Content uploaded by Jonathan Pinckney
Author content
All content in this area was uploaded by Jonathan Pinckney on Nov 11, 2021
Content may be subject to copyright.
1
Close but Not Too Close:
Opposition Network Strategy and Democratization in Zambia
Jonathan Pinckney
United States Institute of Peace
1
Forthcoming in Social Movement Studies
Abstract
Current literature finds that democratization is most likely when opposition social
movements have initiated a political transition. Yet little work has disaggregated bottom-
up transitions to suggest when this effect obtains. This article examines a case not widely
known in the social movements’ literature: the 1991 political transition in Zambia. It
proposes a novel theoretical avenue to help explain this case’s incomplete democratic
transition. The piece re-conceptualizes the opposition movement as a multi-
organizational actor network. I argue that opposition movements face strong incentives to
unify through either centralizing around a single organization or creating a dense network
of multiple, overlapping connections. The strategy they pursue affects the transitional
political landscape. Highly centralized opposition networks are likely to lead to less
democratic outcomes while dense decentralized networks will have more democratic
outcomes. I find that high centralization by the opposition in Zambia undermined
political accountability during the transition and limited Zambia’s democratic progress.
1
The opinions expressed in this article are those of the author alone and do not necessarily represent the views of
the United States Institute of Peace.
2
Introduction
Campaigns ‘from below’ have led to many democratic transitions in recent decades
(Chenoweth & Stephan, 2011; Haggard & Kaufman, 2016). Such campaigns were key players in
the transitions after Communist regimes in the 1980s (Osa, 2003), transitions to multiparty
democracy in Africa in the 1990s (Bratton & Van de Walle, 1997), and the ‘Color Revolutions’
in Eastern Europe in the 2000s (Bunce & Wolchik, 2011). Yet despite increased scholarly
attention we know little about when these movements, after ousting an authoritarian regime,
have initiated a full-fledged democratic transition.
In this article I draw on insights from network theory to suggest one relatively unexplored
avenue: transformation of the opposition’s network structure with the goal of unifying its efforts
during its anti-authoritarian struggle. Opposition movements whose networks become highly
centralized, particularly by absorbing smaller organizations into a single organization, are less
likely to support democratic progress once in power, while more diffuse networks will tend to
facilitate greater democratization.
I then examine the effects of opposition unity strategies in the transition initiated by the
Movement for Multiparty Democracy (MMD) in Zambia in 1991. Through process tracing based
on interviews with key decision-makers I show that the MMD’s unification strategy during its
struggle against the Zambian one-party state negatively impacted the country’s democratic
transition. Through absorption the MMD eliminated many independent sources of civic pressure.
This hindered Zambia’s democratization through demobilizing opposition to the new regime’s
democratic backsliding.
3
This paper makes the following contributions. First, I apply network analysis to
democratization in a novel way, reconceptualizing opposition actors as nodes in a network
spanning oppositional space. Second, I suggest a new avenue for understanding why bottom-up
social movements might sometimes result in failed or incomplete democratic transitions. Third, I
expand the social movement literature’s focus on pro-democracy movements in the developing
world. Fourth, I provide new insight into the democratization process in a counterintuitive case:
Zambia’s 1991 transition.
Bottom-Up Democratic Transitions
Sixty years of democratization literature, beginning with the early modernization
theorists (e.g. Lipset, 1959), has settled on a few stylized facts: High levels of socio-economic
development, few major social or ethnic cleavages, and a usable set of state institutions facilitate
democratic transitions. More recently an economic turn from scholars such as Boix (2003) and
Acemoglu and Robinson (2005) have focused on inequality as a key factor influencing
democratic prospects. The literature is well-developed, even over-determined, with a wealth of
influential findings (Teorell, 2010). However, debate remains on almost all these stylized facts
(Acemoglu et al., 2008; Przeworski, 2000), and scholars’ capacity to predict democratic
transitions’ outcomes remains limited (Teorell, 2010).
One particularly poorly understood set of transitions are those in which the impetus for
change comes from social movements or the political opposition. Recent work by Haggard &
Kaufman (2016) suggests that these transitions tend to result in higher democracy. This argument
is supported by research on democratic transitions and nonviolent resistance, which consistently
finds that transitions initiated through nonviolent resistance are more likely to democratize
4
(Bethke & Pinckney, 2019; Chenoweth & Stephan, 2011; Pinckney, 2020). However, to date,
there has been little research that disaggregates these transitions to understand when bottom-up
transitions democratize and when they do not.
In part, this is because scholarly work on bottom-up transitions tends to aggregate their
initiating forces. Yet, as Przeworski (1991) reminds us, while opposition forces may appear
united during a period of struggle they inherently involve complex relationships of different
factions with varying goals. These relational patterns have major consequences for the long term,
giving rise to patterns of political competition that shape the political transition and the
subsequent regime (Osa, 2003). This complexity calls for a dedicated toolkit. One of the most
powerful sets of tools available for this purpose is network analysis.
Networks and Unity in Contentious Politics
Network approaches have a long history in the study of social movements, both through
examinations of individual-level social networks (McAdam, 1986; Siegel, 2009) and
organizational patterns of conflict and cooperation (Clarke, 2014; Metternich et al., 2013; Osa,
2003). However, there has been little focus on applying this rich set of tools to pro-democracy
campaigns in authoritarian regimes.
One of the most important strategic goals of a social movement is achieving unity (Tilly,
2010). Bunce and Wolchik (2011) point to a unified opposition as one of the key factors in the
successful ‘Color Revolutions’ of the early 2000s. Unity’s importance is shown in the negative
as well: opposition fragmentation is a strategy of most non-democratic regimes (Kefale, 2011). A
unified movement indicates single-peaked preferences among the larger population. In contrast,
if a movement’s message is unclear or divided among many different factions, the preference
5
signaling portion of resistance and public protest is rendered largely inoperable (Wouters &
Walgrave, 2017). Unity also has practical advantages. A unified movement can more easily
prevent defections from its ranks and is also more likely to be able to avoid outbreaks of violence
and maintain a more fully nonviolent conflict posture, which can increase its effectiveness
(Pearlman, 2011). For pro-democracy movements in authoritarian regimes, unity is even more
crucial to counter the strong resource advantages and repressive capacity of the state.
If such a pro-democracy movement is seeking to achieve its goals at least in part through
elections, then unity is necessary to overcome the barriers necessary to achieve victory in an
election that is not free and fair (Cheeseman, 2007). The existence of a unified opposition
coalition is one of the factors most robustly associated with a ‘liberalizing electoral outcome’
(Howard & Roessler, 2006), particularly when complemented by the ‘vertical’ threat of
mobilization from below (Sato & Wahman, 2019). Because of these benefits, opposition
campaigns will face strong incentives to unify, though they will not necessarily achieve it, due to
pressures from countervailing internal incentives towards competition (Beardsworth, 2018), and
coalitions can be fragile without the existence of long-standing well-known groups at their center
(Gandhi & Reuter, 2013), and when their constituent elements have differing policy agendas
(Gandhi & Ong, 2019). If these challenges can be overcome, then unified campaigns are those
most likely to successfully initiate a transition, though not necessarily to lead to full
democratization, as Resnick (2013) highlights, and as I discuss below.
How can opposition movements unify? Conceptualizing the opposition as an inter-
organizational network and drawing on insights from studies of the flow of information through
networks suggests two basic strategies: increasing the relative centrality of a single node or small
number of nodes or increasing the overall density of the network. This distinction in turn
6
provides a useful way to examine the character and likely effects of opposition movements on
democratization. These strategies are not mutually exclusive, and we are likely to observe some
degree of both occurring simultaneously in most opposition campaigns. However, we can
distinguish campaigns based on which of these strategies predominates.
The first strategy involves centralizing the network around a single organization that can act
as an information broker and a decision-making body, as shown in Figure 1. A single node sits at
the center of the movement and dictates the movement’s goals, strategies, and tactics outward.
Opposition campaigns often achieve this through creating a ‘resistance front’ in which a single
body speaks for and directs previously disparate organizations. Centralized network structuring
can be highly efficient in coordinating information. The node at the network’s center serves as a
broker. If information is to diffuse from any external node through the rest of the network then it
will have to pass through the central node. In contrast, the central node can easily diffuse
information to all other nodes in the network. Since it is highly coordinated, centralized
oppositional networks will more easily negotiate with opponents, and achieve concessions.
7
Figure 1: Example of a Highly-Centralized Network
One extreme way to achieve this ideal-type network structure is through fully absorbing
smaller groups into a single organization. Absorption, if feasible, is likely to be the most
effective strategy for unifying. It reduces points at which information flow can be obstructed and
reduces the possibility of organizational fragmentation. Absorption is also likely to lead to
greater effectiveness in mobilization and achieving political objectives. As Metternich and his
co-authors (2013) observe, actors in an oppositional space face a fundamental collective action
problem. All of them would like to see the current government ousted, but would like to free-ride
on others’ effort to achieve this. Absorption reduces the collective action problem at an
organizational level by combining all major opposition organizations in a single node.
An alternate strategy is increasing network density. This involves creating as many
connections as possible between all nodes in the oppositional network, as shown in Figure 2.
Unity is created through increasing the depth and complexity of ties, bringing the network closer
to a fully complete structure. The density strategy is likely to create opposition movements that
8
are highly flexible but that face difficulties in achieving strategic concessions. Non-hierarchical
dense movements are more difficult to negotiate with since no one point in the network has a
greater ability to diffuse information to the rest of the network than any other.
Figure 2: Example of a Dense Network
Opposition campaigns do not select unity strategies randomly, or fully exogenously of
structural factors. For instance, the density of pre-existing civil society will affect what nodes
exist to be unified. Social and political cleavages may also impact certain strategies’
attractiveness and feasibility. Yet they do involve a significant degree of contingency and are not
fully reducible to their structural preconditions. For example, Clarke (2014) shows how a rapid
series of contingent choices by key brokers reconfigured Egyptian social networks, leading to the
highly unexpected 2011 Revolution.
9
Network Structure and Democratization
How do unity strategies affect post-campaign democratization? In this section I argue that
highly centralized opposition networks are likely to undermine democratic prospects, while
dense, complex networks are less likely to do so.
Democratization, the process of moving towards a politics that approximates the ideal of a
system: ‘almost completely responsive to all of its citizens […] considered as political equals’
(Dahl, 1973, p. 2) never takes place overnight. A founding election, even a free and fair one,
does not ensure that a country will stay on that democratizing path (Wahman, 2014). It is a long
road until democracy is truly the ‘only game in town’ (Linz & Stepan, 1996). Because of this,
those in power at a transition’s beginning can significantly shape their countries’ future
trajectories. In particular, new elites, even if they have expressed democratic preferences before
coming into power, face strong incentives to not create a system in which they might lose that
power (Przeworski, 1991). Any electoral uncertainty is likely to motivate new elites to slow or
reverse the democratization process (Wahman, 2014). To prevent new elites from following this
anti-democratic incentive, it is advantageous to have a strong oppositional space that can
mobilize against democratic reversals by new powerholders (Pinckney, 2020).
The opposition campaign’s unity strategy before the transition shapes the oppositional space
during the transition and thus the capacity for the opposition to engage in civic and political
mobilization. When the transition begins, often the people moving into state leadership will be
those who have led the prior struggle. Thus, in opposition campaigns with highly centralized
networks, the center of that network will be removed, leaving a fragmented residual network
with little capacity to engage in concerted collective action. A new opposition network will only
10
be recreated after an often-lengthy rebuilding process, during which a democratic transition may
be derailed. In contrast, democratization is most likely when the oppositional network has
primarily pursued a density-focused strategy. When nodes in this network exit the oppositional
space to come into power their removal does not transform the oppositional network. Rather a
dense web of interconnected organizations remains in place to demand concessions and impose
costs on new rulers. Failure to achieve unity means opposition movements are highly unlikely to
achieve a democratic breakthrough in the first place.
Research Design
I test this theoretical framework in a relevant case: the 1991 Movement for Multiparty
Democracy (MMD) in Zambia. While the MMD is the subject of several excellent studies
(Baylies & Szeftel, 1992; Bratton, 1994; Cheeseman, 2007; Chikulo, 1993; Erdmann &
Simutanyi, 2003; Ihonvbere, 1995a; Rakner, 2003), the case is not well-known in the social
movements literature, nor has a network approach (to my knowledge) previously been used to
examine it. As Beardsworth (2018, p. 124) points out, while some have treated the MMD as a
coalition, this is not accurate. Instead of operating as a temporary coalition of co-equal partners,
the MMD pursued a highly centralized unity strategy, absorbing smaller organizations into a
single, overarching, political front.
If this oppositional network structure was, first, unsuccessful in achieving overthrowing its
opponent regime or, second, highly successful in democratizing after successfully overthrowing
its opponent regime, it would be strong evidence against my theoretical framework. The case of
the MMD is thus a ‘hoop test’ for my theory (Collier, 2011), and a ‘crucial case’ for establishing
its validity (Gerring, 2007).
11
This case study is based on roughly 50 expert interviews conducted with key figures from the
MMD, the single-party authoritarian regime that preceded the MMD, and journalists, civil
society activists, and opposition politicians during the MMD’s rule.
i
I selected interviewees
through an iterative process of identifying perspectives across the major political and social
cleavages of the time, initially through reading the secondary literature and refined through
consultation with my initial interviewees. The goal was to triangulate as complete a picture of the
transitional events as possible (Berry, 2002; Leech, 2002), while ensuring representation of
traditionally under-represented voices, such as women and grassroots activists. Interviews were
semi-structured and roughly an hour. Following advice from McCracken (1988) and Spradley
(1979) I focused on so-called ‘grand tour’ questions, in which I invited the interviewee to give an
overview of the most important events of the MMD’s struggle against the one-party state and the
subsequent political transition.
Many of the issues related to the transition remain politically salient in Zambia, even nearly
thirty years later. Thus, to encourage candor and ensure my interviewees’ protection, when
interviewees gave their informed consent to participate in the research they did so with the
understanding that while their participation in the research would be acknowledged, no specific
quotations or assertions would be attributed to them personally. Thus, all quotes below are solely
linked to a de-identified numeric signifier.
ii
Based on this extensive research, in the section below I trace the MMD’s emergence,
consolidation, victory, and eventual transition to power, focusing on the dynamics of
cooperation, consolidation, and eventual fragmentation through a network lens. In addition to
describing these processes qualitatively, I also produce network graphs detailing the opposition
network structure prior to the MMD’s formation, at key points during its struggle, and in the
12
immediate aftermath of its rise to power. I include as nodes each organization or major social
group that engaged in regular action critical of the regime. I draw undirected edges between
nodes that regularly communicated about and collaborated to achieve shared goals. The size of
the nodes is based on their number of connections to other nodes in the network (degree
centrality).
How to operationalize network connections was a key challenge. Given the time elapsed
between these events and the interviews, and a lack of archival materials specifically from the
MMD, I did not consider it reliable to use a highly-specific protocol for generating these graphs,
such as Osa (2003) does. Such a protocol would likely include significant measurement error.
Instead, I built graphs based on representative, non-controversial accounts. If multiple
interviewees described cooperation and communication between two organizations or groups
that took place multiple times over more than a month and addressed the central goals or
objectives of the organization, then I drew a connection between those groups. This was typically
through formal organizational ties such as written agreements or explicitly coordinated planning
processes.
This means that the number of connections in my graphs is conservative and may understate
informal or irregular forms of cooperation. This focus is warranted given that the key impact of
network structure I am interested in is on unity at a macro level. Creating unified goals and
strategies across large national-level organizations, while it may be facilitated by irregular or
purely informal modes of cooperation, is unlikely to truly be achieved unless those informal
forms of cooperation are matched by more public, formal, or regular forms of cooperation.
13
While the Zambian transition only tests my theory in a single political transition, it provides
multiple opportunities to make causal process observations (CPOs), that is to say ‘insight[s] or
piece[s] of data that provide information about context or mechanism’ (Collier et al., 2004, p.
252). Specifically, it provides opportunities to test (1) whether a disunified Zambian opposition
was ineffective in challenging Zambia’s single-party authoritarian regime, (2) whether the
Zambian opposition intentionally sought to increase its effectiveness through a distinct and
intentional unity strategy, (3) whether its successful unification led to the successful overthrow
of the authoritarian regime, (4) and whether the entry of the MMD into government disrupted the
opposition network such that civic mobilization and accountability were undermined and
negatively impacted the democratic transition. The network graphs at each of these moments
provide evidence of the shape of the opposition network during each of these observations, while
my descriptions provide qualitative evidence that, in each of these cases, the proposed theoretical
relationship obtained. This combination of evidence through multiple CPOs strongly suggests
that the opposition’s network structure played an important role in both initiating the transition
and in undermining its democratization.
Opposition Network Strategy in Zambia
While initially a multiparty democracy, Zambia became a single-party authoritarian regime
under the firm control of founding president Kenneth Kaunda in 1973. While the Kaunda regime
projected domestic unity, there was active opposition throughout its rule (Larmer, 2016).
University students at the University of Zambia (UNZA) engaged in regular protests, and labor
unions, while ostensibly corporatized into Kaunda’s United National Independence Party
(UNIP), engaged in strikes against unpopular government policies (Chiluba, 1994). Wage
demands in particular became more salient during an economic decline in the 1980s (LeBas,
14
2011). Zambia’s church umbrella bodies spoke out for the protection of human rights (Hinfelaar,
2008). Increasing personalization of President Kaunda’s rule through the 1980s also led to many
former UNIP elites being expelled from the party or falling out of favor. Many of these became
vocal opponents of the one-party system of rule.
Thus, by the end of the 1980s there was widespread opposition to the Kaunda regime.
However, these efforts lacked coordination. Multiple lines of evidence support this picture, from
scholarly accounts of the time (Larmer, 2016), to the recollections of key figures from the period
in interviews. Former leaders of each group from this time tend to describe their struggles as
almost entirely independent. As one former trade union leader (Interviewee 25) put it: ‘During
the one party state […] the labor movement was more of, like a de facto opposition. They were
the only voice which was standing up and challenging the government of the day.’ The dominant
mode of operation was fragmented and diffuse.
Figure 3 shows the opposition at this time. Zambia’s trade union umbrella body, the Zambian
Congress of Trade Unions (ZCTU) and the Mineworkers of Zambia (MUZ) engaged in regular
close cooperation.
iii
Similarly the three major Zambian church organization: The Council of
Churches in Zambia (CCZ), Evangelical Fellowship of Zambia (EFZ), and the Zambian
Episcopal Council (ZEC) engaged in regular cooperation, joining together in calls for greater
human rights protections, and jointly organizing independent media sources. As one church
leader (Interviewee 102) stated: ‘The church was very united. […] Catholics and evangelicals
spoke with one voice, and the three mother bodies worked hand-in-hand […] Kaunda wanted to
introduce humanism, and the church raised up against him. It was a united front.’ Former elites
from UNIP were closely connected with the so-called ‘intellectuals’ in the Economics
Association of Zambia (Interviewee 39, Interviewee 52). Student groups were largely on their
15
own, protesting over student-specific demands and occasionally directing their ire at the state
(Larmer, 2016).
Figure 3: The Zambian Opposition Network in the 1980s
A series of events in the late 1980s significantly shifted this picture. First, declining
prices for copper, Zambia’s primary export, undermined state capacity to provide public welfare
goods. Increasing public debt led the government to seek a financial bailout from the IMF, which
led to the adoption of a systematic structural adjustment program (SAP) in 1983 (Simutanyi,
1996). The SAP included significant reductions in government services, particularly the
subsidization of basic staple goods such as corn meal flour. While President Kaunda withdrew
from the SAP in 1987 in response to widespread opposition, the need for another IMF loan led
the Zambian government to once again lift corn meal flour subsidies in 1990. The lifting of
subsidies and subsequent rise in the price of corn meal sparked food riots.
16
Second, glasnost and perestroika in the Soviet Union sparked thinking among Zambia’s
opposition forces that overthrowing the Kaunda regime might be possible. President Kaunda had
closely aligned himself with the Soviet Union through his ideology of ‘humanism,’ a variant on
socialist thinking. The downfall of his international fellow spirits caused many to wonder about
Zambia’s stability.
Third, in June 1990 a group of Zambian military officers attempted to stage a coup. The
coup was quickly suppressed. However, during the few hours where the coup plotters seized
public broadcasting stations and were proclaiming the end of the Kaunda regime widespread
celebratory demonstrations broke out on the streets of Lusaka. These celebrations provided
crucial information that Kaunda’s ostensible popularity was largely a matter of ‘preference
falsification’ (Kuran, 1991) and that opposition to his regime was widespread.
In response to these events, even more diffuse opposition to the Kaunda regime began to
arise. The government’s lack of capacity undermined loyalty to the state, the reaction to the coup
sent a strong signal that this disloyalty was widespread, and the international environment
provided powerful ‘cognitive liberation’ (McAdam, 1982) indicating that resistance, if initiated,
had a good chance of succeeding. These groups were largely small and self-organizing, with few
principles beyond simple opposition to the regime. As one activist who helped form an
independent group (Interviewee 55) put it: ‘We had a cell in Ndola, a small cell, to be honest we
were not even aware that there was [other opposition] happening. We started organizing
ourselves in Ndola with a few colleagues, meeting to strategize on how to oppose the one party
state […] we were meeting almost every day during lunchtime to discuss, strategize.’
17
Second, the various pre-existing opposition forces united to initiate a concerted push for
political change. This coming together was initiated by the intellectuals of the Economics
Association of Zambia (EAZ), particularly EAZ Chair Akashambatwa Mbikusita-Lewanika and
his colleague Derrick Mbita Chitala. These two, recognizing the diverse yet diffuse avenues of
resistance currently in place, organized a two-day conference at the Garden House Hotel in
Lusaka, inviting representatives from all the opposition sectors. The result of the conference was
the formation of a civil society pressure group: The Movement for Multiparty Democracy, which
would bring together these diffuse elements under a single umbrella to advocate for a return to
multiparty politics in Zambia.
iv
The Garden House meeting was the most prominent example of an intentional effort at
network-building and unity-forging during the Zambian transition, and a particularly prominent
example of intentional unity strategy selection. As one leader at the meeting (Interviewee 42)
stated:
Of course there had been […] not resistance, but a kind of opposition from the so-called
civil society organizations, trade unions, and old politicians who had fallen out of favor,
and […] these disparate groups said: Look, I think we need to get together and form a
movement. And that’s how the Movement for Multiparty Democracy started.
Those who attended the meeting came in with varying agendas that were debated
vigorously in the meeting. Among these debates was the strategy for unity. While some argued
for a more diffuse network without a single central named organization, others argued that large-
scale mobilization would be impossible without the creation of a single named organization to
coordinate activity and symbolize the struggle (Interviewee 107, Interviewee 39). What emerged
18
was a cohesive strategy of coordinating the activities of the various opposition forces,
particularly the labor unions, intellectuals, and former UNIP elites into the ‘three-legged iron
pot’ of the MMD (Interviewee 39).
Figure 4 depicts this moment, with the emergence of multiple independent groups
happening simultaneous with the linkage of the previous resistance elements into a single
network through the brokerage of the Economics Association of Zambia.
Figure 4: The Zambian Opposition Network after the Garden House Meeting
The new MMD quickly moved to take advantage of the current of anti-incumbent
popular opinion. They organized rallies across Zambia’s ten provinces and built connections
with independent groups through attendance at these rallies, as well as reaching out directly
through local networks of the various organizations that had become a part of MMD. One of the
biggest of these was the ZCTU, which had previously created a nationwide network of small
19
local groups. While ZCTU itself remained independent, many of these groups simply directly
transitioned into local MMD branches (LeBas, 2011, pp. 160–162).
Simultaneously, MMD was integrating many of the pre-existing groups that had engaged
in resistance within its structure. The upper echelons of MMD were largely made up of former
figures from UNIP, with former Minister Arthur Wina serving as MMD’s first president. The
intellectuals of the EAZ largely moved into the bureaucracy of MMD, with Akashambatwa
Mbikusita-Lewanika serving as MMD’s first National Secretary. There was close integration of
the ZCTU as well, with former ZCTU chair Frederick Chiluba becoming MMD’s first Vice-
President for Mobilization, but with ZCTU itself remaining independent.
Figure 5 depicts this stage in the development of the MMD as a network of resistance
organizations. MMD itself was the central hub in the network and linked out to all the previously
independent local-level organizations. Several of the original nodes, such as the EAZ and former
figures from UNIP, were now all organizationally integrated into the MMD structure.
20
Figure 5: The Zambian Opposition Network in late 1990
In December 1990 President Kaunda, seeing the growth of MMD as a pressure group,
switched tactics and allow for multiparty presidential and parliamentary elections that year, with
the strategy being that Kaunda was still relatively popular and winning a multiparty election
would give him the necessary legitimacy to remain in power. MMD quickly shifted its strategy
from organizing resistance and pressure as a civil society group to functioning as a political party
that would contest the 1991 election. This entailed greater centralization of the network of
resistance organizations. Many of the previously independent local groups were now fully
integrated into the MMD structure, often simply repurposed as local chapters of the MMD
(Interviewee 55). On the eve of the 1991 election, MMD was a behemoth organization, which
had absorbed much of the country’s civil society and political elites (as defections from UNIP
21
increased in the lead-up to the election), and with ‘an effective monopoly on opposition to UNIP’
(Bartlett, 2000, p. 445). Figure 6 depicts this moment.
Figure 6: The Zambian Opposition Network before the 1991 Election
MMD’s strong organization led it to an overwhelming victory in the 1991 election. Their
presidential candidate, former labor leader Frederick Chiluba, received 73% of the vote. Out of
150 seats in the Zambian parliament, MMD candidates won 125, in some cases winning with
over 90%. Their victory was so striking that, even in areas with little campaigning, the MMD
received staggering electoral margins. As one MMD candidate (Interviewee 100) described it:
I stood as an MMD candidate in the remotest part of the country, on the Angolan border,
where politics was not really developed. I hadn’t been there before, and I was hardly
known there. But [despite that] I won 80% of the vote!
22
The MMD’s organizational structure and strategy proved effective in taking the
opportunities present in their environment and transforming them into a stunning moment of
highly mobilized political change.
With such an overwhelming rejection of authoritarianism and support for a pro-
democracy movement, we might expect that Zambia’s democratic transition would have been
relatively quickly consolidated into a stable democratic regime. As Rakner (2011) observes,
Zambia’s transition was in many ways a ‘best-case scenario’ for democratic consolidation.
International observers described the Zambian transition as a ‘model’ (Bratton, 1992; Joseph,
1992), and massive foreign aid flowed in to help consolidate the transition (Rakner, 2003).
However, multiple challenges that link directly to the MMD’s unification strategy during its
struggle against the Kaunda regime undermined this process.
The first problems in Zambia’s transition arose as President Chiluba moved to centralize
power. Under pressure from MMD, UNIP had amended the Zambian constitution just prior to
the 1991 election. Important changes included legalizing other political parties, removing
references to Kaunda’s philosophy of humanism, and limiting the president to two terms.
However, much of the legal structure of the one-party state, including the ‘enormous’ powers of
the executive (Ndulo & Kent, 1996, p. 261), remained in place. Chiluba moved quickly to take
advantage of these powers, as he and many of the new leaders began using state resources to
cement their own political control and to put in place corrupt patronage systems for themselves
and their supporters, beginning a ‘descent into an anti-democratic kleptocracy’ (Bauer & Taylor,
2011, p. 53). Hopes that Chiluba, a former labor leader, would put in place greater protections
for Zambia’s working classes were also not fulfilled, as the MMD, under pressure from
23
international financial institutions (Abrahamsen, 2000), extended the rollback of state services
and protections initiated under President Kaunda.
While there was some popular resistance to these actions, particularly by labor unions
opposed to the harsh terms of the IMF’s structural adjustment plan (Larmer, 2005), this
opposition was ‘demobilized or at the best ineffectual’ (Simutanyi, 1996, p. 825). As one
women’s rights activist (Interviewee 2) put it:
When we changed to the MMD who were so against autocracy [we thought] ok, these
people know that we don’t like autocracy, they will behave differently […] the checks
and balances were not put in place, or what was put in place was not utilized. So the
MMD gained more powers and started misusing it […] by the time the general public
noticed that it was misused it was too late, because the government had become more
powerful than the people.
This lack of mobilization was in large part due to a lack of independent structures
through which mobilization could be channeled. The resistance against the Kaunda regime had
been so deeply centralized within a single organization, with most independent sources of
political pressure integrated into its structures. When this single organization, the MMD, became
the ruling party there were few voices outside of the ruling party to speak out against it. With the
removal of the MMD as a central node in the network of resistance, opposition was largely in the
same diffuse, fragmented place that it had been in prior to the spark of the events of 1990. As
one MMD leader from that time (Interviewee 68) put it, after describing his victory with over
92% of the vote in the 1991 election:
24
That was the strength of MMD. We could do the nasty things that Kaunda couldn’t do
[…] because who was asking questions? Who out of my 8% of [opposing] votes was
going to ask me what I’m doing in a hostile way? Nobody.
The MMD under Chiluba quickly became a single-party electoral authoritarian system in
all but name (Erdmann & Simutanyi, 2003). The Chiluba government used state resources and
the state coercive apparatus to suppress opposition, particularly in response to the so-called ‘Zero
Option’ plan by former UNIP leaders to destabilize the government, which led to the declaration
of a state of emergency and widespread arrests (Ihonvbere, 1995b). Later, when former President
Kaunda attempted to return from political retirement and contest the 1996 presidential election
he was arrested and physically attacked. When these measures failed to dissuade him from
running, he was barred from contesting the presidency through an amendment to the Zambian
constitution barring Zambians whose parents had not been born in Zambia from the presidency.
The measure was transparently written to target former President Kaunda.
Many conscientious members of the MMD left the party during this period, arguing that
it had moved away from its roots and become mired in authoritarianism and corruption.
However, with the opposition network fragmented by MMD’s move from opposition to
government, rebuilding a meaningful opposition structure took time, as well as a new impetus for
action. As Lise Rakner (2003, p. 16) wrote soon after the Chiluba period: ‘Despite the presence
of thirty-six listed parties and around 400 non-governmental associations, there are presently
almost no organizations in Zambia capable of contesting the incumbent’s power.’ As during the
Kaunda period, opposition organizations existed, but they failed to unify in a single coherent
network capable of challenging the regime. The scattered attempts at creating opposition parties
and forming new civil society groups in opposition to the MMD only came together ten years
25
after the MMD’s 1991 victory when President Chiluba began making indications that he would
seek a third term as president.
Alternative Explanations
The preceding narrative shows that the network structure of the Zambian opposition
followed the processes outlined in my theory. Yet multiple objections can be raised to this
narrative. I consider two: one, that the Zambian transition was much more democratizing than I
am arguing, and two, that the negative effects I identify can be fully attributed to other causes.
First, is Zambia’s 1991 transition actually a democratic success story? After all, Zambia
did not return to full-fledged de jure autocracy, and has maintained legal multiparty democratic
competition since 1991. In 2011, 20 years after the events described here, MMD’s third president
Rupiah Banda gave up power after losing an election to the opposition Patriotic Front.
The elimination of de jure one-party dictatorship was an undeniably significant
democratizing step. Yet at least for the ten years of the Chiluba presidency (and arguably up until
the 2011 transfer or possibly up until the present day), it is difficult to argue that Zambia was de
facto democratic. In addition to the anti-democratic steps described above, the Chiluba
government engaged in wholesale repression or attempted illegitimate co-optation of the
opposition throughout its tenure. One interviewee described a memorable meeting with President
Chiluba in which he was promised a limited number of seats if he would stop publicly criticizing
the president (Interviewee 32). The 1996 elections, in which Chiluba won his second term, were
a significant step down in quality from the 1991 elections (Baylies & Szeftel, 1997), and were
widely condemned as not free and fair (National Democratic Institute, 1997; Reynolds, 1999). It
required the mass mobilization of the anti-third term movement to remove Chiluba from office.
26
International democracy indicators largely concur with this assessment. The Varieties of
Democracy project (Coppedge et al., 2021), for instance, while noting a modest improvement in
Zambia’s ‘polyarchy’ score in 1991, categorizes it as an ‘electoral autocracy’ from independence
until the present, with only a brief period of limited ‘electoral democracy’ in the mid-2000s. It is
beyond the scope of this article to fully describe the full thirty years of politics in Zambia since
1991, but for the reasons stated here I consider it reasonable, despite the real progress made, to
characterize the 1991 transition as unsuccessful in democratization.
Second, it is possible that the impacts that I have argued network strategies led to are in
fact merely a spurious correlation due to other factors. What alternate explanations might better
explain this case?
Multiple factors were at work in the movement of the MMD from idealistic opposition
umbrella to corrupt, co-opted ruling party. Many of MMD’s leaders were former UNIP elites, a
factor that helped facilitate an enduring anti-democratic political culture (van Donge, 1995). The
rapidity of the transition also limited the scope for change, leaving in place many aspects of the
old one-party state. This limited scope for change fit well with the priorities of influential
international donors more interested in supporting a ‘disciplined democracy’ that would support
free-market principles than a government that would genuinely represent the will of the Zambian
people (Abrahamsen, 2000). As one interviewee put it:
Some of us proposed to have an interim government whose task would be to change the
constitution before moving forward […] we were talked out of that. Even those
supposedly supportive, democratic agencies like the Carter Center were basically against
27
a radical change. And this tallied very well with the majority of our leaders who truly
were not for any radical change (Interviewee 39).
These factors all limited the extent of change that was possible during the 1991 transition
and the subsequent Chiluba presidency.
However, the choices of opposition network structure in the period prior to Kaunda’s
overthrow also played a significant role in the events of the transition. The incentives shaping
these choices and their consequences closely match the causal processes suggested by the
approach I argue for above. The founders of the MMD clearly understood the benefits of a
having a unified opposition and created a single, highly versatile, expansive organization that
could and did easily incorporate other organizations. The goal, as articulated from the beginnings
of the organization in the Garden House Hotel conference was to have a single vibrant, powerful
center to coordinate previously disparate attempts at resistance to the Kaunda regime. MMD was
stunningly successful in achieving this, as seen in their well-orchestrated campaign and massive
victory in the 1991 election (Bratton, 1992). A unity strategy based on centralization through
absorption successfully achieved its ends.
Yet the fact that the single powerful organization at the center of the oppositional
network then entered government meant that mobilization of any kind, particularly to maintain
accountability for the new ruling party and its leader, was difficult. MMD had been the clearing
house and central coordinating body for resistance to the government. This body was now in
government. With the central broker of the resistance network no longer serving that function,
new networks had to be created. The costs of creating these networks were prohibitively high
and thus beyond the reach of most of Zambia’s opposition forces during the Chiluba presidency.
28
Only when the new shock of Chiluba’s attempt to get a third term mobilized the Zambian
opposition and civil society once again was effective democratic resistance mobilized. This
evidences the difficulties that opposition networks face when they are no longer in opposition.
Often the most powerful actors for resisting are not in positions of power and influence that
cannot be effectively opposed.
One actor who might have been an effective check on MMD was UNIP, the former ruling
party. There is a growing literature showing that authoritarian successor parties can, somewhat
counterintuitively, help facilitate the democratization process by serving as an anchor for post-
transition political competition (Loxton, 2015; Riedl, 2014). MMD’s overwhelming success in
the 1991 election certainly left UNIP a shadow of its former self. Yet UNIP still adopted the path
that some have argued would be most salutary to democratization. Grzymala-Busse (2020)
shows that the strongest democratizing effect of authoritarian successor parties comes when they
exit from power and reinvent themselves as democrats. This was what UNIP did, quickly
conceding the 1991 election and, after a brief hiatus, returning in the mid-1990s to attempt to
counter the MMD. A stronger UNIP might certainly have been a more effective check, making
the centralization of the MMD a less serious challenge for democratization. Yet the MMD’s
centralization and its resulting overwhelming victory in the 1991 election undermined the
possibility of UNIP’s return serving as this effective check.
Other alternate explanations focus on structural factors such as the level of economic
development, enduring political culture, the global mood for change at the end of the Cold War,
and the economic challenges associated with declining copper prices (Rakner, 2003; van Donge,
1995).The influence of these factors is not incompatible with this argument. The key question is
whether they are sufficient to explain not just the ultimate outcome, but each of the causal
29
process observations along the way. The fact that these factors are so slow-moving makes it
difficult to argue that they can. Zambia had similar levels of economic development, a similar
underlying political culture, and similar global geopolitical environment in 1989 before the
MMD came into existence, in 1990 when it dominated the opposition space and resoundingly
defeated UNIP, and in 1991 when it entered government, leaving a highly-fragmented opposition
behind and moving back towards an authoritarian system. Yet the dynamics of opposition
politics radically differed across these moments. Looking longitudinally at each step of the
process of Zambia’s political transition we can clearly see radically different government-
opposition relationships.
Conclusion
After decades of democratization literature and a wealth of models of democratic
transitions, we still have little systematic theory about one of the most important avenues of
democratization: bottom-up social movement campaigns. While there is evidence that this
avenue tends to lead to greater democracy (Chenoweth & Stephan, 2011; Haggard & Kaufman,
2016; Pinckney, 2020), few studies have disaggregated these bottom-up campaigns to explain
when this democratizing effect fails to obtain, as in the 1991 transition in Zambia. In this article I
draw on network theory and the crucial process of unification to explain this outcome.
Unity is facilitated through either increasing a network’s centrality or its density. A
centrality-based strategy may seek to absorb many organizations into a single organizational
structure or give a privileged central position to a single brokering organization, allowing
30
information to flow outwards to the other nodes in the network and preventing conflicting
information from spreading. In a density strategy, the opposition unifies by creating as many
connections as possible between the nodes.
These strategies have different implications for the likelihood that, after a movement has
succeeded in overthrowing a regime, that the overthrow will be followed by democratization. In
this situation, the centrality-based strategy is likely to make democratization more difficult since
the central node is likely to exit the opposition space and enter government, leaving behind an
oppositional network that is highly fragmented, with little capacity for mobilization to keep the
new regime accountable.
This theory helps explain the puzzling failure of democratization in Zambia, in which an
opposition movement pursued an intensive centralization strategy in the lead-up to that country’s
transition to democracy. This strategy, while leading to a stunning electoral victory, ultimately
concluded in a deeply flawed democratic transition.
While causal process observations throughout the Zambian transition bolster the
argument’s plausibility, the evidence leaves many questions for future research. In particular,
future work should examine whether the negative impact of opposition network centralization
identified in Zambia holds across a wider spectrum of cases. What scope conditions limit the
applicability of this framework?
The origins of unity strategies are another important area for further research. While the
accounts of the early MMD meetings do indicate the exercise of a meaningful degree of strategic
choice, it is also certainly true that pre-existing social and institutional structures influence what
unity strategy choices are available and reasonable. How might existing social and political
31
cleavages affect this picture? Centralized networks may be less likely to emerge in societies with
deeply entrenched political cleavages, where there would be greater resistance in the opposition
network to absorption within a single central node.
v
Other pre-existing institutional structures,
such as the non-democratic regime’s degree of institutionalization of a limited opposition, might
also affect what strategies are available to opposition campaigns.
This approach also suggests several policy implications for opposition movements
interested in democratic transitions and outside actors interested in facilitating democratic
breakthroughs. While more systematic research is needed, this research suggests trade-offs that
both oppositions and policymakers should consider carefully. Highly centralized opposition
networks are likely to be more successful in overcoming the difficult collective action problems
in achieving democratic breakthroughs and may even be stunningly effective in bringing about
political change. But this short-term effectiveness comes with long-term costs. Careful attention
to the network structure of an opposition movement during its period of struggle can tell us much
about the likely long-term consequences of that struggle.
32
Notes
i
See the appendix for the complete list of interviewees.
ii
This process was approved by the Institutional Review Board at the University of Denver,
protocol number 962008.
iii
Other trade unions also occasionally participated in resistance actions, but these tended to be
more peripheral. See Rakner, 1992 for more details.
iv
For a detailed description of the Garden House meeting, see Mbikusita-Lewanika, 2003.
v
Thanks to an anonymous reviewer for making this suggestion.
Works Cited
Abrahamsen, R. (2000). Disciplining democracy: Development discourse and good governance
in Africa. Zed Books.
Acemoglu, D., Johnson, S., Robinson, J. A., & Yard, P. (2008). Income and Democracy.
American Economic Review, 98(3), 808–842.
Acemoglu, D., & Robinson, J. A. (2005). Economic origins of dictatorship and democracy.
Cambridge University Press.
Bartlett, D. M. C. (2000). Civil Society and Democracy: A Zambian Case Study. Journal of
Southern African Studies, 26(3), 429–446.
Bauer, G., & Taylor, S. D. (2011). Politics in Southern Africa: Transition and Transformation.
Lynne Rienner Publishers.
Baylies, C., & Szeftel, M. (1992). The fall and rise of multi‐party politics in Zambia. Review of
African Political Economy, 19(54), 75–91. https://doi.org/10.1080/03056249208703954
33
Baylies, C., & Szeftel, M. (1997). The 1996 Zambian elections: Still awaiting democratic
consolidation. Review of African Political Economy, 24(71), 113–128.
https://doi.org/10.1080/03056249708704242
Beardsworth, N. A. (2018). Electoral coalition-building among opposition parties in Zimbabwe,
Zambia and Uganda from 2000 to 2017 [PhD Thesis]. University of Warwick.
Berry, J. M. (2002). Validity and reliability issues in elite interviewing. PS: Political Science &
Politics, 35(4), 679–682.
Bethke, F. S., & Pinckney, J. (2019). Nonviolent Resistance and the Quality of Democracy.
Conflict Management and Peace Science, Onlinefirst.
Boix, C. (2003). Democracy and Redistribution. Cambridge University Press.
Bratton, M. (1992). Zambia Starts Over. Journal of Democracy, 3(2), 81–94.
https://doi.org/10.1353/jod.1992.0027
Bratton, M. (1994). Economic Crisis and Political Realignment in Zambia. In J. Widner (Ed.),
Economic Change and Political Liberalization in sub-Saharan Africa (pp. 101–128).
Johns Hopkins University Press.
Bratton, M., & Van de Walle, N. (1997). Democratic experiments in Africa: Regime transitions
in comparative perspective. Cambridge University Press.
Bunce, V. J., & Wolchik, S. L. (2011). Defeating Authoritarian Leaders in Postcommunist
Countries. Cambridge University Press.
Cheeseman, N. (2007). The rise and fall of civil-authoritarianism in Africa: Patronage,
participation, and political parties in Kenya and Zambia [PhD Thesis]. University of
Oxford.
34
Chenoweth, E., & Stephan, M. J. (2011). Why civil resistance works: The strategic logic of
nonviolent conflict. Columbia University Press.
Chikulo, B. C. (1993). End of an era: An analysis of the 1991 Zambian presidential and
parliamentary elections. Politikon, 20(1), 87–104.
https://doi.org/10.1080/02589349308704989
Chiluba, F. (1994). Democratisation in Zambia (Master’s Thesis). University of Warwick.
Clarke, K. (2014). Unexpected Brokers of Mobilization: Contingency and Networks in the 2011
Egyptian Uprising. Comparative Politics, 46(4), 379–297.
Collier, D. (2011). Understanding Process Tracing. PS: Political Science and Politics, 44(4),
823–830.
Collier, D., Brady, H. E., & Seawright, J. (2004). Sources of Leverage in Causal Inference:
Toward an Alternative View of Methodology. In H. E. Brady & D. Collier (Eds.),
Rethinking Social Inquiry: Diverse Tools, Shared Standards. Rowman & Littlefield
Publishers.
Coppedge, M., Gerring, J., Knutsen, C. H., Lindberg, S. I., Teorell, J., Alizada, N., Altman, D.,
Bernhard, M., Cornell, A., Fish, M. S., Gastaldi, L., Gjerløw, H., Glynn, A., Hicken, A.,
Hindle, G., Ilchenko, N., Krusell, J., Luhrmann, A., Maerz, S. F., … Ziblatt, D. (2021).
V-Dem Country-Year Dataset v11.1 [Varieties of Democracy Project]. V-Dem Country-
Year Dataset V11.1. https://doi.org/10.23696/vdemds21
Dahl, R. A. (1973). Polyarchy: Participation and opposition. Yale University Press.
Erdmann, G., & Simutanyi, N. (2003). Transition in Zambia: The Hybridisation of the Third
Republic. Konrad-Adenauer-Stiftung.
35
Gandhi, J., & Ong, E. (2019). Committed or conditional democrats? Opposition dynamics in
electoral autocracies. American Journal of Political Science, 63(4), 948–963.
Gandhi, J., & Reuter, O. J. (2013). The incentives for pre-electoral coalitions in non-democratic
elections. Democratization, 20(1), 137–159.
Gerring, J. (2007). Is there a (viable) crucial-case method? Comparative Political Studies, 40(3),
231–253.
Grzymala-Busse, A. (2020). Consequences of authoritarian party exit and reinvention for
democratic competition. Comparative Political Studies, 53(10–11), 1704–1737.
Haggard, S., & Kaufman, R. R. (2016). Dictators and democrats: Masses, elites, and regime
change. Princeton University Press.
Hinfelaar, M. (2008). Legitimizing Powers: The Political Role of the Roman Catholic Church,
1972-1991. In M. Hinfelaar, J. B. Grewald, & G. Macola (Eds.), One Zambia, Many
Histories (pp. 127–143). Brill.
Howard, M. M., & Roessler, P. G. (2006). Liberalizing electoral outcomes in competitive
authoritarian regimes. American Journal of Political Science, 50(2), 365–381.
Ihonvbere, J. O. (1995a). From Movement to Government: The Movement for Multi-Party
Democracy and the Crisis of Democratic Consolidation in Zambia. Canadian Journal of
African Studies, 29(1), 1–25.
Ihonvbere, J. O. (1995b). The “Zero Option” Controversy in Zambia: Western Double Standards
vis a vis Safeguarding Security? Africa Spectrum, 30(1), 93–104.
Joseph, R. (1992). Zambia: A model for democratic change. Current History, 91(565), 199.
36
Kefale, A. (2011). The (un)making of opposition coalitions and the challenge of democratization
in Ethiopia, 1991–2011. Journal of Eastern African Studies, 5(4), 681–701.
https://doi.org/10.1080/17531055.2011.642525
Kuran, T. (1991). Now out of never: The element of surprise in the East European revolution of
1989. World Politics, 44(1), 7–48.
Larmer, M. (2005). Reaction & Resistance to Neo-liberalism in Zambia. Review of African
Political Economy, 32(103), 29–45.
Larmer, M. (2016). Rethinking African politics: A History of Opposition in Zambia. Routledge.
LeBas, A. (2011). From Protest to Parties: Party-Building and Democratization in Africa.
Oxford University Press.
Leech, B. L. (2002). Asking questions: Techniques for semistructured interviews. PS: Political
Science & Politics, 35(4), 665–668.
Linz, J. J., & Stepan, A. (1996). Problems of democratic transition and consolidation: Southern
Europe, South America, and post-communist Europe. JHU Press.
Lipset, S. M. (1959). Some social requisites of democracy: Economic development and political
legitimacy. American Political Science Review, 53(1), 69–105.
Loxton, J. (2015). Authoritarian Successor Parties. Journal of Democracy, 26(3), 157–170.
https://doi.org/10.1353/jod.2015.0052
Mbikusita-Lewanika, A. (2003). Hour for Reunion: Movement for Multi-Party Democracy,
Conception, Dissension, and Reconciliation. African Lineki Courier.
McAdam, D. (1982). Political process and the development of black insurgency, 1930-1970.
University of Chicago Press.
37
McAdam, D. (1986). Recruitment to High-Risk Activism: The Case of Freedom Summer.
American Journal of Sociology, 57(4), 64–90.
McCracken, G. (1988). The Long Interview. Sage.
Metternich, N. W., Dorff, C., Gallop, M., Weschle, S., & Ward, M. D. (2013). Antigovernment
networks in civil conflicts: How network structures affect conflictual behavior. American
Journal of Political Science, 57(4), 892–911.
National Democratic Institute. (1997). The November 18, 1996 National Elections in Zambia: A
Post-Election Assessment Report. National Democratic Institute.
https://www.ndi.org/sites/default/files/870_zm_report_111896.pdf
Ndulo, M. B., & Kent, R. B. (1996). Constitutionalism in Zambia: Past, Present and Future.
Journal of African Law, 40(2), 256–278.
Osa, M. (2003). Solidarity and Contention: Networks of Polish Oppositions. University of
Minnesota Press.
Pearlman, W. (2011). Violence, Nonviolence, and the Palestinian National Movement.
Cambridge University Press.
Pinckney, J. (2020). From Dissent to Democracy: The Promise and Perils of Civil Resistance
Transitions. Oxford University Press.
Przeworski, A. (1991). Democracy and the market: Political and economic reforms in Eastern
Europe and Latin America. Cambridge University Press.
Przeworski, A. (2000). Democracy and development: Political institutions and well-being in the
world, 1950-1990 (Vol. 3). Cambridge University Press.
Rakner, L. (1992). Trade Unions in Processes of Democratization: A Study of Party Labour
Relations in Zambia. Chr. Michelsen Institute.
38
Rakner, L. (2003). Political and economic liberalisation in Zambia 1991-2001. Nordic Africa
Institute.
Rakner, L. (2011). Institutionalizing the Pro-Democracy Movements: The Case of Zambia’s
Movement for Multiparty Democracy. Democratization, 18(5), 1106–1124.
Resnick, D. (2013). Do electoral coalitions facilitate democratic consolidation in Africa? Party
Politics, 19(5), 735–757.
Reynolds, A. (1999). Electoral Systems and Democratization in Southern Africa. Oxford
University Press.
Riedl, R. B. (2014). Authoritarian origins of democratic party systems in Africa. Cambridge
University Press.
Sato, Y., & Wahman, M. (2019). Elite coordination and popular protest: The joint effect on
democratic change. Democratization, 26(8), 1419–1438.
Siegel, D. A. (2009). Social Networks and Collective Action. American Journal of Political
Science, 53(1), 122–138.
Simutanyi, N. (1996). The politics of structural adjustment in Zambia. Third World Quarterly,
17(4), 825–839. https://doi.org/10.1080/01436599615407
Spradley, J. P. (1979). The Ethnographic Interview. Waveland Press.
Teorell, J. (2010). Determinants of Democratization: Explaining Regime Change in the World,
1972–2006. Cambridge University Press.
Tilly, C. (2010). Regimes and Repertoires. University of Chicago Press.
van Donge, J. K. (1995). Zambia: Kaunda and Chiluba—Enduring Patterns of Political Culture.
In J. A. Wiseman (Ed.), Democracy and Political Change in Sub-Saharan Africa (pp.
193–219). Routledge.
39
Wahman, M. (2014). Democratization and electoral turnovers in sub-Saharan Africa and beyond.
Democratization, 21(2), 220–243.
Wouters, R., & Walgrave, S. (2017). Demonstrating Power: How Protest Persuades Political
Representatives. American Sociological Review, 82(2), 361–383.
