionary than the rise of a
new chief executive from outside the ruler’s family. Sheila Carapico
notes that completed succession means the ruling coalition “remains
34
Ibid., 161, 166, 215.
35
Michael Herb, All in the Family: Absolutism, Revolution, and Democracy in the Middle Eastern
Monarchies (Albany, N.Y.: State University of New York Press, 1999), 8. On the king’s dilemma, see
Huntington (fn. 3), 168.
36
David Waldner has differentiated between low-intensity elite conflict, in which elite status is
not threatened, and high-intensity elite conflict, in which elites face permanent loss of elite status.
For treatment of leadership succession, I treat conflict intensity as a product of elite strategies, rather
than as an independent variable. David Waldner, State Building and Late Development (Ithaca, N.Y.:
Cornell University Press, 1999), 29.
h e r e d i t a r y s u c c e s s i o n 607
intact. . . . Policies might change, but political arrangements emerge
unscathed, without struggle or upheaval.”
37
In this way, the son-heir
provides a “focal point” for elite consensus once he is tapped for suc-
cession by the father-ruler.
38
Additionally, senior leaders may expect to
hold sway over an ostensibly inexperienced and pliable heir. Under this
set of circumstances, hereditary succession becomes the most palatable
of many possible outcomes, far preferable to the regime’s disintegration
or to a series of purges by a wrathful victor.
i
d e n t i f y i n g co n t r a s t ca s e s f o r th e o r y te s t i n g
Historical comparisons and anecdotal observations support Tullock’s
expectation, but no study has yet tested his proposition across the uni-
verse of succession outcomes in the developing world.
39
To evaluate
whether succession patterns accord with Tullock’s thesis, I contrast he-
reditary succession with alternative outcomes—the “contrast cases” to
hereditary succession.
40
The identification of cases that provide varia-
tion in the dependent variable (succession outcome) begins with two
logical requisites for hereditary succession: political opportunity for
succession (remaining in power to the point of grooming a successor)
and biological opportunity for hereditary succession (presence of an
adult male heir). These factors remove those instances in which the
outcome of hereditary succession was already precluded and thereby
hone in on the relevant counterfactuals.
p
o l i t i c a l op p o r t u n i t y
In the full universe of autocracies, hereditary successions seem rare. As
illustrated above, most rulers are ousted long before they might choose
to depart and thus they never transfer power to a desired successor.
Consider, for example, Syria, where commentators remarked on the
father-son succession in 2000 as unusual but said little about the his-
toric rarity of orderly leadership transition in any form.
37
Sheila Carapico, “Successions, Transitions, Coups, and Revolutions,” Middle East Policy 9 (Sep-
tember 2002).
38
Thomas C. Schelling, The Strategy of Conflict (Cambridge, Mass.: Harvard University Press,
1960), 57.
39
Peter Kurrild-Klitgaard, “The Constitutional Economics of Autocratic Succession,” Public Choice
103 (April 2000), 79; and Gordon Tullock, “Undemocratic Governments,” Kyklos 55 (2002), 262.
40
Barbara Geddes, “How the Cases You Choose Affect the Answers You Get: Selection Bias in
Comparative Politics,” Political Analysis 2 (1990), 131; Gary King, Robert O. Keohane, and Sidney
Verba, Designing Social Inquiry: Scientific Inference in Qualitative Research (Princeton: Princeton Uni-
versity Press, 1994), 129; and David Collier and James Mahoney, “Insights and Pitfalls: Selection Bias
in Qualitative Research,” World Politics 49 (October 1996), 67.
608 w o r l d p o l i t i c s
Hafez al-Assad commandeered the Baath Party in November 1970
and became president of Syria the following year.
41
Over subsequent
decades Assad held onto his office despite military defeat, popular re-
volt, a power grab by his brother, and the country’s loss of Soviet pa-
tronage.
42
In 2000, then sixty-nine years old, Assad arranged for the
presidency to pass to his oldest remaining son, Bashar. Bashar’s prior
return from medical training in England and subsequent grooming for
a leadership role in Syrian politics had drawn intense scrutiny.
43
Yet
the focus on the successor missed a fundamental point about the out-
going ruler. Hafez al-Assad engaged in a deliberate transfer of power
without being forcibly removed from office or assassinated. A nonvio-
lent succession of this kind was unprecedented in modern Syria. In the
period 1947 to 1970, fifteen coups had been carried out there; Assad’s
numerous predecessors had been ousted or slain long before they had
the chance to contemplate a hereditary or nonhereditary succession.
44
Thus, beyond being a case of hereditary succession, Bashar al-Assad’s
inauguration was an instance of a prepared succession at the end of
a ruler’s natural life. The novelty of Bashar’s ascent is exaggerated by
contrasting it with the string of prior Syrian rulers who were unable to
orchestrate either hereditary or nonhereditary transitions of power.
While political survival is a precondition for hereditary succession,
an autocrat’s survival in office may also be influenced by the succession
process, particularly once an heir apparent emerges. If elites balk at
the prospect of hereditary succession they may conspire to remove the
leader and choose his successor from among their own ranks. Tullock
predicts otherwise, but to evaluate his theory one must consider the
possibility that political survival is endogenous to the ruler’s succession
plans. In order to pin down whether hereditary succession wards off
dissension or invites it, I examine, where possible, how elites respond
to the ruler’s designated heir apparent, a figure who typically emerges
late in the dictator’s tenure. Was the ruler’s designated heir apparent
accepted or opposed by the surrounding elites, the figures Tullock ex-
pects to rally behind a filial appointee?
Of course some events are censored from observation and in those
cases whether elites would have gone through with a hereditary suc-
41
James T. Quinlivan, “Coup-Proofing: Its Practice and Consequences in the Middle East,” Inter-
national Security 24 (Autumn 1999), 134.
42
Steven Heydemann, Authoritarianism in Syria: Institutions and Social Conflict, 1946–1970 (Ithaca,
N.Y.: Cornell University Press, 1999), 2–4.
43
On Bashar’s heir apparency in the late 1990s and succession during June–July 2000, see Joshua
A. Stacher, “Adapting Authoritarianism: Institutions and Co-optation in Egypt and Syria” ( Ph.D.
diss., University of St. Andrew’s, 2007), 122–28.
44
Quinlivan (fn. 41), 134.
h e r e d i t a r y s u c c e s s i o n 609
cession cannot be determined. One example of this is when a ruler has
groomed his son, but he, the ruler, is removed by nonelite actors either
through a revolutionary process or foreign conquest.
45
There are, how-
ever, cases that are not censored in which elite dissent was apparent
and in which elite opposition to hereditary succession did contribute to
the ruler’s downfall. Hereditary succession plans were a salient factor in
the removal of Alfredo Stroessner in Paraguay and Todor Zhivkov in
Bulgaria. The cases of Stroessner and Zhivkov are therefore included
among the contrast cases.
b
i o l o g i c a l op p o r t u n i t y
Rulers who elude their foes and stay in power for decades are still not
guaranteed the option of hereditary succession. They may have the
chance to oversee succession, but may lack a viable heir, an adult son. At
that point biological opportunity presents a secondary constraint and
screens out many would-be dynasts. While the question of producing
offspring is rarely a subject of political science discussions, the vagaries
of genetics and matrimony have precluded hereditary successions for
more than a few resilient autocrats. Houari Boumédienne in Algeria (r.
1965–78) and António Salazar in Portugal (r. 1932–68) were bachelor
presidents. Francisco Franco in Spain (r. 1939–75) and Josip Tito in
Yugoslavia (r. 1945–80) had no son in their presidential families. These
long-ruling autocrats perforce enacted nonhereditary successions. To
determine which leaders could entertain the option of a hereditary suc-
cession, I looked for the presence of a viable heir.
46
In addition, among the universe of 258 rulers, the average age upon
entry into power was forty-nine years, twenty-nine days. The youngest
ruler was Jean-Claude “Baby Doc” Duvalier, who succeeded his father
when he was nineteen years old. The second youngest ruler who came to
power through an orderly succession was thirty-one-year-old Norodom
45
Two empirical cases in which succession was screened from observation are the December 1989
ouster of Nicolae Ceaus¸escu in Romania and the March 2003 deposition of Saddam Hussein in Iraq.
In Romania, Nicu Ceaus¸escu had been the clear frontrunner to succeed his father. In Iraq, Hussein’s
heir apparent was his son Qusay. But hereditary succession plans in these cases were not a proxi-
mate cause of the ruler’s removal. These cases are therefore not considered as confirming or discon-
firming the present theory. On dynastic socialism in Romania and the overthrow of Ceaus¸escu see
René de Flers, “Socialism in One Family,” Survey: A Journal of East & West Studies 28 (Winter 1984);
Vladimir Tismaneanu, “Ceaus¸escu’s Socialism,” Problems of Communism 34 ( January–February 1985);
and Chirot (fn. 32), 256. On Saddam Hussein’s regime and Qusay’s promotion see “Iraq-Government
Saddam Hussein’s Second Son Named to Governing Council,” EFE News Service, June 9, 2001,
http://global.factiva.com (accessed June 18, 2007).
46
I treat biological opportunity as exogenous. For example, I do not seek to explain why António
Salazar never married or why Francisco Franco sired no sons. Rather, I note those facts as they affect
the question of why the Portuguese and Spanish regimes did not have hereditary successions while
others, such as the Syrian regime, did.
610 w o r l d p o l i t i c s
Sihanouk of Cambodia.
47
Accordingly, I operationalized “viable heir”
as a son who was no younger than thirty years of age.
48
an a l y z i n g t h e da t a
The political and biological preconditions of hereditary succession help
to discern relevant contrast cases. After considering political longevity
to the time when a ruler begins to groom his successor and biological
opportunity as the existence of a son to fill that role, the data pro-
vide fifteen nonhereditary successions (Table 2), the analytic foil to the
hereditary examples of Table 1. These contrast cases help to evaluate
Tullock’s theory and assess the conditions under which leaders pass
power to sons, pass over them in favor of another designee, or attempt
to pass power to sons and are thwarted by discontent elites.
e
x p l a i n i n g Va r i a t i o n i n su c c e s s i o n ou t c o M e s
The cases in Table 2 provide variation in the dependent variable: con-
trast cases in which rulers with the political and biological opportu-
nity to bequeath power to a son are instead succeeded by an alternative
figure. These are the relevant counterfactual data for testing whether
nonfamilial elites support hereditary succession and for determining
whether sons play a role in preserving extant power arrangements.
Tullock’s work offers one variable for explaining nonhereditary succes-
sion—the presence of an accepted organizational method for selecting
leaders and designating a focal point.
Elites outside of the father-son dyad prefer the regime’s continuation
over an unconstrained power struggle that endangers their posts. But
if a practice for the nonviolent selection of a ruler already exists, lead-
ers will defer to that method over a tradition-breaking filial appoint-
ment. In essence, leaders will opt for the least risky of the two courses,
hereditary succession or nonhereditary succession. Their perception of
the potential costs depends on the precedent of the incumbent’s instal-
lation. Did the present ruler come from the party or vice versa? If an
organizational mechanism for leadership choice has already been used,
47
In the data set there are six rulers who took office through coups while between the ages of
twenty-five and thirty: Valentine Strasser (Sierra-Leone), Michel Micombero (Burundi), Muammar
Qadhafi (Libya), Samuel Doe (Liberia), Gnassingbé Eyadéma (Togo), and Marien Ngouabi (Congo-
Brazaville). Youth appears to be less a barrier to seizing power than it is for being granted power by
an incumbent.
48
Following this operationalization, I exclude the Haitian case from subsequent inferences. Al-
though the Duvaliers are an iconic example of hereditary succession, François Duvalier’s grooming of
his son appears to be exceptional. Inclusion of Haiti in the analysis would strengthen the postulated
relationship between institutional antecedents and succession practices. See David Nicholls, “The
Duvalier Regime in Haiti,” in Chehabi and Linz (fn. 23).
h e r e d i t a r y s u c c e s s i o n 611
elites will expectedly treat that practice as the default. In the absence of
such a process—that is, when the ruler was either a personal appointee
or the party’s founder—the ruler may promote his son while garnering
the deference of regime elites who lack an accepted alternative route.
The combined set of twenty-two succession cases provides the data
to evaluate whether hereditary succession predominates in the regimes
where rulers predated parties and elites had no prior practice in choos-
ta b l e 2
no n h e r e d i t a r y su c c e s s i o n s : t h e co n t r a s t ca s e s
(1946–2006)
Potential Hereditary
Year Country Ruler Successor Actual Successor
1953 Soviet Union Joseph Stalin Vassily Stalin Nikita Kruschev
(r. 1922–53) (r. 1953–1964)
1982 Soviet Union Leonid Brezhnev Yuri Brezhnev Yuri Andropov
(r. 1964–82) (r. 1982–1984)
1985 Albania Enver Hoxha Ilir Hoxha or Ramiz Alia
(r. 1944–85) Sokol Hoxha (r. 1985–1992)
1985 Sierra Leone Siaka Stevens James Stevens Joseph Saidu Momoh
(r. 1967–85) (r. 1985–93)
1985 Tanzania Julius Nyerere Charles Makongoro Ali Hassan Mwinyi
(r. 1961–85) Nyerere (r. 1985–95)
1988 Taiwan Chiang Ching-kuo Chiang Hsiao-wu Lee Teng-hui
(r. 1975–88) (r. 1988–2000)
1989 Bulgaria Todor Zhivkov Vladimir Zhivkov Petar Mladenov
(r. 1956–89) (r. 1989–90)
1989 Iran Ayatollah Ruhollah Ahmad Khomeini Ali Khamenei
Khomeini (r. 1989–present)
(r. 1979–89)
1989 Paraguay Alfredo Stroessner Gustavo Stroessner Andrés Rodríguez
(r. 1954–89) (r. 1989–93)
1992 China Deng Xiaoping Deng Pufang Jiang Zemin
(r. 1978–92) or Deng Zhifang (r. 1992–2004)
1992 Laos Kaysone Saysomphone Khamtai Siphandon
Phomvihane Phomvihane (r. 1992–2006)
(r. 1975–92)
2003 Malaysia Mahathir Mokhzani Mahathir Abdullah Badawi
Mohammed (r. 2003–present)
(r. 1981–2003)
2004 China Jiang Zemin Jiang Mianhang Hu Jintao
(r. 1992–2004) (r. 2004–present)
2005 Mozambique Joaquim Chissano Nyimpine Chissano Armando Guebuza
(r. 1986–2005) or N’naite Chissano r. 2005–present)
so u r c e s : Consult references in case synopses in this article and data set (available from author).
612 w o r l d p o l i t i c s
ing from among their ranks. Table 3 sorts the cases into two columns
based on institutional history and whether the ruler predates the party,
and places them in rows based on succession outcomes.
Institutional variations match the hypothesized outcomes in seven-
teen of the twenty-two successions (77 percent), and were thus consis-
tent with a political theory of hereditary succession. Additionally, with
regard to the behavior of nonfamilial elites, in two of five attempted
hereditary successions where parties predated the ruler, nonhereditary
succession occurred despite the ruler grooming his son, i.e., elites re-
jected the incumbent’s heir apparent. A close look at all of the cases il-
luminates the behavior of nonfamilial elites in these abortive hereditary
successions and in the other observed outcomes.
h
e r e d i t a r y su c c e s s i o n : ru l e r pr e d a t e s pa r t y
In five of the eight hereditary successions, rulers predated the party
in operation. In these cases—the Dominican Republic, Taiwan, Azer-
baijan, Singapore, and Togo—surrounding elites facilitated the plan to
install the ruler’s son.
General Rafael Leónidas Trujillo took power in the Dominican Re-
public through a coup d’état on February 23, 1930.
49
His opponents
cowed, Trujillo soon won the presidency in a single-candidate race.
50
He then created the Dominican Party, which was also permitted no
competitor and whose accounts Trujillo managed personally.
51
While
in his sixties, Trujillo began positioning his son Ramfis as heir appar-
ent.
52
When a small band of aggrieved affiliates assassinated the pres-
ident on May 30, 1961, Ramfis returned home from vacationing in
Paris, took command of the country’s armed forces, and spearheaded
the regime’s retribution against his father’s assailants.
53
Although the
regime at that point was visibly shaken by the death of its leader, the
conspirators failed to catalyze a country-wide revolt or even a military
coup. During this uncertain phase the security and military apparatus
remained cohesive and brutally effective at repressing dissent. As Jona-
than Hartlyn writes: “It took some eight months, with considerable
49
Frank Moya Pons, “The Dominican Republic Since 1930,” in Leslie Bethell, ed., The Cambridge
History of Latin America, vol. 8, Latin America since 1930: Spanish South America (New York: Cam-
bridge University Press, 1991), 509.
50
Jonathan Hartlyn, “The Trujillo Regime in the Dominican Republic,” in Chehabi and Linz (fn.
23), 91.
51
Pons (fn. 49), 510, 512, 521; and Crassweller (fn. 32), 99–100.
52
“Guarding the Heir,” Time, February 17, 1958, http://www.time.com/time/printout/0,8816,
862920,00.html (accessed May 17, 2007).
53
“Ramfis in Power,” Time, June 16, 1961, http://www.time.com/time/magazine/article/0,9171
,895364,00.html (accessed May 17, 2007).
h e r e d i t a r y s u c c e s s i o n 613
pressure and oversight by the United States, to move to a provisional
government that represented a clear break with Trujillo.”
54
Crucial to
this transition was the imminent threat of U.S. invasion—underscored
by American military ships looming off the coast of Santo Domingo—
which on November 18 compelled Ramfis Trujillo to resign and de-
part.
55
Ramfis’s exit was followed by the rise of theretofore nominal
president Joaquín Balaguer.
56
In regards to succession, elites in Taiwan, Azerbaijan, Singapore, and
Togo operated in a political context analogous to that of their coun-
terparts in the Dominican Republic—albeit with the absence of U.S.
pressure after the succession took place. As in Trujillo’s regime, top
officials in the regimes of Chiang Kai-shek, Heydar Aliyev, Lee Kuan
Yew, and Gnassingbé Eyadéma did not have an institutional precedent
for selecting a successor from their ranks and, when the time came,
they deferred to the ruler’s choice of heir apparent. Chiang Kai-shek’s
54
Hartlyn (fn. 50), 103.
55
In his discussion of hereditary succession Tullock asks, “Why did it take the direct intervention
by the American government after the death of Trujillo to prevent his son from remaining in control
of the family hacienda?” Tullock (fn. 10), 163.
56
Linda Greenhouse, “A Colonel at 6,” New York Times, December 29, 1969, 29.
ta b l e 3
su c c e s s i o n pr a c t i c e s a n d in s t i t u t i o n a l hi s t o r y
Succession Outcomes Ruler Predates Party Party Predates Ruler
Hereditary succession Rafael Trujillo (1961) Anastasio Somoza García (1956)
(year of ruler’s exit) Chiang Kai-shek (1975) Kim Il-sung (1994)
Heydar Aliyev (2003) Hafez al-Assad (2000)
Lee Kuan Yew (2004)
Gnassingbé Eyadéma (2005)
Abortive hereditary Alfredo Stroessner (1989)
succession Todor Zhivkov (1989)
(year of ruler’s exit)
Nonhereditary Julius Nyerere (1985) Joseph Stalin (1953)
succession Ayatollah Khomeini (1989) Leonid Brezhnev (1982)
(year of ruler’s exit) Enver Hoxha (1985)
Siaka Stevens (1985)
Chiang Ching-kuo (1988)
Deng Xiaoping (1992)
Kaysone Phomvihane (1992)
Mahathir Mohamed (2003)
Jiang Zemin (2004)
Joaquim Chissano (2005)
Portion of hereditary
succession in column 5 of 7 (71.4 percent) 3 of 15 (20 percent)
614 w o r l d p o l i t i c s
rout from mainland China in 1949 concluded the country’s civil war
and placed the defeated Chiang in control over the new Taiwanese
state; Chiang controlled the ruling Kuomintang (
k M t ), which exer-
cised “total political dominance” over the island’s population.
57
This
control persisted through Chiang’s lifetime. By the time he reached
his mid-seventies, the unrivaled premier had begun preparing to be-
queath this power to his son. Chiang Ching-kuo was made minister
of defense in 1965, clarifying his status as frontrunner to succeed his
father.
58
The younger Chiang’s influence grew steadily; in 1972 he was
appointed prime minister and later that year, when his father fell ill, he
became de facto chief executive.
59
When Chiang Kai-shek died at the
age of eighty-seven on April 5, 1975, Chiang Ching-kuo continued to
rule the country, notwithstanding Vice President C. K. Yen’s formal
assumption of the presidency.
60
After an interim period in which he
governed from the premiership, Chiang Ching-kuo was nominated by
the
k M t for president and elected in 1978.
61
Heydar Aliyev founded the New Azerbaijan Party in 1993 as he
consolidated his grasp on the presidency.
62
By 1999, the septuagenarian
autocrat was staging his son Ilham’s eventual succession.
63
Maneuvers
to manipulate the constitution and provide a legal veneer for the dy-
nastic handover continued through the elder Aliyev’s final months, cul-
minating in Ilham’s dubious election to the presidency on October 15,
2003.
64
Less than two months later Heydar Aliyev passed away with
Ilham settled in his new post.
65
Lee Kuan Yew founded the People’s Action Party and led Singapore
into independence, first from the British and then from Malaysia in 1965.
66
57
Stephan Haggard and Robert R. Kaufman, The Political Economy of Democratic Transitions
(Princeton: Princeton University Press, 1995), 276–77.
58
“Taiwan’s No. 2 Man: Chiang Ching-kuo,” New York Times, January 14, 1965, 4.
59
Tillman Durdin, “Chiang’s Role Shrinking,” New York Times, March 22, 1971, 13; “Nationalist
China’s New Premier: Chiang Ching-kuo,” New York Times, May 27, 1972, 4; and “Premier in Taipei
Running Regime,” New York Times, April 1, 1973, 6.
60
“Chiang Kai-shek is Dead in Taipei at 87,” New York Times, April 6, 1975, 1, 47.
61
Ian Buruma, “Reclaiming Taiwan,” New York Times, September 18, 1988, http://global.factiva
.com (accessed December 23, 2006); and “History of Taiwan Ruling Nationalist Party,” Reuters News,
March 17, 2000, http://global.factiva.com (accessed December 23, 2006).
62
“Azerbaijan: Political Forces,” Economist Intelligence Unit: ViewsWire, June 26, 2006, http://
global.factiva.com (accessed December 23, 2006).
63
David Stern, “Dynasty in Making as Aliyev’s Son Charts Path to Azeri Presidency,” Agence
Franc-Presse, December 22, 1999, http://global.factiva.com (accessed May 17, 2007).
64
Susan B. Glasser, “President’s Son Sweeps Election in Azerbaijan,” Washington Post, October
16, 2003, http://global.factiva.com (accessed June 21, 2007); and Guy Dinmore, “U.S. Stands by Its
Decision on Azeri Poll Result,” Financial Times, October 23, 2003, http://global.factiva.com (accessed
June 21, 2007).
65
“Azerbaijan: Political Forces” (fn. 62).
66
Yong Mun Cheong, “The Political Structures of the Independent States,” in Nicholas Tarling,
ed., The Cambridge History of Southeast Asia, vol. 2, The Nineteenth and Twentieth Centuries (Cambridge:
h e r e d i t a r y s u c c e s s i o n 615
In the 1980s he began positioning his son Lee Hsien Loong to suc-
ceed him. The younger Lee was elected to parliament in 1984, became
Minister of Trade and Industry in 1986, and rose to Deputy Prime
Minister in 1990. At the time, many Singaporeans correctly suspected
that “Lee [Kuan Yew] was, and would remain, the government’s domi-
nant force.” The Lees’ continued influence overshadowed the formal
premiership of Goh Chok Tong (1990–2004). In those years, Lee
Hsien Loong was effectively premier-in-waiting while Goh Chok
Tong served as premier and Lee Kuan Yew oversaw both men from
the newly created post of senior minister.
67
On August 12, 2004, Lee
Hsien Loong was sworn in as prime minister. Much like the handover
from Chiang Kai-shek in the early 1970s, this “pre-mortem” transition
ensured stability even as Lee Kuan Yew, then in his eighties, remained
elder statesman.
68
Finally, General Gnassingbé Eyadéma (then a colonel), seized
power in Togo on January 13, 1967, and formally became president
on April 14.
69
In 1969 he formed the Rally of the Togolese People,
which remained the only permitted party until 1991.
70
After surviving
the tumultuous 1990s, Eyadéma began grooming his son Faure Gnass-
ingbé for the presidency.
71
Faure was elected to parliament unopposed
in 2002; the following year he was the presumptive beneficiary of a
constitutional change—pushed through by Eyadéma—that lowered
the minimum age for the presidency from forty-five to thirty-five.
72
In July 2003 Faure joined his father’s cabinet and acquired the min-
istry overseeing Togo’s lucrative phosphate mines.
73
Despite the overt
Cambridge University Press, 1993), 412, 414, 450–55, 457; Michael Leifer, Dictionary of the Modern
Politics of Southeast Asia, 3rd ed. (London: Routledge, 2001),164–65; and Andrew T. H. Tan, A Political
and Economic Dictionary of South-East Asia (London: Europa Publications, 2004), 159–160.
67
Michael Haas, “A Political History,” in Michael Haas, ed., The Singapore Puzzle (Westport,
Conn.: Praeger, 1999), 26; Vasil (fn. 16), 135; Chris Lydgate, Lee’s Law: How Singapore Crushes Dissent
(Melbourne: Scribe Publications, 2003), 222.
68
John Burton, “Singapore’s New Prime Minister Faces Tough Task,” Financial Times, August
12, 2004, http://global.factiva.com (accessed June 22, 2007); and “World in Brief,” Washington Post,
August 13, 2004, Dow Jones & Reuters, http://global.factiva.com (accessed June 22, 2007).
69
Harvey Glickman, ed., Political Leaders of Contemporary Africa South of the Sahara: A Biographical
Dictionary (Westport, Conn.: Greenwood Press, 1992), 76
70
Yomi Durotoye, “Republic of Togo,” in George E. Delury, ed., World Encyclopedia of Political
Systems and Parties, 3rd ed., (New York: Facts on File, 1999), 3:1110–11.
71
Jacques Lhuillery, “Eyadema: Africa’s Longest Serving Ruler Who Evokes Strong Passions,”
Agence France-Presse, May 29, 2003, http://global.factiva.com (accessed December 23, 2006).
72
“Ruling Party ‘Assured of Winning the Majority of Seats’ in Elections,” BBC Monitoring Africa,
October 22, 2002, http://global.factiva.com (accessed May 14, 2007); and “Assembly Revises Consti-
tution … To Give More Terms for Presidents,” All Africa, January 8, 2003, http://global.factiva.com
(accessed May 14, 2007).
73
“Togo’s President Gives Son Ministry in New Cabinet,” Reuters News, July 30, 2003, http://
global.factiva.com (accessed May 14, 2007).
616 w o r l d p o l i t i c s
expansion of Faure’s portfolio, President Eyadéma remained coy about
his succession plans, all the while garnering the tacit support of other
possible contenders in the regime.
74
When Eyadéma died at the age of
sixty-seven while out of the country for urgent medical care on Febru-
ary 5, 2005, leaders of the Togolese armed forces immediately declared
Faure the country’s new president.
75
In the following weeks the ruling
elite around Faure cemented his hold on power through a series of legal
maneuvers and a legitimating election, and thereby lowered the risk of
a chaotic internal conflict.
76
Throughout these five cases, the leader’s departure posed an un-
precedented challenge to the regime’s continuation. Consistent with
Tullock’s expectation, the ruler’s grooming of his son as heir appar-
ent was actively facilitated by the surrounding elite. Particularly in the
Dominican Republic, Azerbaijan, and Togo, relatively young and inex-
perienced leaders with little to no independent institutional base were
supported by nonfamilial elites due to the implicit security the heir’s
presence offered these top officials.
h
e r e d i t a r y su c c e s s i o n : pa r t y pr e d a t e s ru l e r
In most of the regimes where rulers came to power via parties, those
practices constrained the subsequent successions. Nicaragua, North
Korea (Democratic People’s Republic of Korea), and Syria are three
exceptions. Rulers in these countries subordinated the preexisting party
to their personal authority and successfully promoted their sons as heirs
apparent. Surrounding nonfamilial elites did not obstruct hereditary
succession and in Syria they clearly regarded the ruler’s designee as the
optimal course.
Anastasio Somoza García climbed to the presidency of Nicaragua
through a stilted election in 1936 in which he was jointly nominated by
the country’s historically leading parties, the Liberal Party and the Con-
servative Party.
77
Competition between the Liberal and Conservative
parties had characterized Nicaraguan politics since the country gained
independence in 1821.
78
Under the custodianship of a U.S. military
intervention, Liberals were in power in the late 1920s. In 1932 newly
elected Liberal President Juan Bautista Sacasa grudgingly appointed
74
“Has Eyadema Now Found a Successor?” All Africa, August 14, 2003, http://global.factiva.com
(accessed May 14, 2007).
75
“Togo Parliament Leader Unable to Return Home after President Dies,” Agence France-Presse,
February 5, 2005, http://global.factiva.com (accessed May 14, 2007).
76
“Togo: Gnassingbe Digging in as the New Front Man for Togo’s Long-Ruling Elite,” All Africa,
February 21, 2005, http://global.factiva.com (accessed May 14, 2007).
77
John A. Booth, “Republic of Nicaragua,” in Delury (fn. 70), 2:802.
78
Arthur S. Banks, Political Handbook of the World: 1975 (New York: McGraw-Hill, 1975), 246.
h e r e d i t a r y s u c c e s s i o n 617
Somoza chief director of the National Guard. Four year later Somoza
marginalized Sacasa and positioned himself for a carefully engineered
presidential bid.
79
In June the Liberal Party nominated Somoza for
the presidency, which he won six months later in a race boycotted by
the opposition. As the country’s traditional parties fell under Somoza’s
sway, they were renamed the Liberal National Party and Conservative
National Party.
80
The Liberal National Party was to be Somoza’s titu-
lar political organization. Upon taking office in January 1937, Somoza
merged the positions of president and National Guard chief director.
He went on to rule for nearly two decades, actively grooming his sons
for succession during the later years of his tenure. In 1956 Somoza ap-
pointed his eldest son Luis Somoza Debayle, First Designate (heir ap-
parent), and his younger son Anastasio Somoza Debayle, acting chief
director of the National Guard. When Samoza was assassinated in
September 1956, the sons were prepared to carry on for him. At that
point, the Liberal National Party and National Guard supported the
Somoza brothers in maintaining the dictatorship against the country’s
enfeebled but persistent opposition movements.
81
Kim Il-sung’s dictatorship in North Korea represents the sole case of
hereditary succession despite the presence of a communist party. Kim
had not founded the Workers’ Party of Korea. Rather, he and the party
were installed by occupying Soviet forces after they invaded the Ko-
rean peninsula in 1945.
82
Given this background, it would have been
conventional for Kim’s successor to emerge through the party organi-
zation. Instead, Kim infused the party with his relatives while brutally
eliminating his rivals. As Mark Suh writes, these purges yielded an
unusually narrow elite for a communist party regime. “By 1961, Kim
Il-sung’s absolute power position was consolidated, and by 1970 all
the politburo members were identified as belonging to his faction or
family.”
83
Relatives as distant as Kim’s great-grandfather were idolized
and even his cousins, nieces, and nephews occupied influential party
posts.
84
This extraordinary imposition of family over party presents its
79
Victor Bulmer-Thomas. “Nicaragua since 1930,” in Leslie Bethell, ed., The Cambridge History of
Latin America, vol. 7, Latin America since 1930: Mexico, Central America, and the Caribbean (New York:
Cambridge University Press, 1991), 331–34.
80
Richard Millett, Guardians of the Dynasty (Maryknoll, N.Y.: Orbis, 1977), 181–82; and John A.
Booth, “The Somoza Regime in Nicaragua,” in Chehabi and Linz (fn. 23), 132–33.
81
Bulmer-Thomas (fn. 79), 342–44.
82
Mark B. Suh, “Korea (Democratic People’s Republic/North Korea),” in Nohlen, Grotz, and
Hartmann (fn. 6), 2:395.
83
Ibid., 396.
84
Tai Sung An, North Korea in Transition: From Dictatorship to Dynasty (Westport, Conn.: Green-
wood Press, 1983), 137, 141.
618 w o r l d p o l i t i c s
own lessons for the political process of hereditary succession. Heredi-
tary succession was predicated on the thorough expunging of tradi-
tional political elites who could have opposed Kim’s cult of personality.
President Kim designated his son Kim Jong-il as heir apparent in 1980.
These plans drew the opprobrium of other communist nations and
strained Kim’s own erratic fealty to communist ideology.
85
(In 1973 the
regime had conspicuously removed the term “hereditary succession”
from the state’s official Dictionary of Political Terminologies.
86
) When
the elder Kim passed away in 1994 at eighty-two, his son’s succession
marked the culmination of a plan begun much earlier.
In Syria, Hafez al-Assad did not eliminate the Baath Party after he
seized power, but he ruled over the party as much as through it. By 1991
Assad had outlasted most of his domestic opponents and was far along
in preparing his elder son Basil to succeed him.
87
Yet this plan collapsed
when Basil perished abruptly in an automobile accident in 1994. The
mantle of president-in-waiting then passed to Bashar, the president’s
next eldest son and a previously innocuous figure. Bashar’s ascent did
not evoke resistance from top Baathists and was instead abetted by the
surrounding elite when Hafez al-Assad’s health declined in the later
1990s. Baath Party leaders have described their support for Bashar’s
succession—his father’s clear wish—as a regime-preserving choice that
reduced the risk of internal feuding. As Eyal Zisser recounts: “Syria’s
senior leadership, aware of Hafiz al-Assad’s impending demise, wished
to deflect any possible threat, domestic or external, to the existence and
stability of the regime. Bashar remained the best possible option.”
88
In Nicaragua, North Korea, and Syria, rulers presided over hereditary
successions without triggering a backlash from the party organizations
that predated their tenure. As the contrasting cases of nonhereditary
succession illustrate, such a course is unusual.
a
b o r t i V e he r e d i t a r y su c c e s s i o n : pa r t y pr e d a t e s ru l e r
Where rulers predated parties and groomed their sons, all five were
successfully followed by their filial heirs apparent. By contrast, two out
of the five autocrats who attempted hereditary succession in parties
that predated their tenure were unsuccessful. Rather than support the
85
Ibid., 4, 155.
86
The prior edition called hereditary succession “a reactionary custom of exploitative societies . . .”
Ibid., 150–51.
87
Chris Hedges, “Fist May Be of Iron, but Is Assad’s Hand Weak?” New York Times, December
17, 1991, http://global.factiva.com (accessed June 27, 2007); and Caryle Murphy, Washington Post,
“New High Profile for Assad’s Son Suggests Heir Apparency,” January 24, 1992, http://global.factiva
.com (accessed June 27, 2007).
88
Eyal Zisser, “Does Bashar al-Assad Rule Syria?” Middle East Quarterly 10 (Winter 2003).
h e r e d i t a r y s u c c e s s i o n 619
son’s promotion to the top post, elites obstructed hereditary succes-
sion and rallied against the incumbent. Elite dissent in these two cases,
Paraguay under Stroessner and Bulgaria under Zhivkov, bridges the
modal cases of hereditary succession (where rulers predated parties)
and nonhereditary succession (where parties predated rulers). Rulers
in this set sought to circumvent prior institutional practices but they
lacked the command over the organization that Somoza, Kim, and As-
sad had achieved. Regime officials subverted the incumbents’ plans to
install their sons and oversaw nonhereditary succession instead.
Alfredo Stroessner came to power in Paraguay in May 1954 by forc-
ibly removing embattled president Federico Cháves and then receiv-
ing the nomination of the historically dominant Colorado Party, which
was founded in 1887. Stroessner was inaugurated president on August
15, 1954.
89
Eventually elected president seven times as the Colorado
standard-bearer, Stroessner made the party, the military, and the state
apparatus the pillars of his regime.
90
In 1984, with Stroessner in his
early seventies, the Colorado Party began discussing possible succes-
sors.
91
In a series of increasingly contentious party conferences, Colo-
rado traditionalists argued the next president should come from within
the party. Another faction, closer to Stroessner and less tightly wedded
to the party, advocated that Gustavo Stroessner, an air force officer,
succeed his father in the event of the dictator’s death.
92
By late 1988, it
was clear which course President Stroessner favored: Gustavo had been
promoted to the rank of colonel and the elder Stroessner contemplated
making him vice-president as well.
93
These preparations alienated the
Colorado Party’s leadership, which began to support the president’s dis-
affected associate, General Andrés Rodríguez. The backlash against an
incipient hereditary succession reached its climax when Rodríguez pro-
pelled Stroessner’s violent ouster on February 3, 1989.
94
The Colorado
89
Paul H. Lewis, “Paraguay since 1930,” in Bethell (fn. 49), 252.
90
Thomas G. Sanders, The Fall of Stroessner: Continuity and Change in Paraguay (unknown: Uni-
versities Field Staff International, 1989), 1; and Alain Rouquié, “The Military in Latin American
Politics since 1930,” in Leslie Bethell, ed., Latin America: Politics and Society Since 1930 (Cambridge:
Cambridge University Press, 1998), 199–200.
91
Carlos Miranda, The Stroessner Era: Authoritarian Rule in Paraguay (Boulder, Colo.: Westview
Press, 1990), 130–32.
92
Riordan Roett, “Paraguay after Stroessner,” Foreign Affairs 68 (Spring 1989), 124–42; Richard
Snyder, “Explaining Transitions from Neopatrimonial Dictatorships,” Comparative Politics 24 (July
1992), 391; and Lewis (fn. 89), 265.
93
Alfredo Seiferheld, “Stroessner, Hemisphere’s Most Durable Leader, Marks 33rd Anniversary,”
Associated Press, August 14, 1987, http://global.factiva.com (accessed April 22, 2007); “Government
Denies Stroessner Ailing,” Latin American Southern Cone Report, November 17, 1988, http://global
.factiva.com (accessed April 22, 2007); and “Stroessner Seeks Approval to Promote Son to Colonel,”
Reuters News, December 14, 1988, http://global.factiva.com (accessed April 22, 2007).
94
Alan Riding, “Just a Question of Whose Snout is in the Trough,” New York Times, February 6,
1989, http://global.factiva.com (accessed April 22, 2007).
620 w o r l d p o l i t i c s
Party then nominated Rodríguez as its presidential candidate and he
was overwhelmingly elected in the voting that quickly followed.
95
Al-
though this leadership change marked the end of the Stroessner era, it
perpetuated the Colorado Party’s influence in Paraguayan politics. The
country’s subsequent four presidents have all come from this party.
96
The portents of a similar internal clash appeared earlier in Bulgaria,
where Todor Zhivkov ruled for thirty-five years as first secretary of the
Central Committee of the Bulgarian Communist Party (
b c p ). Founded
in 1919, the
b c p had been Bulgaria’s ruling party ever since the country’s
“people’s republic” had been established under Soviet auspices in 1947.
97
Zhivkov came to power in 1954 from within the
b c p ’s hierarchy, fol-
lowing the course of his two predecessors who also headed the Central
Committee.
98
Over the subsequent decades, he consolidated his hold
and, for a communist party chief, became conspicuously nepotistic, a
tendency that was poorly received by his fellow communists. After his
wife died in 1971, Zhivkov’s daughter Lyudmila assumed the role of
Bulgaria’s first lady. Lyudmila Zhivkova was appointed by her father to
be head of national culture in 1975; she joined the party central com-
mittee that same year and became a full member of the
b c p ’s politburo
in 1979, leading outside observers to dub her “the second most power-
ful person in the country.”
99
In July 1981, before the thirty-eight-year-
old Zhivkova could be further groomed to succeed her father, she died
suddenly under circumstances that remain disputed—suspected causes
of death range from suicide to politically motivated assassination.
100
After Zhivkov’s aspirations for his daughter were cut short, the ag-
ing ruler began prodding his younger son Vladimir along the path trod
by Lyudmila.
101
In July 1989, Vladimir, thirty-six years old at the time,
was brought into the
b c p ’s central committee to head a department of
cultural affairs.
102
A mere four months later, Vladimir’s rise was cur-
95
Lewis (fn. 89), 264–66.
96
“State Department Issues Background Note on Paraguay,” US Fed News, May 1, 2007, http://
global.factiva.com (accessed June 20, 2007).
97
Arthur S. Banks, Political Handbook of the World: 1975 (New York: McGraw-Hill, 1975), 42–43.
98
Carl Beck, William Jarzabek, and Paul H. Ernandez, Political Succession in Eastern Europe: Four-
teen Case Studies (Pittsburgh: University of Pittsburgh, 1976), 9–10.
99
“President’s Daughter Plays Major Role in Bulgaria,” New York Times, November 9, 1980,
http://global.factiva.com (accessed June 20, 2007); “Obituary: Lyudmila Zhivkova, Bulgaria Culture
Aide,” New York Times, July 22, 1981, http://global.factiva.com (accessed June 20, 2007); and John D.
Bell, The Bulgarian Communist Party from Blagoev to Zhivkov (Palo Alto, Calif.: Hoover Institution,
1986), 128.
100
Michael Simmons, “Country Debates the Sudden Death of Former Leader’s Daughter,” Guard-
ian, July 24, 1991, http://global.factiva.com (accessed June 20, 2007).
101
Bell (fn. 99), 129.
102
“Zhivkov’s Son Wins Promotion,” Financial Times, July 10, 1989, http://global.factiva.com (ac-
cessed May 31, 2007).
h e r e d i t a r y s u c c e s s i o n 621
tailed when long-time foreign minister Petar Mladenov deposed Todor
Zhivkov.
103
Within a week of the elder Zhivkov’s “resignation,” he,
his son, and their closest loyalists were expelled from the
b c p ’s central
decision-making bodies.
104
Much like the situation in Paraguay, non-
hereditary succession had been promoted by the ruling party whose
ranking members strove to preserve the organization’s influence. Fol-
lowing Zhivkov’s removal, the
b c p ’s rechristened itself the Bulgarian
Socialist Party and went on to take a healthy majority in the elections
of June 1990. Mladenov himself resigned shortly afterward under pub-
lic pressure.
105
Meanwhile, Zhivkov was placed under house arrest for
crimes committed during his tenure. He was freed in 1997 and died
the following year.
106
The nonhereditary transitions overseen by Paraguay’s Colorado Party
and the Bulgarian Communist Party shed light on the conditions that ob-
struct dynasticism. In both countries, powerful and long-ruling autocrats
were not only foiled in their attempts to install their sons, they were
themselves ousted, in large part because of their succession schemes.
Party elites in both regimes enjoyed a long experience in choosing suc-
cessors from within the party and they did not view hereditary succes-
sion as the optimal route for preserving their status. When Stroessner
and Zhivkov positioned their sons as heirs apparent they departed from
the organizations’ prior practices and drew the ire of party elites.
n
o n h e r e d i t a r y su c c e s s i o n : ru l e r pr e d a t e s pa r t y
Among rulers who preceded party institutions, nonhereditary succes-
sion occurred in only two cases, Tanzania in 1985 and Iran in 1979.
In Tanzania, Julius Nyerere was the founder and undisputed leader of
the Chama Cha Mapinduzi (
c c M ) or “Revolutionary Party.” Originally
103
Mary Bhattia, “Bulgaria’s Zhivkov Quits After 35 Years; Foreign Minister, 53, Replaces East
Bloc’s Longest-Serving Leader,” Washington Post, November 11, 1989, http://global.factiva.com (ac-
cessed June 20, 2007); Judy Dempsey, “Bulgaria’s Hardline Leader Quits,” Financial Times, November
11, 1989, http://global.factiva.com (accessed June 20, 2007); and Rada Nikolaev, “A Year of Crucial
Change in Bulgaria,” Report on Eastern Europe 1 (January 5, 1990), 7.
104
“Party Plenum Makes Changes in Politburo and Central Committee,” BBC Monitoring Service,
November 18, 1989, http://global.factiva.com (accessed June 20, 2007).
105
“New Bulgarian Leader Says He’s for Free Elections,”San Francisco Chronicle, November 18,
1989, http://global.factiva.com (accessed June 20, 2007); “Thousands Demand End to Monopoly of
Power,” Times, December 11, 1989, http://global.factiva.com (accessed April 30, 2007); “New Lead-
ers Hold Faith with Communism,” Independent–London, December 14, 1989, http://global.factiva
.com (accessed April 30, 2007); Rada Nikolaev, “Political Maneuvering before the Roundtable Talks,”
Report on Eastern Europe 1 ( January 26, 1990), 6–7; and Andrei Pantev, “The Historic Road of the
Third Bulgarian State,” in Iliana Zloch-Christy, ed., Bulgaria in a Time of Change: Economic and Politi-
cal Dimensions (Brookfield, Vt.: Avebury, 1996), 20–21.
106
Vessela Sergueva, “Todor Zhivkov, Last of Eastern Europe’s Communist Dinosaurs,” Agence
France-Presse, August 6, 1998, http://global.factiva.com (accessed April 30, 2007).
622 w o r l d p o l i t i c s
created by Nyerere as the Tanganyika National Union in July 1954, the
party spearheaded Tanzania’s campaign for independence from Great
Britain, which was achieved on December 9, 1961.
107
For over two de-
cades in power Nyerere demonstrated his devotion to social welfare and
pluralism, leading one biographer to write: “Although not immune to the
authorization of force in politics. . . [Nyerere] called ordinary citizens to
their better instincts, toward nonracialism, cooperation, a code of moral
conduct, and social justice.”
108
This same pursuit of collective welfare
characterized Nyerere’s self-abnegating decision to begin democratiz-
ing Tanzania and to institutionalize the
c c M by naming Ali Hassan
Mwinyi his successor at a special party congress in August 1985.
109
He
then retired from political life, gradually receding from party activities
while continuing to practice the principles he had pursued in office.
110
Conditions surrounding nonhereditary succession in Iran were quite
different. As the spiritual leader of the revolution that ousted the shah,
Ayatollah Ruhollah Khomeini enjoyed a preeminent status upon his
return from exile on February 1, 1979. With Khomeini already in his
late seventies at that time, succession was a pressing concern in the
Islamic Republic of Iran’s first years. The country’s new constitution
created an eight-six-member Assembly of Experts to select a succes-
sor, but the Assembly deferred to Khomeini’s choice. Grand Ayatollah
Hossein Ali Montazeri was one of only fifteen or so grand ayatollahs in
Iran and well known to be Khomeini’s preferred heir. During Khomei-
ni’s lifetime, in 1985, the Assembly formally recognized Montazeri for
that role.
111
Due to his clerical credentials, Montazeri overshadowed
all other contenders for succession, including Khomeini’s son Ahmad,
a midranking cleric. Although Ahmad Khomeini was rumored to be a
107
Terry M. Mays, “United Republic of Tanzania,” in Delury (fn. 70), 3:1090–93.
108
Glickman (fn. 69), 213.
109
Glenn Frankel, “Nyerere Resignation to End 23-Year Era in East Africa: Tanzania Needs New
Leadership, He Says,” Washington Post, December 9, 1984, http://global.factiva.com (accessed Novem-
ber 15, 2006); and Hunter R. Clark, “Making a Graceful Exit; Nyerere’s Peaceful Departure Ends an Era
in African History,” Time, November 4, 1985, http://global.factiva.com (accessed November 15, 2006).
110
Samuel P. Huntington, The Third Wave: Democratization in the Late Twentieth Century (Nor-
man, Okla.: University of Oklahoma Press, 1991), 130.
111
Bernard Gwertzman, “U.S. Aide Says Iranian Authorities Seem Committed to Free Captives,”
New York Times, November 1, 1980, http://global.factiva.com (accessed May 22, 2007); John Kifner,
“Iran Says Its Troops Have Mounted Counteroffensive,” New York Times, January 6, 1981, http://
global.factiva.com (accessed May 22, 2007); “Montazeri Chosen to be Khomeini’s Successor,” As-
sociated Press, November 23, 1985, http://global.factiva.com (accessed May 22, 2007); and Wilfried
Buchta, Who Rules Iran? The Structure of Power in the Islamic Republic (Washington, D.C.: Washington
Institute for Near East Policy, 2000), 54, 92.
112
“The Son Also Rises,” Newsweek, June 9, 1980; “Khomeini’s Regime: The Squabbling Intensi-
fies,” Newsweek, November 17, 1986; “Iran; Rafsanjani Dons His Wolf’s Outfit,” Economist, April
29, 1989; and “The Day of Infinite Sadness,” Economist, June 10, 1989, http://global.factiva.com (ac-
cessed December 18, 2006).
h e r e d i t a r y s u c c e s s i o n 623
possible successor after his father’s death, he was never heir apparent.
112
Rather, Montazeri remained Khomeini’s formally designated successor
until March 1989, at which point a falling out between the two clerics
cost Montazeri his status. When Khomeini passed away less than three
months later, the Assembly of Experts selected President Ali Khame-
nei as the country’s new supreme leader.
113
The orderly assumption of office by Mwinyi in Tanzania and
Khamenei in Iran occurred in dissimilar contexts, the first under a
ruler who intentionally pursued gradual democratization, the second
within an Islamic republic where religious credentials were supposedly
the hallmark of authority. Notably in both cases, the ruler’s choice of
a nonhereditary appointment was accepted by the surrounding elites.
Thus, while Nyerere and Khomeini did not chart a course to heredi-
tary succession, the path of succession they stipulated was followed by
regime leaders. The rarity of founding rulers intentionally inaugurating
nonhereditary succession stands in relief against the general trend, in
which rulers pass power to sons unless constrained by antecedent in-
stitutional practices. The final subset displays this pattern in ten cases
where parties predated rulers.
n
o n h e r e d i t a r y su c c e s s i o n : pa r t y pr e d a t e s ru l e r
Rulers who reach their posts through existing institutions establish or
further engrain the organizational practices for selecting subsequent
leaders. In such contexts elites have experience with a prior succession
and the ruler’s departure does not presage an unregulated power strug-
gle. To the contrary, intrafamilial succession would be a path-breaking
departure from earlier practices where the ruler comes from the party
and the party’s members look among themselves for his successor. This
logic fits the majority of nonhereditary succession cases, six of which
(in addition to Bulgaria) took place in communist regimes.
The communist leadership transitions that followed Joseph Stalin,
Leonid Brezhnev, Enver Hoxha, Deng Xiaoping, Kaysone Phomvi-
hane, and Jiang Zemin evince a general trend by which rulers’ political
heirs come from within the conventional ranks of the party hierarchy.
Potential dynasts in these cases were succeeded by party members who
113
“President Backs Ousting of Khomeini’s Successor,” Reuters News, March 31, 1989, http://
global.factiva.com (accessed May 22, 2007); Kamran Fazel and Andrew Gowers, “Iranian Left Clips
the Pragmatists’ Wings,” Financial Times, April 20, 1989, http://global.factiva.com (accessed May 22,
2007); Mohammad Zargham, “Khomeini Leaves Iran a Nation Groping for Direction,” Reuters News,
June 4, 1989, http://global.factiva.com (accessed May 22, 2007); and Bahman Baktiari, Parliamentary
Politics in Revolutionary Iran: The Institutionalization of Factional Politics (Gainesville, Fla.: University
Press of Florida, 1996), 171–74.
624 w o r l d p o l i t i c s
were not their sons. There is evidence that Stalin and Brezhnev at-
tempted to promote their sons and that Jiang Zemin tried to protect his
son’s business interests, but none of their offspring came within reach
of the Soviet or Chinese premiership.
In the Soviet Union, the party
chose the successors of Stalin and Brezhnev.
114
In the People’s Republic
of China, Deng and Jiang cultivated heirs apparent from within the
party and steadily ceded power to them: Deng to Jiang, Jiang to Hu
Jintao.
115
In Albania, Hoxha groomed neither of his sons and during
his final years passed power to his premier and designated successor
Ramiz Alia.
116
Finally, among the cases of nonhereditary succession
in communist regimes, in Laos in 1992 Kaysone Phomvihane was fol-
lowed by long-time heir presumptive General Khamtai Siphandon as
leader of the Lao People’s Revolutionary Party, an outgrowth of Ho
Chi Minh’s Communist Party of Vietnam.
117
Although communist parties are well known for their organizational
resilience, there are several noncommunist cases where institutions
structured leadership changes. In Sierra Leone, Taiwan, Malaysia, and
Mozambique, rulers with viable adult heirs passed power via the same
organizations through which they had reached their offices. The first
of these, Siaka Probyn Stevens, led the All People’s Congress (
a p c )
of Sierra Leone to an electoral victory in March 1967 in which the
party won thirty-two of sixty-six seats and subsequently chose Stevens
114
Helen Rapaport, Josef Stalin: A Biographical Companion (Santa Barbara, Calif.: ABC-Clio,
1999); “Leonid Brezhnev Presents Awards to His Sons,” Xinhua General Overseas News Service, May
12, 1979, http://global.factiva.com (accessed December 18, 2006); and “Brezhnev’s Son Retired from
Government Trade Post,” United Press International, August 8, 1986, http://global.factiva.com (ac-
cessed December 18, 2006).
115
Louise Branson, “Deng’s Third Successor in Three Years Faces Uncertain Future,” Strait Times,
October 23, 1989, http://global.factiva.com (accessed May 17, 2007); Nicholas D. Kristof, “China
Whispers of Plots and Man Called Yang,” New York Times, October 24, 1989, http://global.factiva
.com (accessed May 17, 2007); Jim Abrams, “Communist Party Plenum Meets to Set Economic
Course,” Associated Press, November 6, 1989, http://global.factiva.com (accessed May 17, 2007);
Louise Branson, “Deng Making Way for Jian ‘Like the Time Mao Anointed Hua,’” Sunday Times,
November 12, 1989, http://global.factiva.com (accessed May 17, 2007); Damien McElroy, “Anxious
to Protect Business Interests, China’s Leaders Delay Handover,” Daily Telegraph, August 17, 2002,
http://global.factiva.com (accessed May 17, 2007); and Geoffrey York, “A New Star in the East,” Globe
and Mail, November 2, 2002, http://global.factiva.com (accessed May 17, 2007).
116
“Enver Hoxha Dies; Albanian Leader,” New York Times, April 12, 1985; and “Dictator’s Sons
Are Bitter, Idle, Unrepentant,” Associated Press, January 16, 1993, http://global.factiva.com (accessed
December 18, 2006).
117
Alan Dawson, “Aging Laotian Leaders Likely to Step Aside,” Bangkok Post, March 27, 1989,
http://global.factiva.com (accessed May 31, 2007); Andrew Sherry, “Kaysone’s Death Leaves Power
Vacuum in Laos,” Agence France-Presse, November 22, 1992, http://global.factiva.com (accessed May
31, 2007); Yong Mun Cheong, “The Political Structures of the Independent States,” in Tarling (fn.
66), 406; C. M. Turnbull, “Regionalism and Nationalism,” in Tarling (fn. 66), 624–26; and Michael
Leifer, Dictionary of the Modern Politics of Southeast Asia, 3rd ed., (London: Routledge, 2001), 91,
149–50, 161–62, 215, 253.
118
Glickman (fn. 69), 275–57; and Steven Metz, “Republic of Sierra Leone,” in Delury (fn. 70),
3:981–82.
h e r e d i t a r y s u c c e s s i o n 625
as premier.
118
Stevens ruled Sierra Leone for nearly two decades and
transformed the country into a presidential system and a single-party
state. In 1985, at eighty years old, Stevens stepped down from office
and the
a p c nominated Joseph Saidu Momoh, whom Stevens had been
preparing for succession during the prior four years, as its presidential
nominee.
119
As with the ruling party in Sierra Leone, Taiwan’s Kuom-
intang provided the vehicle for the country’s leadership succession in
the 1980s. Years before his death, Chiang Ching-kuo foreswore the
option of hereditary succession, embraced the mantle of institutional
reformer, and designated Lee Teng-hui as vice-president and official
successor in the event of his death.
120
When Chiang passed away in
January 1988 at the age of seventy-seven, Lee was immediately sworn
in as president of Taiwan. That summer he was made
k M t chairman at
the party’s congress.
121
It is noteworthy that in Malaysia Mahathir did not replicate Singa-
porean Lee Kuan Yew’s model, but instead ceded power to his deputy
in the ruling United Malays National Organization (
u M n o ).
122
Ma-
hathir’s rise to the premiership in 1981 had been Malaysia’s third non-
hereditary succession and the third such transition within
u M n o .
123
Although party leaders enjoyed great discretion in facilitating or ob-
structing would-be successors, no
u M n o chief had ever designated his
son. Despite his arbitrariness as premier, Mahathir operated within
this tradition and ultimately continued it. In 1999, he expelled his
long-time deputy and heir apparent Anwar Ibrahim from
u M n o and
oversaw Anwar’s eventual imprisonment through a widely discredited
119
“The All People’s Congress, the Only Party in Sierra Leone, Will Meet Today to Select a Presi-
dent to Succeed Mr. Siaka Stevens,” Times, August 1, 1985, http://global.factiva.com (accessed May
22, 2007); and “Tomorrow Sierra Leone Is to Elect a New Leader to Succeed President Siaka Stevens,
the 80-Year Old Who has Held Power for 17 Years,” Financial Times, September 30, 1985, http://
global.factiva.com (accessed May 22, 2007).
120
Fox Butterfield, “Taiwan: A New Sense of Confidence,” New York Times, May 5, 1984, http://
global.factiva.com (accessed June 19, 2007); “Chiang Rules Out Family Succession, Military Rule
for Taiwan,” Associated Press, December 25, 1985, http://global.factiva.com (accessed November 1,
2006); “Collective Leadership Likely in Taiwan After Chiang,” Wall Street Journal, December 31,
1985, http://global.factiva.com (accessed November 1, 2006); and “Obituary of Chiang Ching-kuo,
Taiwan’s Popular Reformer,” Financial Times, January 14, 1988, http://global.factiva.com (accessed
June 19, 2007).
121
James McGregor and Adi Ignatius, “Taiwan President’s Death Intensifies Political Uncertain-
ties,” Wall Street Journal, January 14, 1988, http://global.factiva.com (accessed June 19, 2007); and
Andrew Browne, “Lee Teng-hui Voted Taiwan Nationalist Party Chairman,” Reuters News, July 8,
1988, http://global.factiva.com (accessed June 19, 2007).
122
On Mahathir’s dominance, see William F. Case, “Can the Halfway House Stand? Semidemoc-
racy and Elite Theory in Three Southeast Asian Countries,” Comparative Politics 28 ( July 1996); and
Dan Slater, “Iron Cage in an Iron Fist: Authoritarian Institutions and the Personalization of Power in
Malaysia,” Comparative Politics 36 (October 2003).
123
Harold Crouch, Government and Society in Malaysia (Ithaca, N.Y.: Cornell University Press,
1996), 98–104.
626 w o r l d p o l i t i c s
judicial process. Mahathir then promoted a lesser-known party leader,
Abdullah Badawi in Anwar’s stead. In the fall of 2003, at seventy-seven
years old Mahathir relinquished his roles as premier and party presi-
dent, ceding the political stage to Abdullah.
124
In Mozambique, Joaquim Chissano’s entry and exit bore semblance
to Mahathir’s trajectory within
u M n o . Chissano had been a high-rank-
ing member of the Liberation Front of Mozambique (
f r e l i M o ) and
was chosen by the party to succeed President Samora Machel upon his
death on October 9, 1986.
125
Just as f r e l i M o ’s leaders selected Chis-
sano, they choose his successor by tapping a new presidential candidate
from the party in fall 2004.
126
As in Malaysia, succession came from
the party, rather than from the ruler’s family.
These ten cases illustrate the conditions under which rulers are most
likely to pass over sons rather than pass power to them. Consistent
with Tullock’s theory, when elites have a precedent within the party
for designating the successor, their collective security does not depend
on the ruler’s choice. Under these conditions an incumbent’s pursuit of
hereditary succession may be perceived as a threatening aberration to
prior convention, pitting elites against the ruler as seen in Paraguay and
Bulgaria.
c
o M p a r a t i V e iM p l i c a t i o n s
There has been little cross-national research on hereditary succession.
Recent analyses tend to treat the process in isolation without contrast-
ing it to deliberate nonfamilial power transfers. That focus implies a
large field of successions in which rulers either elect to eschew dynasti-
cism or are compelled to do so by recalcitrant elites. A comparison of
hereditary successions and their relevant contrast cases shows why this
view should be revised. When there is no precedent for choosing a
leader from within the ruling organization, elites may support heredi-
tary succession as a method for preserving their influence and mitigat-
ing the uncertainty that surrounds an aging ruler and an impending
leadership change. In seventeen of the twenty-two cases of possible he-
reditary succession, outcomes closely matched variations in antecedent
124
John Burton, “UMNO to Support Mahathir Successor,” Financial Times, June 23, 2003, http://
global.factiva.com (accessed June 19, 2007); and Alan Sipress, “Malaysian Strongman Leaving Mixed
Legacy,” Washington Post, October 27, 2003, http://global.factiva.com (accessed June 19, 2007).
125
John A. Wiseman, Political Leaders in Black Africa: A Biographical Dictionary of the Major Politi-
cians since Independence (Brookfield, Vt.: Edward Elgar, 1991), 48–49; Glickman (fn. 69), 138–42;
Robert J. Griffiths, “Republic of Mozambique,” in Delury (fn. 70), 2:762–64; and Michael Krennerich,
“Mozambique,” in Nohlen, Krennerich, and Thibaut (fn. 11), 645.
126
“Mozambique Set for Landmark Polls to Choose Peacemaker Chissano’s Successor,” Agence
France-Presse November 30, 2004, http://global.factiva.com (accessed December 18, 2006).
h e r e d i t a r y s u c c e s s i o n 627
institutions: sons succeeded fathers when the ruler’s authority predated
the party’s. Of the seven cases where there was no precedent for choos-
ing a leader from within the ruling organization, five rulers embarked
on hereditary succession. In twelve of fifteen cases where rulers came
from a party, elites deferred to that organization to name the leader’s
successor. The data thus comport with the political explanation ven-
tured by Tullock. That same theory can help comparativists gauge the
likelihood that current autocrats will be replaced through hereditary
succession.
There are a number of rulers presently grooming sons for succession.
Three closely watched cases are Equatorial Guinea, Yemen, and Libya.
Notably, rulers in these cases are now in their mid-sixties and elites in
these countries have no prior experience managing succession through a
ruling party. In Equatorial Guinea, President Teodoro Obiang Nguema
took power in 1979 through a military coup and established the ruling
Equatorial Guinea Democratic Party in 1987.
127
President Ali Abdul-
lah Saleh ruled North Yemen beginning in 1978 and unified Yemen
after 1990, the year in which he created the still-dominant General
People’s Congress Party.
128
Based on the record considered above, Obi-
ang and Saleh can harbor substantial hope that their colleagues in the
regime will support their sons as successors when the time comes. Fi-
nally, in Muammar Qadhafi’s Libya there is no ruling party apparatus
through which regime leaders may expect an alternative to Qadhafi’s
third son and heir apparent, the London School of Economics-trained
Seif al-Islam al-Qadhafi.
129
Father-son transitions in these cases would
be consistent with the pattern of prior cases.
c
o n c l u s i o n : a ne w ge n e r a t i o n o f au t o c r a t s ?
Rulers around the world have undertaken hereditary successions that
augur not a new wave of democracy but a fresh generation of dictators.
What explains this phenomenon? Earlier scholarship posited that he-
127
“Obiang Dissolves Parliament to Pave Way for First Free Elections,” Associated Press Political
Service, July 3, 1993, http://global.factiva.com (accessed December 23, 2006); and Adrien Feniou,
“Ruling Party in Equatorial Guinea Re-Elects President to Its Helm,” Global Insight Daily Analysis,”
July 10, 2006, http://global.factiva.com (accessed December 23, 2006).
128
Abbas Ghaleb, “Ruling Coalition Seen Winning in Yemen,” Associated Press, April 28, 1993,
http://global.factiva.com (accessed December 23, 2006); and Muhammad Sudam, “Yemen’s Saleh
Gets Expected Presidential Win,” Reuters News, September 25, 1999, http://global.factiva.com (ac-
cessed December 23, 2006).
129
Dirk Vandewalle, Libya since Independence: Oil and State-Building (Ithaca, N.Y.: Cornell Uni-
versity Press, 1998), 16; “Libya President’s Son Urges Islamic Rebels to Accept Autonomy,” BBC
Monitoring Asia Pacific, December 15, 1999, http://globa.factiva.com (accessed June 27, 2007); and
Lisa Anderson, “Rogue Libya’s Long Road,” Middle East Report 241 (Winter 2006), 45.
628 w o r l d p o l i t i c s
reditary succession depends not simply on the will to power of geriatric
dictators and their sons, but on the reception of those ambitions by
the surrounding extra-familial elite. The above evidence supports that
theory. In the absence of prior party practices for leadership selection,
ruling elites tend to favor hereditary succession over a tumultuous free-
for-all. By contrast, they are most likely to balk at a filial successor
when the ruler himself emerged from the party.
The present investigation reopens the understudied subject of suc-
cession in modern autocracies. Its findings point to areas for further
research. Future work can expand upon the present analysis by relaxing
the scope conditions on cases and considering alternative explanations
for the institutional patterns shown above, including looking at re-
gional demonstration effects and underlying economic factors. Integra-
tive case analysis might also be a productive step; the parallels between
Tullock’s and Herb’s theories suggest modern autocracies and formal
monarchies may have more in common than often thought. Addition-
ally, the present consideration of institutional precedents can be em-
bedded in research on the origins of parties, extending the causal chain
back into the earlier histories of the regimes under consideration.
130
Finally, comparativists may choose to study the impact of hereditary
succession on subsequent regime outcomes, from democratization to
enduring authoritarianism.
Recognizing that much work remains, this comparative study of
succession responds to a question not readily addressed by an electoral-
ist or transitions approach: Who will form the next generation of lead-
ers in countries where autocracy persists? The above findings offer one
provisional answer: Sons will succeed fathers whose power predates the
party’s.
130
Benjamin Smith, Hard Times in the Lands of Plenty: Oil Politics in Iran and Indonesia (Ithaca,
N.Y.: Cornell University Press, 2007), 200–201.
