Notes

Discussion Paper 27/2018
Cultural Values, Attitudes, and

Democracy Promotion in Malawi
How Values Mediate the Effectiveness of Donor

Support for the Reform of Presidential Term
Limits and Family Law
Daniel Nowack
Cultural values, attitudes, and democracy
promotion in Malawi
How values mediate the effectiveness of donor support
for the reform of presidential term limits and family law
Daniel Nowack
Bonn 2018
Discussion Paper / Deutsches Institut für Entwicklungspolitik
ISSN (Print) 1860-0441
ISSN (Online) 2512-8698
Die deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie;

detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar.
The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed
bibliographic data is available in the Internet at http://dnb.d-nb.de.
ISBN 978-3-96021-084-9 (printed edition)

DOI:10.23661/dp27.2018
Printed on eco-friendly, certified paper
Daniel Nowack is a researcher in the research programme “Transformation of Political (Dis-)Order:

Institutions, Values and Peace” at the German Development Institute / Deutsches Institut für Entwicklungs-
politik (DIE).
Email: [email protected]
Published with financial support from the Federal Ministry for Economic Cooperation and
Development (BMZ)
© Deutsches Institut für Entwicklungspolitik gGmbH

Tulpenfeld 6, 53113 Bonn
 +49 (0)228 94927-0
 +49 (0)228 94927-130
Email: [email protected]
www.die-gdi.de
Contents
Foreword
Abbreviations
Executive summary 1
1 Introduction 3
2 Research design: theory and method 4
2.1 Discussion of concepts: democracy, democratisation, democracy promotion,
and values 4
2.2 Actor constellations, values, and the effectiveness of democracy promotion 7
2.2.1 Organisational resources 8
2.2.2 Congruence of attitudes 8
2.2.3 The use of material incentives: how values mitigate the logic of consequences 9
2.2.4 Social status of international actors: how values mitigate the logic of
appropriateness 9
2.3 Methodology 10
2.4 Case selection 12
3 Country analysis: democratic reforms in Malawi 13
3.1 The political background of Malawi, 1992 to 2017 13
3.2 Case 1: Muluzi’s attempt at a third term 16
3.2.1 Process description 16
3.2.2 Explanatory factors 19
3.3 Case 2: reform of the marriage law 22
3.3.1 Process description 22
3.3.2 Explanatory factors 25
4 Appraising the hypotheses: explaining the effectiveness of support to
democracy 31
4.1 Organisational resources are a necessary but not a sufficient condition for
effective democracy support 31
4.2 Popular attitudes mediate effective democracy support 32
4.3 A value-fit between donor and recipient countries seems to condition the
effectiveness of instruments of the logic of consequences, but definitive
evidence is missing 33
4.4 Instruments of the logic of appropriateness work if the donor enjoys high
esteem, but this is sensitive to context 34
5 Conclusions and recommendations 36
References 39
Appendix 1 45
Appendix 2 48
Appendix 3 49
Figures
Figure 1: Dimensions of one actor group 8
Figure 2: Value indices for Malawi: egalitarianism-hierarchy and individualism-
collectivism 15
Figure 3: Actor constellation during Muluzi’s third-term bid, 2002-2003 18
Figure 4: Actor constellation during the process of the Marriage, Divorce, and
Family Relations Act 26
Figure 5: Criminalisation of child marriage in Africa in 2015 28
Tables
Table 1: Attitudes on term limits and one-man rule 21
Table 2: Attitudes on gender equality 27
Table 3: Attitudes towards donors in Malawi 34
Table 4: Attitudes towards donors and NGOs’ influence 35
Foreword
This Discussion Paper has been written as part of the DIE research project “What is
Democracy’s Value? The Influence of Values on the Effectiveness of Democracy
Promotion”, which is supported by the German Federal Ministry for Economic
Cooperation and Development (BMZ). The project contributes to research that informs the
implementation of Goal 16 of the Sustainable Development Agenda as well as theory-
building on external influence on the evolution of political regimes. This paper is one of
two country case studies on how social values mediate the effectiveness of donor
interventions to support democratic rights. More information and related Discussion
Papers can be found at https://www.die-gdi.de/en/research/projects/details/what-is-
democracys-value-the-influence-of-values-on-the-effectiveness-of-democracy-promotion/.
Many thanks are due to colleagues and peers who assisted in the development of this
paper. These include my colleagues at the DIE, Dr Julia Leininger, Dr Merran Hulse,
Charlotte Fiedler, and Dr Christine Hackenesch and the attendees of the workshop
“Wither the End of the Liberal Order: Do Societal Values Influence the Effectiveness of
Democracy Support?”, held at the German Development Institute / Deutsches Institut für
Entwicklungspolitik (DIE) from 7 to 8 December 2017, all of whom provided valuable
comments on earlier drafts of this paper. This research would not have been possible
without the openness and the willingness to help of the interviewees in Malawi who
deserve my deepest gratitude.
Bonn, 25 September 2018 Daniel Nowack

Abbreviations
AFORD Alliance for Democracy (Malawi)
AU African Union
BMZ Bundesministerium für wirtschaftliche Zusammenarbeit und Entwicklung (German
Federal Ministry for Economic Cooperation and Development)
CCAP Church of Central Africa Presbyterian
CCJP Catholic Commission for Justice and Peace (Malawi)
CEDEP Centre for the Development of People (Malawi)
CHRR Centre for Human Rights and Rehabilitation (Malawi)
CSO civil society organisation
IMF International Monetary Fund
LGBTI Lesbian, Gay, Bisexual, Transsexual, & Intersex
MCP Malawi Congress Party
MP Member of Parliament
NGO non-governmental organisation
OECD Organisation for Economic Co-operation and Development
PAC Public Affairs Committee (Malawi)
SADC Southern African Development Community
UDF United Democratic Front (Malawi)
UK United Kingdom
UN United Nations
US United States
Cultural values, attitudes, and democracy promotion in Malawi
Executive summary
Democratisation as a historical process began in the 19th century and is continuing in the
21st. As one aspect of this, donor countries of the Organisation for Economic Co-
operation and Development (OECD) are promoting democratic standards around the
globe. Democracy promotion now constitutes a central part of development cooperation
between donor and partner countries. A number of scholars, however, argue that the
current practice of democracy promotion is inherently flawed. By promoting democracy as
a culturally invariant political order, donor countries are glossing over significant cultural
differences in other countries and this hampers the successful and effective promotion of
democratic institutions. Yet, how cultural differences affect the promotion of democracy is
still not sufficiently understood.
This Discussion Paper is part of the larger research project “What is Democracy’s Value?
The Influence of Values on the Effectiveness of Democracy Promotion”, which aims at
understanding how societal values and attitudes influence the effectiveness of international
democracy promotion in African countries. The project looks at how social values and
political attitudes mediate the promotion of democracy in two specific realms: attempts by
heads of state to circumvent presidential term limits; and reforms to legislation in the
realm of family law and LGBTI (Lesbian, Gay, Bisexual, Transsexual, Intersex) rights.
This Discussion Paper focuses on two cases situated in Malawi: the attempt of former
president Bakili Muluzi to alter presidential term limits in Malawi in 2002; and the reform
of Malawi’s family and marriage law in 2015. In both cases, donors engaged in democracy
promotion in different ways and to different degrees. In the former case, donor countries
and organisations warned the Muluzi government to heed the democratic process and
thereby seconded popular attitudes. In the latter, donor countries and organisations played
an important role in coordinating and mobilising domestic actors towards the reform of
Malawi’s family and marriage law.
Both cases are analysed using in-depth process tracing investigating how donors took part
in the constellations of actors engaged in the reform of either the term limit or the family
law. The results of the analysis are fourfold: First, although the supply of organisational
resources to domestic actors played an important role in both cases, this was not in itself
sufficient to guarantee the success of democracy promotion. Rather the coordination of
partner organisations played a crucial role in rendering democracy promotion effective.
Second, popular mass attitudes played a crucial role in determining the outcome of the two
reform processes analysed here. Third, democracy promotion involving the employment
of conditionality instruments was effective when the values of the donors matched those
of domestic audiences. Finally, when donors were held in esteem in Malawi this aided the
effectiveness of democracy promotion by means of public statements and other
appropriateness-focused instruments. However, donor esteem did not trump popular
attitudes which emphasises that the effectiveness of democracy promotion is sensitive to
the specific issue it addresses.
These findings prompt a number of recommendations for donors wishing to promote

democracy. First, donors need to assess the space available for intervention on an issue-
by-issue basis. Values and attitudes in partner countries cannot simply be generalised
across issues but are uniquely linked to certain social and political issues. Where attitudes
German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 1

Daniel Nowack
and values in partner countries do not match the goals of donors, it is necessary for donors
to support an open discourse. If, in turn, local attitudes and values do favour donors’ goals,
domestic civil society actors in partner countries are still in need of support from donors
for the purpose of mobilisation and coordination. Furthermore, interventions making use
of conditionality seem to necessitate a concurrence of attitudes and values between donor
and partner countries.
2 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

1 Introduction
Democracy promotion is a highly normative undertaking. This expresses itself in repeated

negotiation and renegotiation between domestic and international actors. Evidence shows
that recipients of democracy promotion do not endorse it uncritically and may even
challenge it (see, for instance, Poppe & Wolff, 2013; Khakee, 2009). Following this
evidence, a whole new literature strand on the contestation of Western norms of liberal
democracy has opened up in recent years (Wolff & Zimmermann, 2016). At the core of
contestation lies the need to align promoted institutions and formal rules with the political
beliefs and notions of domestic actors. This emphasises the significance of cultural values
and attitudes in the promotion of democracy, even if the exploration of this significance
can still be considered a blank spot (Leininger, 2016).
The literature on democracy promotion and norm diffusion has established that democracy
promotion proceeds through two major types of channel, one based on rationalist
conditionality and one based on normative persuasion or socialisation (Beichelt, 2012;
Börzel & Risse, 2003, 2009; Checkel, 2001). Yet, much of this literature does not allow for
the fact that both channels of democracy promotion – rationalist conditionality and
normative persuasion – do not address unformed recipient actors. Instead, individuals and
organisations in recipient countries have fully formed ideas, beliefs, values, attitudes, and
opinions. These stem from socialisation into the local culture, and influence the effective-
ness of both the conditionality and the persuasion approach to democracy promotion.
The paper addresses this literature gap by asking how societal values and attitudes
influence the effectiveness of both democracy-promotion channels. By doing so, it
provides further differentiation of the understanding of what instruments work when and
how in democracy promotion. This will provide donors with more information on how to
form better strategies for their promotion of democracy and how to tune the instruments
they use to the different scenarios they face. The methodology applied is a qualitative case
study design that uses structured process tracing to identify the mechanisms behind how
value orientations influenced democracy promotion in Malawi in the two empirical cases.
In the first case, democracy promotion in the context of the debate on limits to the
presidential term of office in the period 2002-2003 is investigated. The second case looks
at the role of values and democracy promotion in the reform of Malawi’s marriage and
family law, a process that culminated in the passing of the Marriage, Divorce, and Family
Relations Act in 2015.
The next section will discuss both theory and the method applied. In subsection 2.1, we
will start with a discussion of key concepts and will deduce four hypotheses to guide the
empirical analysis in subsection 2.2. The next two subsections introduce the
methodological approach of structured process-tracing and the two cases selected. The
empirical section 3 begins by providing a briefing on the historical, political, and
economic background of Malawi. Subsections 3.2 and 3.3 form the analysis. Each is
structured by first providing a description of the process at hand and of the actor
constellations involved; this is then followed by a discussion of factors that explain the
process outcomes. The results of the process tracing of both cases is synthesised in Section
4, where they are discussed in the light of the hypotheses previously outlined. The
concluding section, Section 5, provides a list of recommendations for democracy
promotion in developing cooperation.

Daniel Nowack
2 Research design: theory and method
2.1 Discussion of concepts: democracy, democratisation, democracy promotion,

and values
In its broadest understanding, the term “democratisation” describes an “expansion of a set of

political ideals” (Harrison, 2007). The spread of these “ideals” happens both in the form of
notions at the individual micro-level as well as in the form of institutional design at the
societal macro-level. Although it is not only a process that takes place on the scale of
societies but also on the level of global discourse, it is most often investigated as a specific
country-based process in which a given political system transitions towards democracy.
Democratic transitioning is not a smooth, linear process. Although phases of
democratisation can be distinguished, there is no clear-cut sequence of democratisation. Any
political system – formally democratised or not – may drift back and forth within a spectrum
spanning democracy and autocracy (Carothers, 2002).
Political competition and equality appear to be essential conceptual characteristics in most

scholarly attempts to define democracy. Political competition channelled through the
electoral principle ensures contestation, and hence inclusiveness. The significance of
political competition was upheld early on in 1942 by Schumpeter (see Schumpeter, 2003)
who conceded in a minimalist definition that democracy is a political system in which
decision-making power is acquired “by means of a competitive struggle for the people’s
vote” (p. 269). Political competition also features prominently in Dahl’s widely accepted
conception (Dahl, 1971, accepted for example in Vanhanen, 2000; Lijphart, 2012) which −
along with contestation − includes participation and civil liberties. It is worthwhile noting
that the principle of democratic contestation does not only encompass electoral
competition over political office, but also includes policy contestation by civil society
(Becker, 1983; Warren, 2011). Unfortunately, political competition is a fragile principle that
can easily be undermined. Low quality of elections, for instance − that is, electoral processes
that are not conducted properly − can turn political competition into a farce (Ham &
Lindberg, 2015), while violent conflict may inhibit and disrupt political competition.
For this reason, the principle of political competition needs to be tied to the principle of
equality. In the electoral arena, this amounts to the implementation of the principle of “one
person, one vote” and of universal suffrage. Outside the electoral arena, however, these need
to be met by the “rule by laws, not men”. The purpose of the rule of law is to inhibit
arbitrariness as well as abuse of power, and hence to facilitate political competition on equal
footing. Oppression of opposition actors, either through violence or through non-violent
forms such as pursuing them through the courts, is clearly outside the rule of law as it tilts
the balance of power towards the incumbent’s advantage. Rule of law and the political
rights, civil liberties, and the accountability mechanisms supporting them are commonly
accepted as core features of any democratic political system (O’Donnell, 2004). Hence, in
addition to political competition, equality is the second cornerstone of democracy.
Strengthening these two fundamental principles of democracy in recipient countries is the

general goal of democracy promotion. Democracy promotion is defined as activities
undertaken by external actors that aim to “enable internal actors to establish and develop
democratic institutions that play according to democratic rules” (Grimm & Leininger,
2012, p. 396). It is carried out either directly through building the capacity of democratic

actors and institutions in the recipient country, or indirectly by establishing favourable

conditions for democratisation such as economic stability, peace, or education for vibrant
civil society participation.
Despite its increase, no well-established findings exist on what works best in promoting
democracy (Burnell, 2007). Democracy promotion comes with an array of different
instruments and approaches, but the mechanisms in which they work are still not well
understood (Beichelt, 2012). As briefly sketched out above, they can be broadly sorted
into two categories. These relate to two different logics of social action that delineate
motivations for actor behaviour. The logic of consequences reflects instruments that work
as incentives and conditionality. It perceives domestic actors in recipient countries as
rational who, in line with a rational choice approach, calculate costs and benefits
(Beichelt, 2012). Thus, democracy promoters attempt to incentivise them, for instance
through an increase in foreign aid, or inversely sanction them, by cutting aid. Immaterial
incentives are also possible, for example by offering diplomatic support on an unrelated
issue of international relations. In contrast, the logic of appropriateness works through
socialisation, social learning, and normative persuasion (Börzel & Risse, 2009; Checkel,
2001). It encompasses instruments that address actors according to their intrinsic
conceptualisations of the obligations they have, given their identity or social role. Instead
of taking a course of action due to the expectation of extrinsic consequences and results,
recipient actors make decisions based on intrinsic − that is, internalised − social norms and
the obligations that come with their identity (March & Olsen, 1989, 2009). Such
instruments of the logic of appropriateness are, for example, statements of condemnation
or support, pleas, demands, and requests or references to international law and standards.
To truly understand the effectiveness of the instruments of both logics, contextual factors
domestic to recipient countries need to be accommodated into any democracy promotion
analysis (Leininger, 2010). As mentioned in the introduction, democracy promotion is in
itself a normative process − a process of cultural change. The logic of appropriateness
testifies to this as it presupposes a common understanding of social norms. But most
scholars do not usually take account of normative contextual factors such as culture
differences between recipient and donor countries. Fish (2009) is a notable exception to this.
He investigates how the position of recipient countries towards democracy and towards the
promoting donor country is likely to affect the prospective outcomes of democracy
promotion. This paper links up to this in that it explores the role of value orientations and
attitudes as independent variables.
There is no established and widely regarded conceptualisation or definition of values and

of attitudes − and the difference between them − in the literature. Nevertheless, most
conceptualisations acknowledge that “values” are broad notions transcending specific
situations but serving as a “guiding line” in life, while “attitudes” are more situation-
specific and issue-focused. Hofstede (2003, p. 347) regards values as “broad preferences
for one state of affairs over others […]. They are opinions on how things are and they also
affect our behaviour”. Maleki and Hendriks (2015) follow Hofstede, while Fischer and
Schwartz (2011, p. 1128) see values as “abstract beliefs about desirable goals, ordered by
relative importance, that guide individuals as they evaluate events, people, and actions”.
The conceptualisation of values used in this paper also acknowledges the “guiding line-
characteristic” of values. They are thus seen as broad but to a degree individually variable
cultural traits that serve as heuristics or “rules of thumb” (Nunn, 2012, p. 109). They are

Daniel Nowack
abstract, intersubjective notions that arise from culture and are thus transmitted in on-
going socialisation processes, both horizontally and vertically. As such, they are a product
of social learning, and change only slowly and most times inter-generationally (Henrich &
McElreath, 2003). Our conceptualisation of values is much in line with how they are
generally conceptualised within the literature on political psychology (Feldman, 2013).
According to that, values are organised cognitively in the human mind as an evaluation
system. The interaction between the values in this evaluation system can shape behaviour
through different channels, among others, by informing the attitudes individuals hold.
Attitudes in turn are concrete positions towards direct, external phenomena, often social in
some way, that translate into personal opinions. While they are often roughly on a par with
beliefs, they are often more abstract preferences. While attitudes position individuals on
questions of social interaction and order, preferences are positions toward more market-like
transactions and trade-offs. While values change slowly and often inter-generationally,
attitudes can change faster and more frequently as they are not simply informed by the
values an individual holds but also by the conflict between these values. An example might
clarify this conceptualisation: Ecological sustainability and responsibility can serve as a
value for an individual. This value shapes the individual’s attitude on how society deals with
agricultural production. In particular, it is likely to create a disapproval of intensive livestock
farming. Subsequently, the preference formed through this is a preference for organic meat –
or no meat at all.
Scholars and researchers widely acknowledge that values are not directly observable
(Halman, 2007). Because values are more abstract than attitudes, they are more difficult to
observe than attitudes. Hence, the most common approach in measuring values is by
bundling attitudes in order to detect their underlying latent value orientation. Methodo-
logically, this is done by using attitudinal items asked in large-n surveys and by indexing or
scaling them. The outcome is often a single measure reporting a value dimension.
There are two prominent value orientations in the cross-cultural psychology literature that
promise to be especially important for political questions. These are the value orientations
of individualism-collectivism and egalitarianism-hierarchy.
Hofstede’s (2003) individualism-collectivism dimension describes a value continuum

ranging from an emphasis on individuals, self-reliance, and the importance of the self above
group membership (that is, individualism) to an emphasis on the collective in contrast to the
self, and the integration of individuals into strong, cohesive and united groups
(collectivism). The individualism-collectivism value orientation maps particularly on
competition. Individualistic values put a greater emphasis on competition than collectivistic
values do. In contrast, in collectivism, people act more pro-socially out of inherent
motivations (Moorman & Blakely, 1995). Consequently, collectivism is associated with
better in-group coordination (Wagner, 1995; Boroş, Meslec, Curşeu, & Emons, 2010), but
this is something that does not however reflect at the societal level (Marcus & Le, 2013).
Hence, a collectivistic value orientation might not raise the stakes against a society-wide
opposition towards, for instance, an incumbent president seeking an unconstitutional third
term. In contrast, the individual value orientation is associated with greater social capital,
such as engagement in associations or voluntary work, as well as with a greater trust
towards other persons (Allik & Realo, 2004; Van Hoorn, 2014).
A similar well-established value orientation is provided by Schwartz’ cultural value theory

(1992, 2006). Schwartz developed seven value orientations based on universal problems

that all societies face. Among these seven value orientations is the dimension of
egalitarianism-hierarchy. A hierarchic value orientation compels people to take an
unequal distribution of power, roles, and resources as given and legitimate. Authority,
humility, social power, and wealth are the major values of hierarchic value orientations. In
contrast to this, egalitarian value orientations emphasise that people recognise one another
as equals, and help one another. Dominant values in this orientation are help, honesty,
social justice, responsibility, and equality.
2.2 Actor constellations, values, and the effectiveness of democracy promotion
As noted above, democracy support as an element of development cooperation is seen as

being carried out by actors external to recipient countries. Traditional research approaches
to democracy promotion emphasise the role of external, international actors such as donor
countries or international organisations by differentiating between internal and external
actors. Hence, actors are grouped according to whether they are organisations in the
recipient country or not, creating a group of domestic actors that “receive” democracy
promotion and a group of international actors that “deliver” democracy promotion
(Beichelt, 2012).
However, this approach distorts the actual real-world process of democracy promotion.
There is, in fact, no clear-cut line between international and domestic actors. Many
international organisations, international non-governmental organisations (NGOs) as well
as donor country representations in recipient countries are staffed with domestic
personnel. Often enough international NGOs found country-specific “daughter NGOs”
which renders them domestic entities in legal terms. Even in the case of donor country
representations such as embassies, it is easy to see that they are not detached from the
domestic political and societal system of the recipient country but are, in fact, part of it.
The external-internal distinction in approaches to democracy promotion can thus be
rightly criticised (Leininger, 2010).
Instead of making an external-internal distinction, this paper follows Leininger (2010) in

employing the alternative conceptualisation of pro and contra actor groups in democracy
promotion. Processes of democracy promotion are often tied to a specific political process
or policy. The positions that actors of democracy promotion take towards these issues are
often not unanimous. Contrary to that, democracy promotion proceeds as a discourse, or
even struggle, over rival positions. These positions do not even need to be simply in
favour of or opposed to democracy per se; instead they might be trying to achieve the
same end by different means, such that conflicts over specific objectives arise (Grimm &
Leininger, 2012). It hence makes sense to conceptualise actors of democracy promotion as
either pro or contra groups. Thus, when for instance the issue at hand of democracy
promotion concerns a specific act of legislation, it makes sense to divide participating
actors into actor groups that favour or oppose the particular piece of legislation. Each actor
group encompasses a diversity of actors along the dimensions of international versus
domestic, and state versus non-state actors (see Figure 1): international state actors (upper
left of Figure), international non-state actors (upper right), domestic non-state actors (lower
right), and domestic state actors (lower left). The advantages of this conceptualisation are
that the specific country context can be taken into account much more easily by shifting
the focus of the analysis onto the dynamics between pro and contra factions. It also

Daniel Nowack
acknowledges the existence of international actors that do not promote but in fact oppose
democracy, and that actors of the recipient country do not unanimously welcome or refute
democracy promotion. Besides this, a pro and contra distinction also allows us to zoom in
on how the domestic and international actors of each actor group work together in pushing
for their respective goals.
Figure 1: Dimensions of one actor group (pro and contra distinction not included here)
Source: Author
Before the background of democracy promotion as a discourse between pro and contra
factions is dealt with, the question arises as to what eventually influences which group will
succeed in realising its position. Several hypotheses concerning which actor group
succeeds as well as concerning the effectiveness of democracy promotion arise.
2.2.1 Organisational resources
The literature on social movements and civil society shows that access to resources and
resource mobilisation are important factors for societal actors in achieving their goals (see,
for instance, Edwards & McCarthy, 2007; Jenkins, 1983; McCarthy & Zald, 1977). The
same logic is likely to apply to the actor groups involved in the promotion of democracy.
Greater access to resources capacitate actor groups to push for their goals. Resources may
include monetary resources such as funding; physical capital such as buildings, vehicles,
or office equipment; and human capital in the form of well-educated and skilled personnel
as well as knowledge and information − for instance of how legislative processes work.
Therefore, we posit Hypothesis 1, namely that the more organisational resources an actor
group controls, the more likely it will achieve its aims.
2.2.2 Congruence of attitudes
The interactions of actor groups in democracy promotion do not take place in an empty
space. Instead, they take place against the background of popular attitudes prevalent in the

recipient country. For any actor group in instances of democracy promotion, the road to
success leads via the actors that have decision-making power. Hence, an actor group will
aim at convincing these decision-makers, who may be parliamentarians or government
members, or, in an autocratic setting, actors that control the reigns of physical power such
as army generals. Besides attempting to convince them, actor groups also leverage their
arguments by exerting pressure on these decision-makers and power wielders. This
happens through mobilising popular support for the actor group’s position. If the actor
group’s position is already close to attitudes prevalent in the population, mobilisation is
much easier to achieve. In a comparative case study on Kyrgyzstan and Kenya, Fiedler
(2018) finds, for instance, that international democracy support is more effective when its
goals connect to the broader population.
Hence, Hypothesis 2 assumes that the greater the match between an actor group’s
position and prevailing political attitudes at the societal level, the more likely it will
achieve its aims.
2.2.3 The use of material incentives: how values mitigate the logic of
consequences
Hypotheses 1 and 2 make statements about the differential chances for success of whole
actor groups. These actor groups, however, may involve international actors that promote
democracy by using the logic of consequences. The use of instruments of the logic of
consequences likely influences the total actor group’s outlooks on achieving its aim.
Instruments of the logic of consequences, that is, material incentives, are less likely to be
effective if the values promoted by international actors in democracy promotion are
incongruent with the predominant societal values in the recipient country. Attempts to “buy
off” opposition might even weaken the standing of the actor group that applies material
incentives or sanctions. In contrast, attempts work best if a commonality on values already
exists.
Hence, according to Hypothesis 3, the greater the fit between the values promoted by
international actors and domestic social values, the more likely it is that the use of
material incentives will encourage change in the direction supported by the international
actors.
2.2.4 Social status of international actors: how values mitigate the logic of
appropriateness
Fish (2009) argues not only that cultural disposition towards democracy is an important
predictor of the effectiveness of democracy promotion, but that the prevailing attitude
towards the democracy-promoting countries bears importance, too. As most democracy-
promoting countries are “Western”, he reasons that the general disposition of a country’s
population towards “The West” is important for the success of democracy promotion
efforts.
However, although he explains his argument intuitively, he does not flesh out a clear
mechanism. Why should the generally ascribed social status of democracy-promoting actors
in recipient countries make much difference? The differentiation between instruments of the

Daniel Nowack
logic of consequences and the logic of appropriateness provides for an answer. As

mentioned earlier, the logic of appropriateness encompasses calls that address the
recipient’s identity and role as well as the obligations that follow on from these.
Obviously, such calls do work better when they come from actors who are revered in the
recipient country. In contrast, calls by international actors who have only a low standing in
the recipient country, and who might even be mocked, are more likely to be derided. The
recipient country would risk its credibility, if it suddenly heeded the calls of actors of
minor status and prestige. In contrast, following the calls of the more prestigious will
transfer some prestige also onto the recipient country.
Hypothesis 4 thus suggests that the higher the social status attributed to international
actors, the more likely it is that change will occur in the direction advocated by the
international actors.
2.3 Methodology
The outcome variable of interest is the effectiveness of democracy promotion.

Unfortunately, measuring democracy promotion effectiveness is inevitably ambiguous.
The debate on formalism in democratisation exemplifies this well: a successful electoral
process, for instance, does not necessarily mean that democracy in a country is robust.
Instead, evaluations always need to be contextualised.
To capture the effectiveness of democracy promotion, instances in which democracy

promotion could strengthen essential characteristics of democracy as a political system
need to be identified. As rationalised above, the two core dimensions of democracy are
equality, and peaceful political competition. Over the last years there have been two types
of occurrences that are of relevance to these two core dimensions: attempts aimed at
unconstitutional third terms of office; and legal reforms in respect to equal rights for
groups that suffer legal discrimination.
Where the former are concerned, in recent years many young democracies have seen
attempts by incumbent presidents to extend their presidential careers by a third term in
breach of term limits set by the Constitution. Such patterns have been especially
observable in Africa (Posner & Young, 2007; Tull & Simons, 2017). At the time of
writing, attempts to change, or abolish, presidential term limits are becoming visible in
Zambia, Gabon, and Burundi. 1 Such attempts are efforts to remove the constitutional
checks that ensure democratic competition, and thus make for paths towards autocracy.
With regard to legal reforms for equal rights, we choose to focus on gender equality and
LGBTI equality. Gender equality is still an unachieved goal in many countries of the
world. Although this is true for both well-established as well as young democracies, social
institutions tend to be more discriminatory in not yet consolidated democracies and in
non-democracies. According to the Organisation for Economic Co-operation and
Development’s (OECD) Social Institutions and Gender Index (SIGI) 61 per cent of the
1 See Africa Research Bulletin, Political, Social, and Cultural Series, (2018), 55(1) under
https://onlinelibrary.wiley.com/toc/1467825x/2018/55/1.

108 countries surveyed have social institutions in place that discriminate against women
on a medium to very high level. Restricted civil liberties, restricted access to resources and
assets, and a family code discriminating against women, among others, are particularly
prevalent in many African countries (OECD [Organisation for Economic Co-operation
and Development], 2014).
Similar to the state on gender equality, in many parts of the world legal discrimination
against LGBTI persons is still in place. For the year 2017, the International Lesbian, Gay,
Bisexual, Trans and Intersex Association (ILGA) counted a total of 72 countries in which
acts of homosexuality are criminalised. In some of these countries, the punishment for
homosexual acts is very drastic, such as in Saudi Arabia where the penalty is the death
sentence. Although many countries around the world have some kind of protection
measures in place, same-sex marriages are only legal in a minority of 22 countries (Carroll
& Mendos, 2017).
Cases in which presidents attempted to abolish the presidential term limit specified in a
country’s constitution, as well as cases of legal reform for or against more gender equality
posit occurrences, in which the two core features of democracy, namely equality and
competition, are at stake. To evaluate the outcomes of such occurrences for democracy,
we apply the following operationalisation rules. In respect to the third-term cases: failed
attempts to remove a third-term limit by incumbent presidents does no damage to
democracy, while attempts that succeed damage democracy. Attempts are regarded as
failed when the incumbent president eventually aborts his attempt or when a motion to
remove the term limit fails in parliament or other institutions of the political system, such
as the constitutional court. In addition, cases in which an incumbent succeeds in cowing
institutions so that he is allowed to run for another term but in which he loses out against
his election competitor are also regarded as failed third-term attempts. This reflects the
rationale that the electorate has neither been cowed nor deceived and that the incumbent
has eventually been forced to accept the judgement of the electorate. Hence, in such cases
the democratic process worked.
With regard to gender law reforms, any law reform needs to be assessed on its overall
impact, as one legal stipulation in the reform text could improve women’s standing in
society while another one in the same reform text could have the opposite effect. Hence,
an overall assessment of its impact on democracy is facilitated by a case-sensitive
investigation of the net effect of the reform measures in light of its – possibly – competing
stipulations. Concerning operationalisation, we focus on whether the reform enters into
force legally. This means we stop short of investigating the behavioural impact for change
on the ground but concentrate on the behaviour of the actors of the political system.
Hence, a reform is operationalised as successful when it is passed by parliament, assented
to by the head of state, and not “shot down” by any other political institution such as the
constitutional court. The limitation to the aspect of passing a law reform versus also
investigating its actual implementation does no harm to the analysis. This is so because the
final outcome variable of interest is the effectiveness of democracy promotion.
This approach is coherent with our theoretical departure. We are interested in how
democracy promotion can effectively influence behaviour within the political system.
Both the logic of consequences and the logic of appropriateness use their leverage in
forming motivations by addressing and creating reasons for behaviour. Hence, in order to

Daniel Nowack
infer the motivations of actors, it is best to use the methodological qualitative approach of
structured process tracing. Process tracing is derived from cognitive psychology in order
to infer the causal intermediate steps in a process and can well be applied to trace the
impact of ideational factors on motivations (Jacobs, 2015). Bennett and Checkel (2015, p.
6) define it as a method that “attempts to identify the intervening causal process – the
causal chain and causal mechanism – between an independent variable (or variables) and
the outcome of the dependent variable”. It does so by “drawing descriptive and causal
inferences from diagnostic pieces of evidence – often understood as part of a temporal
sequence of events or phenomena” (Collier, 2011, p. 824). These diagnostic bits and pieces
of evidence constitute “causal process-observations” (Mahoney, 2012, p. 571) that in
combination with scientifically backed-up or universally true generalisations uncover the
causal mechanism leading to the dependent variable. The characteristic element of process
tracing is its diagnostic reasoning as illustrated by Collier (2011) in his applied introduction
to process tracing.
Carrying out structured process tracing necessitates a case study design in which “thick
data” on causal-process observations can be gathered and analysed qualitatively. In order
to draw general conclusions from the analysis, this paper will comparatively investigate
both a law reform for equality and a third-term attempt within one country in order to
control for country-specific effects.
To bring in the influence of values and attitudes, secondary quantitative survey data will
be presented. The nature of attitudes as direct positions on questions of social organisation
means that they can be presented with survey data. This contrasts to value orientations that
are more general and abstract dispositions. In order to operationalise value orientations,
we constructed two three-items-based value indices that reflect the value dimensions of
individualism-collectivism (Hofstede, 2003) and of egalitarianism-hierarchy (Schwartz,
1992, 2006). We chose these two value dimensions as they reflect orientations towards the
two essential core characteristics of democracy: competition and equality.
To construct these value indices we identified a number of items from the Afrobarometer
survey (Afrobarometer: A pan-African series of national public attitude surveys on
democracy, governance, and society, 2018) that can be hypothesised to map these two
value dimensions. Because not all of the items identified were asked about in all the same
rounds, it was only possible to use three items for each index. We then applied principal
components analysis to extract an index score for every respondent (see Appendix 3 for
more information on the selected items and the method employed). The indices range
from 0, indicating egalitarian values or individualist values respectively, to 1 indicating
hierarchical or collectivist values respectively.
2.4 Case selection
The within-country comparison limits the population of cases to those countries in which
i.) incumbent presidents at least once attempted to overstep a constitutional term limit, and
ii.) a legal reform of the status of women or LGBTI persons was attempted. In addition,
donors should be active and operative in the country so that the effectiveness of their
democracy promotion activities in the respective cases can be assessed.

For case selection, we compiled data on attempts to extend or abolish term limits as well as
to reform family and LGBTI law in African countries (including North Africa) for the
period 1990 to 2016. We then classified each episode according to operationalisation rules
as either “successful” or “failed”. We identified 48 episodes in which term limits were
reached. Incumbents unsuccessfully attempted to extend their term in 6, and succeeded in 13
of these episodes. Regarding our second case − legal reforms for gender or LGBTI equality
− we identified 172 episodes, 139 of which resulted in legislation being successfully turned
into law. Ninety-one of these episodes were an improvement for women or LGBTI persons,
while 11 were not.
Ultimately Malawi was selected as a country case worth studying more closely. In 2015, it
had passed a legal reform that aimed at harmonising its various marriage and family laws.
In addition, its first democratic president, Bakili Muluzi, attempted to abolish the
constitutionally set limit on his term of office after serving his second term, but eventually
failed to achieve his aims in 2002/2003. What is more: not only are OECD donor
countries very active in Malawi, but the country also presents itself as an important case of
a democratically stable country without a history of past conflict.
To facilitate the process-tracing analysis, we collected and analysed primary newspapers

through online archives for the periods under analysis. Data gathered from the newspapers
was complemented by 21 in-depth semi-structured interviews with representatives of
donor countries, civil society actors, ministries, and members of the Malawian parliament
(MPs) conducted in Lilongwe from April to May 2017 (see Appendix 2).
3 Country analysis: democratic reforms in Malawi
3.1 The political background of Malawi, 1992 to 2017
Shortly after gaining its independence, Malawi was turned into a one-party state ruled by
the Malawi Congress Party (MCP) under the leadership of Dr Hastings Kamuzu Banda,
who declared himself “Life President” in 1970. During his dictatorship, Banda created a
centralised rent-economy in which any spheres outside of the government were drastically
oppressed using a special security apparatus.
As with so many other countries that slipped into autocracy after independence, change
came in the early years of the 1990s. Throughout Banda’s regime, a diaspora of political
opposition had aimed at bringing him down. Throughout the 1980s, Malawi’s economy
had been faltering and distress increased when a drought hit the country in 1992. Church
leaders from all congregations began criticising the regime publicly in pastoral letters
because of mismanagement and oppression. Amidst an emerging atmosphere of criticism,
a major opposition leader returned to Malawi, openly called for multi-party democracy,
and founded the Alliance for Democracy (AFORD). Thus encouraged, students took to the
streets in public protest against autocracy while workers of many different economic
branches went on strike all over the country. When the AFORD leader was arrested by
security forces, this sparked another spike of opposition which eventually forced Banda to
call for a referendum on multi-party democracy in 1993, of which almost two-thirds of the
Malawians voted in favour (Ihonvbere, 1997).

Daniel Nowack
When the first presidential elections under the new multi-party regime were held in 1994,
Bakili Muluzi, leader of the United Democratic Front (UDF) emerged as the winner. The
democratic transition of Malawi had been regarded as an astounding success as it had not
involved much violence and seemed to be carried by popular groundswell. However, it is
often argued that, while Malawi facilitated a transition from autocracy to democracy
formally, no true end was put to the clientelistic political rule established by Banda after
independence. Such a “transition without transformation” (Cammack, 2017, p. 664) made
Malawi vulnerable to backsliding. Subsequently, Malawi faced another crossroads when
Muluzi attempted to abolish the constitutional term limit in 2002, although he did not
succeed in doing so.
Slowing economic growth in 2008 and 2009 meant that the country was heading towards a
critical crisis in 2011, when civil society started to criticise the Mutharika administration
that followed Muluzi’s presidency for mismanagement and misconduct. In response, the
government cracked down on civil society, so that several civil society leaders went into
hiding (Cammack, 2017; Wroe, 2012).
Malawi faced another critical juncture, when in 2012 president Bingu wa Mutharika
passed away due to a heart attack. According to the Constitution, Vice-President Joyce
Banda – who had earlier been sidelined by Mutharika and founded her own party – was to
take over, but during a period of several tense weeks it seemed as if her presidency might
be contested by Mutharika’s followers (Patel & Wahmann, 2015). The Malawi Defence
Forces (MDF) emerged as a critical actor upholding the Constitution, as they openly
declared their support for, and aim to protect, Joyce Banda (personal interview 3). 2After
starting off well and serving two years as Acting President, Joyce Banda was eventually
defeated in the elections of 2014 by Mutharika’s younger brother, Peter Mutharika, who
will face the next elections in 2019.
As can be inferred from this brief historical description, Malawi is a young democracy with
fledgling institutions. It is a Presidential Republic with a unicameral parliament and a single
tier of local government situated at District Level. Its political system is often described as
“defective” (Chinsinga, 2008, p. 11). “Defective democracies” are characterised on the one
hand by sufficiently strong political institutions that can implement democratic elections and
uphold the core functions of the democratic system, but, on the other hand, major
deficiencies exist in the rule of law as well as in horizontal accountability as powerful
cliques tend to dominate the political arena (Merkel & Croissant, 2004).
In economic terms, Malawi is among the poorest countries of the world. In the United
Nations’ Human Development Index it was ranked 170th out of 188 countries in 2015,
coming in just behind Afghanistan. Hence, it is no surprise that Malawi is highly aid
dependent. The most important donors are the European Union, Norway, the United States,
and the United Kingdom. The ratio of received net Official Development Assistance (ODA)
to government expense reached a record high of 112 per cent in 2012. Although this has
subsided since then, the figure for 2015 was still very high at 88.2 per cent. Against this
background, it is also not surprising that donors play an important role in the politics of
Malawi. For instance, during Malawi’s democratic transition in 1994, donors played a key
2 For a list of personal interviews, see Appendix 2.

role in exerting pressure on the Banda regime by withholding aid (Ihonvbere, 1997;
Brown, 2005). However, Malawi’s relationship to donor countries has become strained in
recent years (Wroe, 2012). Donors suspended direct budgetary aid after a major fraud
scandal erupted in September 2013. This had major impact on service delivery and
operation by the government leading to delay in salaries for public servants and a
suspension in the recruitment of new staff, among other things (personal interview 1).
Figure 2 shows the distribution of Malawian respondents’ scores in relation to the value
indices that we constructed. For both value orientations, Malawians tend more towards
egalitarian and individualist values. In both cases, more than half of all respondents have
scores smaller to .5, thus are closer to egalitarian and individualist values than to
hierarchical or collectivist values. Because slightly more than half of all respondents even
score below .4 for the egalitarian-hierarchical value orientation, it seems that egalitarian
value-orientations are slightly more frequent than individualist values compared to their
opposite value-orientations.
Figure 2: Value indices for Malawi: egalitarianism-hierarchy and individualism-collectivism
Source: Author
These results deviate from other assessments of cultural values in African societies. Data
on cultural value orientations for African societies are scarce, and no studies on Malawi
exist to which our cultural value orientation indices can be compared. However, data by
Hofstede (2003) and by Schwartz (2006) suggest that African societies are fairly
collectivistic and hierarchic. The differences between this general assessment and our
index scores probably stem from differences in samples. Both Hofstede (2003) and
Schwartz (2006) use specific samples for their surveys (IBM employees in the case of
Hofstede, and teachers and students in the case of Schwartz), while we used representative
Afrobarometer (2018) survey data.

Daniel Nowack
3.2 Case 1: Muluzi’s attempt at a third term
3.2.1 Process description
Table A1 in Appendix 1 provides an overview of the process of Muluzi’s third-term

attempt in 2002/2003. Nevertheless, this section will provide a brief description of how
the debate on Muluzi’s third term unfolded, and which positions different actors took
throughout the debate.
According to Morrow (2006), the idea that Muluzi might run another term as president
was floated for the first time by representatives of his party (UDF) and civil service
officials just after his election to the presidential office in 1999. But the debate on
abolishing the presidential term limit did not really take off until early 2002, when the
UDF unsuccessfully attempted to change the provision that constitutional amendments
necessitate a two-thirds majority (Hussein, 2004). From then onwards, the UDF started
actively promoting the abolishment of the term limit and the institution of an open end to
presidential terms. The party founded a commission with the goal to advocate for another
term for Muluzi and consult with presidents of other countries who had successfully
abolished constitutional term limits (Banda, 2002b). Muluzi himself publicly stated that he
would not seek another term unless asked so by the people of Malawi (Kanguru, 2002),
but also argued that he had “a lot of development plans that [would] benefit Malawians”
and that he wanted “to finish them” (Morrow, 2006, p. 156). Meanwhile, considerable
pressure was exerted on opposition politicians and opposing businessmen. Allegedly,
many opposition politicians were bought off with either land holdings or cash (Morrow,
2006). Eventually, even the leaders of the two major opposition parties, AFORD and the
MCP, went over to the pro-open term faction after they had negotiated with Muluzi over a
possible government of national unity after Muluzi’s successful additional term bid
(Chinsinga, 2008). Meanwhile, the government issued a ban on demonstrations to curtail
popular opposition led by church and other civil society organisations towards the open
term-bid. Although the ban on demonstrations was declared illegal by a judge of Malawi’s
High Court, Muluzi ordered the people to ignore the ruling and, in response to government
pressure, the High Court overturned the ruling (Morrow, 2006).
Eventually, the Open Term Bill was presented to Parliament in July 2002 as a Private
Member’s bill by an AFORD party member, although it failed to be passed by only three
votes (Morrow, 2006). However, the issue reappeared on the public agenda when the UDF
announced it would propose a modified version of the bill that would not abolish the
presidential term limit but would only extend it by a third term (IRIN [United Nations
Integrated Regional Information Networks], 2002c). After the defeat of the Open Term
Bill, civil society opposition remained alerted. Although the ban on demonstrations still
applied, it was defied by civil society organisations, and allegedly Muluzi ordered his
party followers to intensify buying off opposition MPs (Mwase, 2002b). The Third Term
Bill was eventually gazetted in September, but meanwhile internal dissent cracked open
within the UDF, as leading party members started opposing Muluzi’s conduct and general
party leadership (Mweninguwe, 2002; IRIN, 2002e; Jamieson, 2002). As the internal party
support ebbed away, civil society and donor pressure became stronger throughout the
second half of 2002. Meanwhile part of the third-term debate revolved further around the
government ban on public protests and demonstrations after Muluzi ferociously declared
about whoever wanted to demonstrate that he would “demolish all of [them] together”

(IRIN, 2002d). Violent clashes between demonstrators and police who made use of
teargas and gunfire to disperse the protesters ensued in November (African Church
Information Service, 2002d).
Hence, after much heightened tension at the end of the year 2002 and during a special
parliamentary session in January 2003, Muluzi and his followers had to face the fact that
they could not possibly muster a majority for the Third-Term Bill. Subsequently the bill was
sent back to the parliament’s Legal Affairs Committee for revision, a de facto withdrawal
(Morrow, 2006). Two months later, Muluzi announced that Bingu wa Mutharika as the UDF
candidate running for president in the upcoming elections.
Figure 3 shows the constellation of actors that were active throughout Muluzi’s third-term
bid. Each sphere presents either the pro-third term faction (left) or the contra-third term
faction (right).
The contra-third term faction consisted mainly of civil society, including human rights
organisations such as the Centre for Human Rights and Rehabilitation (CHRR), workers’
associations under the umbrella of the Malawi Congress of Trade Unions, several church
organisations such as the Church of Central Africa Presbyterian (CCAP), as well as
specialised associations such as the Law Society which has ties to academia. It should be
noted that the graphical depiction may distort the size of the contra-faction. This is
especially with regard to civil society organisations (CSOs) which had organised
themselves under two large umbrella organisations, the Forum for the Defence of the
Constitution, and the Human Rights Consultative Committee which was comprised of 50
CSOs alone (Dulani, 2011, p. 125). In addition to these civil society actors, the two major
opposition parties also appear in the contra-faction, as they were effectively split over the
issue of Muluzi’s attempt at a third term after he had succeeded in drawing large numbers
of them into his faction.

Figure 3: Actor constellation during Muluzi’s third-term bid, 2002-2003
Source: Author
In contrast to the contra-third term faction, no international state actors backing the pro-
third term faction were found although some newspapers rumoured that Muluzi received
direct support from Libya’s Gaddafi regime (Jamieson, 2002; Namingha, 2002a) and
similar rumours on support from US American charismatic church groups existed on the
behalf of non-state international actors (Morrow, 2006). However, such claims could not
be further substantiated. Instead, the pro-third term faction consisted mainly of the UDF
and the members of the AFORD and MCP opposition parties that Muluzi and his
followers could convince and draw into their faction. In addition, the pro-third term
faction received some support from charismatic and Pentecostal churches as well as from
some, though not all, Muslim organisations of Malawi. Importantly, the traditional
authorities of Malawi represented by the chiefs’ council also backed Muluzi’s bid after he
had raised their stipends by 25 per cent in the debate leading up to the open term bill
tabling (Morrow, 2006).
From this description of the process, it becomes clear that the proximate cause of why
Muluzi’s bid for another presidential term failed lay in the waning support from both UDF
and opposition MPs. When Malawi’s parliament voted on the Open Term Bill, Muluzi
missed the mark by only three votes, but half a year later conditions had changed
considerably, so much so that he did not even dare to attempt to introduce the Third Term
Bill in parliament. It is worthwhile shedding more light on what turned the tide.
3.2.2 Explanatory factors
Two factors played out particularly during the process of Muluzi’s third-term bid. On the
one hand, civil society, including non-government organisations (NGOs) and other CSOs
as well as church organisations, was very active in mobilising popular attitudes for protest.
On the other hand, the fact that donors warned the Muluzi government about the
importance of adhering to the democratic process complemented protests on the ground.
Civil society & popular attitudes
Throughout the third-term bid, civil society played a major role in expressing and
mobilising opposition to Muluzi. First and foremost were Malawi’s church organisations.
In the early 1990s, these had already been instrumental in fuelling Malawi’s democratic
transition and saw − and still see − themselves as occupying the role of “democratic
watchdogs”. This pertains for instance to the inclusion of contents dealing with good
governance in the educational curricula of future priests (personal interview 13). In
addition, many pastoral letters and statements publicised by the various church
organisations deal with issues of governance and democracy. For instance, already one
year before Muluzi’s third-term bid started in earnest did Catholic bishops publish a
pastoral letter cautioning against manipulation of the Constitution, similarly to CCAP
which published a letter on democratic culture in April 2001 (Morrow, 2006).
When Muluzi’s third-term bid became more apparent in the first months of 2002, the
Catholic Church issued its Lentine Pastoral letter in April which read that “any
amendment to the Constitution should be preceded by broad public consultations and
consensus. On fundamental issues, no change should be done without a referendum”
(Mnela, 2002). The churches also continued to clash with Muluzi after the defeat of the

Daniel Nowack
Open Term Bill when Muluzi accused them of “[exercising] questionable ambitions”
(Mwase, 2002a). The Catholic Commission for Justice and Peace (CCJP), for instance,
published a statement in October 2002 condemning Muluzi’s attempt to push his bid for
another term, calling “upon all the MPs who now hold power to either destroy or build our
nation, to examine their conscience […] by firmly and categorically VOTING NO! NO!
NO! NO! to the third term bill” (“The Catholic Commission”, 2002). The churches’
campaign against Muluzi’s third term was not limited to the Catholic Church only, but
was an endeavour which cut across congregations. The Evangelical Lutheran Church in
Malawi (ELCM) publicly opposed Muluzi’s bid (African Church Information Service,
2002c), while the Public Affairs Committee (PAC) was one of the major actors of the
contra-third term faction, bringing opponents of the third term together in, for instance,
evening dinners (Morrow, 2006). As an umbrella-body including multiple congregations
and faiths, it also includes major Muslim organisations. However, as such, the PAC was not
immune to internal fragmentation either. Fault lines existed, especially between the
Christian and Muslim organisations (Mwanyongo & Kang’ombe, 2002). This led some
country observers to argue that the opposition of Christian church organisations to Muluzi’s
third-term bid was fuelled by reluctance to accept yet another term under a Muslim
president rather than by their concern for the respect of Malawi’s Constitution (Meinhardt
& Patel, 2003). Indeed, representatives of the CCAP interviewed stated that, at least for
some Christian constituencies, Muluzi’s religious congregation played a role in their
opposition to his bid for another term. But Morrow (2006) warns against emphasising
Christian-Muslim rifts too much, arguing that Muslim church leaders had already fallen
out with Muluzi one year before the tabling of the Open Term Bill.
From the very beginning, church organisations cooperated with non-governmental

organisations that were active in the promotion of human rights, such as the Centre for
Human Rights and Rehabilitation (Mnela, 2002). CSOs like this were active in liaising
with the church organisations, academia, the Malawi Law Society, and especially in
awareness-raising in local communities (personal interviews 10; 13). Demonstrations later
in the debate were often organised by NGOs. Thus, the Forum for the Defence of the
Constitution for instance organised the demonstrations in November 2002 that led to
heavy clashes with police forces. Church organisations and NGOs could thereby count on
popular attitudes. According to survey data from the Afrobaromter (2018), 76.33 per cent
of Malawians agreed either very strongly or strongly with having a limit on the number of
presidential terms. This political attitude, which appears to be comparably stable from
2003 to 2014 (see Table 1), probably stems from Malawi’s past experience with Banda’s
“Life Presidency”. It is matched by a strong attitude against autocracy shared by almost
four-fifths (77.92 per cent) of the population. Afrobaromter data also show that Malawians
were split in their attitudes towards Bakili Muluzi as president in 2003, with one-half
trusting the president “not at all” or “a bit”, and another half “somewhat” or “a lot”.

Table 1: Attitudes on term limits and one-man rule

2003 2008 2012 2014
Term limit?
Agree with term limit 13.58% 12.5% 13.96% 6.08%
Agree very strongly with term limit 62.75% 46.58% 62.9% 71.29%
Abolish elections and parliament?
Disapprove 62.17% 68.17% 68.92% 75.5%
Disapprove strongly 15.75% 12.00% 18.28% 8.46%
N 1,200 1,200 2,407 2,400
Notes:
Term limits: “Which of the following statements is closest to your views? Choose Statement A or Statement B:
A: The President of Malawi should be able to serve as many terms in office as he wishes.
B: In Malawi, the President must obey the law, including the constitution, for example by serving no more than two
terms in office.”
Abolish elections and parliament:
“There are many ways to govern a country.
Would you disapprove or approve of the following alternatives? Elections and parliament are abolished so that the
president can decide everything.”
Source: Afrobarometer, 2018
Donor interventions
Tacit donor pressure complemented the protests and demonstrations of the Malawian civil
society. Donors engaged in a “carrot-and-stick” strategy vis-à-vis the Muluzi government
by withdrawing aid, cutting debt and condemning the behaviour of Muluzi’s government
rhetorically.
In the run-up to Muluzi’s third-term bid, Malawi’s relation to traditional donors had
gradually worsened. At the end of the 1990s, donors such as the World Bank, the
International Monetary Fund (IMF), and the United Kingdom had pushed for reforms
towards greater transparency and accountability (Brown, 2005). But, finding their
demands not met satisfactorily, the EU, the United Kingdom and the United States ended
direct budget support in 2001 (Banda, 2001). Denmark took a more extreme measure in
early 2002, withdrawing not only budget support but all aid (Banda, 2002a), citing as the
reason that “political developments continue to move in a negative direction” (IRIN,
2002a) while the Malawi government accused Denmark of cutting aid due to a change of
government in Copenhagen. Other donors, such as the United Kingdom and the IMF took
the opportunity to warn Muluzi’s government about economic mismanagement and a lack
of transparency as well as governance (“Britain cautions Muluzi”, 2002; Ganthu, 2002).
Much of how donors talked in the first half of 2002 was hence rather stiff as they argued
that they were waiting for the results of an IMF assessment on Malawi’s economic and
governance performance before deciding whether to release further aid (African Church
Information Service, 2002a).
Amid this tense atmosphere, several donors reverted their course in the summer of 2002 by
cutting Malawi’s debt. This reversion was announced with rhetorical but allegedly not
causal connection to Muluzi’s bid for another presidential term. After a warning
condemnation by the EU of how the third-term debate was handled by the Muluzi

Daniel Nowack
government, both Germany and Italy cut proportions of Malawi’s debt while at the same
time warning Malawi’s government to heed the democratic participatory process (IRIN,
2002b; “Germany, Italy write-off”, 2002). By the same token, the United States warned
Malawi against taking autocratic measures to suppress popular participation (African
Church Information Service, 2002b), while the United Kingdom joined the choir just briefly
after the Open Term Bill was defeated in parliament by stating that aid to Malawi was not
“lost”, yet without referring to the third-term debate (Chafunya, 2002). According to
Resnick (2012), who cites information by the Economist’s Intelligence Unit, donors also
took more explicit steps, threatening Muluzi’s government with additional aid cuts if the
ban regarding demonstrations on the third–term issue was not lifted. However, his claim
could not be substantiated. Morrow (2006), too, argues that donors like the United States did
not only exert open, but also concealed, pressure on the Malawian government. Opposition
party leaders and civil society activists, too, emphasised the role donor pressure played
during the third-term debate (Namingha, 2002b; IRIN, 2002d; 2002e).
The role of donors also comes into play in supporting civil society organisations belonging
to the contra third-term faction. Many of the NGOs who rallied to this faction were directly
funded by donor governments, such as the CHRR which received support from the United
Kingdom and the EU at the time (personal interview 10). However, this partly led to serious
backlashes against the activity of such NGOs as they were accused of acting as “foreign
agents” (“Donors fuel political feud”, 2002). By the same token, the funding that contra-
third term organisations such as the Forum for the Defence of the Constitution, the PAC, or
the CCJP received from the Norwegian Church Aid was attacked as foreign meddling, and
official state donors were pressured to declare that their funding was not specifically tied to
political activist campaigns (“Donors fuel political feud”, 2002). Such suspicions were
further fuelled by the actions of the Danish development organisation Danida, which had
withdrawn completely from Malawi in January 2002 but then channelled funding to the
PAC for campaign purposes through a Zambian NGO (Morrow, 2006).
3.3 Case 2: reform of the marriage law
3.3.1 Process description
The second case selected for study is Malawi’s Marriage, Divorce, and Family Relations
Act which outlawed child marriage and harmonised existing legislation on marriage and
family law, thus slightly improving the legal standing of women (see Table A2 in
Appendix 1 for an overview of the process).
In the course of Malawi’s transition to democracy, the female MPs in Malawi’s parliament
(a minority) had formed an informal group that cross-cut parties and advocated in
parliamentarian debates and dealings for women’s and children’s rights. This informal
group was formalised as the parliament’s Women’s Caucus in 1996 (Chiweza, Wang, &
Maganga, 2016). Two years later, civil rights activists backed by support from the United
Nations Development Programme (UNDP) and seconded by the Malawian government
founded the Gender Coordination Network, an NGO which collaborated extensively with
the female MPs of the Women’s Caucus.
These actors were vital in putting the Marriage, Divorce, and Family Relations Bill on track.
One of the Women’s Caucus goals, as declared in its by-laws, was the harmonisation of

laws pertaining to women’s rights (Chiweza, Wang, & Maganga, 2016). To this end, the
Malawian government instituted a Special Law Commission on Gender, (henceforth
called the Gender Commission) in September 2001 (Law Commission, 2006). 3 After its
foundation, the Gender Commission conferred with the Women’s Caucus and civil society
stakeholders for women’s rights to identify which parts of Malawi’s legislation on
women’s rights needed to be reformed. The resulting Gender Law Reform Programme
identified succession and inheritance, marriage and divorce, and general gender equality
as major areas and laid out a work programme spanning ten years (Mwale, 2012). In the
next four years, the Gender Commission carried out a review of existing laws, as well as
consulting NGOs and local communities (personal interview 3).
The Gender Commission’s work resulted in a report with an accompanying draft bill that
harmonised the disparate legislative acts on marriage and divorce in Malawi (Law
Commission, 2006). The stated goal of the draft bill was to harmonise legislation on the
three forms of marriage prevalent in Malawi: the statutory marriage, the customary
marriage, and the marriage by repute or permanent cohabitation. Although the draft bill
did not create full harmonisation – as the legislative texts on the three forms of marriage
were simply “bundled” into one bill rather than harmonised – it nevertheless proposed
three important legal changes: These were i) an improvement of the rights of women in
marriage by repute or permanent cohabitation; ii) the criminalisation of polygamy; and,
iii) by raising the legal marriage age from 15 to 18, the criminalisation of early marriage. 4
Especially the two later propositions, outlawing polygamy and child marriage, rendered
the debate on the Marriage, Divorce, and Family Relations Bill value-sensitive. Both
polygamy and child marriage is common and deeply rooted in Malawian culture (MHRC
[Malawi Human Rights Commission], 2006). Thus, reform of the family and marriage law
in Malawi faced substantial political inertia. Hence, when the Marriage, Divorce, and
Family Relations Bill was introduced to parliament for the first time in 2010, Parliament
sent the bill back for revision to the Gender Commission despite a comfortable
government majority (CHRR [Centre for Human Rights and Rehabilitation], 2011).
For the next two to three years, changing the bill lay dormant. Although civil society and
non-governmental organisations kept on campaigning against child marriage, no concerted
efforts to reintroduce the bill were attempted. However, momentum gathered after UN
Women, the United Nations organisation promoting gender equality and women
empowerment, opened its Malawi office and became operative in 2012/2013. In
3 Malawi’s democratic Constitution sets out a programme to carry out a wide and comprehensive review
and reform of its existing laws. To achieve this, the Law Commission was founded in 1996 as a
parastatal body with the mandate to review all laws, and draft suggestions for reform. This is not done
by the Law Commission itself but by specially instituted thematic commissions in which various
stakeholders and experts, such as academics, church and civil society members, and ministry officials,
sit seconded by programme officers of the Law Commission (personal interview 3).
4 Legally speaking, marrying below the age of 18 does not constitute an act of child marriage if the
legally specified age of the child is not set at 18 but at 15, say. Although the United Nations Convention
on the Rights of Children specifies 18 as the age of transition from childhood to adulthood, this only
applies if a majority is not achieved earlier under national law (UNOHCHR [United Nations Office of
the High Commissioner of Human Rights], 1989). Nevertheless, in much of the academic and non-
academic literature, child marriage denotes marriage under the age of 18 (see, for instance, Lee-Rife,
Malhotra, Warner, & McGonagle Glinski, 2012). This same definition is applied in this paper.

Daniel Nowack
cooperation with the Ministry for Gender, UN Women pushed the Marriage, Divorce, and
Family Relations Bill back onto the legislative agenda. By mid-2013, legislation to reform
the marriage age was again in discussion in the media (personal interview 15). UN
Women and the Gender Ministry consulted with both parliament’s Women’s Caucus and
the Social Affairs Committee, and set up meetings with the other parliamentary committees
as well as consultations with the Gender Commission and actors from civil society.
Eventually, the bill was tabled in the Cabinet in October 2014, and subsequently
introduced to parliament by the Ministry of Justice. A majority of MPs expressed their
concern about the stipulations of the bill that would criminalise polygamy (personal
interviews 2; 16). Because it was feared and sensed that the bill would not gain approval
by Parliament if the stipulations on polygamy remained, they were dropped before the bill
was tabled in Parliament in February 2015. On the day of the vote, MPs voted on the bill
in a roll call. The presence of civil society representatives on Parliament’s grandstand who
“kept their eyes” on the MPs voting below suggests that some measure of social control was
exerted (personal interviews 2; 5). In the end, the bill was passed unanimously as opposing
MPs, who demanded more space and time for discussion, abstained from the vote.
After it had been passed by Parliament, President Peter Mutharika signed the Bill in April
2015. However, the discussion on it did not subside but continued with the issue that it
stood in conflict with Malawi’s Constitution (“New marriage law deemed invalid”, 2015).
The Constitution still stipulated that children are persons younger than 16 years, and that
marriage between persons between 15 and 18 is possible with parental consent
(Government of Malawi, 2016). Changes made to the Constitution in 2010 made it possible
to amend some of the Constitution’s subsections through a two-third’s parliamentary vote
rather than through popular referenda. Hence, after passing the Marriage, Divorce and
Family Relations Act, the focus of CSOs and NGOs advocating against child marriage
shifted towards amending the Constitution through parliamentary vote.
In order to exert pressure for the constitutional amendment, some Malawian civil society
organisations filed a complaint with the African Committee of Experts on the Rights and
Welfare of the Child (ACERWC), a body of the African Union (AU). During the November
2016 AU summit meeting in Gambia, representatives of the Malawian Ministry of Justice
responded to the complaint and pledged that Malawi would align its Constitution with
regional and international standards (personal interview 11; Human Rights Watch, 2016).
By mid-February 2017, Parliament voted into effect the Constitutional Amendment Act No.
3 raising the age of the child to 18, receiving presidential assent in April of the same year.
The debate about the Marriage, Divorce, and Family Relations Act was characterised by a
large pro-reform alliance made up of government agencies, donor institutions, and inter-
and transnational actors. Part of their success was due to the fact that they faced only
fragmented opposition that had not yet coordinated itself (see Figure 4). Central domestic
state actors were the parliamentarians of the Women’s Caucus seconded by the Social
Affairs Committee, as well as the ministries and the Law Commission who are by
definition key players in the formal law-drafting process. Local chiefs, formally belonging
to Malawi’s government structure, also played a key role. Through enacting by-laws that
forbade child marriage in their districts, many chiefs prepared the ground for the Marriage,
Divorce, and Family Relations Act in MPs’ local constituencies. On the non-state
domestic side, a broad and varied coalition of civil society actors was motivated to push

for law reform but faced substantial transaction and coordination costs. This broad
coalition found important counterparts in international, non-state actors, especially among
UN agencies as well as international NGOs. Some of the non-state domestic civil society
organisations also strategically instrumentalised access to bodies and organisations of
regional and international law such as the Southern African Development Community
(SADC), the AU, or the UN Human Rights Committee. Important for the work of all
domestic pro-reform actors, either state or non-state, were traditional state-donors who
supplied them in particular with funding and financial support.
In contrast to the dense web-structure of the pro-reform actors, the opposition actors were
less organised. Opposition within parliament focused mainly on a handful of MPs who
opposed the bill for various reasons. Members of the Legal Affairs Committee for instance
opposed the bill on the ground of its being in conflict with the Constitution. The Law
Society, a Malawian association of lawyers also raised this concern. Then again, the Law
Society cooperated with some of the pro-reform civil society organisations on legal issues
and questions, hence blurring the line between both camps (personal interview 1; 2; 6; 16).
Figure 4 illustrates this with the dashed lines that connect some of the opposing actors
with the actors favouring reform of the law. Several domestic civil society organisations
criticised the Act for not making provisions for gay marriages (Fahamu, 2015). Popular
criticism as well as concerns were sometimes voiced during consultations, both those
undertaken by the Law Commission but also in the media, for example in call-in radio
shows (personal interviews 2; 3; 5; 9). The main opposition faced by the pro-reform camp,
however, originated among individual parliamentarians.
3.3.2 Explanatory factors
Three explanatory factors stand out in explaining the protracted but eventual success of
the Marriage, Divorce, and Family Relations Act. First, the process of the Marriage,
Divorce, and Family Relations Act played out against a background of domestic changes
in attitudes and practices. Next, it was embedded in an international climate that opposed
child marriage. International organisations and treaties propelled the diffusion of anti-child
marriage norms. Finally, international organisations and non-government organisations as
well as donors played an important role in capacitating domestic actors.
Domestic attitudes and practices
While the attitude of the masses is fairly favourable, there still seems to be strong opposition
in some parts of Malawian society towards gender equality. Set against this, a change in the
attitudes of the cultural elite, Malawi’s traditional chiefs, was crucial in supporting the
Marriage, Divorce, and Family Relations Act. According to Afrobaromter data, attitudes in
Malawi towards gender equality are split but show a tendency towards supporting equal
rights for women. As Table 2 shows, the proportion of Malawians supporting gender
equality rather than traditional, customary laws for women oscillated between 57 per cent
and 72 per cent for the period 2003 to 2012. On the flipside, this meant that two in five
(2005) or one in four (2012) Malawians still tended to oppose gender equality. Survey data
from the more recent Afrobarometer round of 2018 shows strong support for equal access to
land for women (80 per cent). Overall, this suggests that the application of customary law to
women’s rights is a vanishing minority opinion in Malawi.

Figure 4: Actor constellation during the process of the Marriage, Divorce, and Family Relations Act
Source: Author
Table 2: Attitudes on gender equality

2003 2005 2012
Customary
29.75% 41.58% 26.98%
(Statement A)
Equality
68.42% 56.59% 71.86%
(Statement B)
N 1,200 1,200 2,407
Notes:
Question: Which of the following statements is closest to your view? Choose Statement A or Statement B.
A: Women have always been subject to traditional laws and customs, and should remain so.
B: In our country, women should have equal rights and receive the same treatment as men do.
Source: Afrobarometer, 2018
This is seconded by the findings of the consultations undertaken by the Gender

Commission during the drafting process of the Marriage, Divorce, and Family Relations
Act. Many community leaders and members stated that, although early or polygamous
marriage was customarily practiced in their respective region, they actually would not
prefer it for their own children if they had a choice (personal interview 3).
However, these attitudes do not square well with the high occurrence of child marriage in
Malawi, where the proportion of women married before their 18th birthday was 47 per
cent in 2015 (Demographic and Health Survey, 2017). Community leaders attempting to
forbid child marriage in their constituencies also reported strong opposition within
communities, such as the threat of physical attacks (personal interview 17), suggesting a
strong preference for women’s traditional roles within some parts of the Malawian
population. A possible explanation for this contradicting evidence is that attitude-
behaviour correspondence is not as robust and linear as hitherto assumed, a fact that is
often reported in much of social psychology literature (refer, for instance, to Itzchakov,
Uziel, & Wood, 2018; Jerolmack & Khan, 2014; Glasman & Albarracín, 2006).
Besides the attitude of the masses, an important change in attitude towards child marriage
concerns Malawi’s chiefs. In the years running up to the passing of the Marriage, Divorce,
and Family Relations Act, Malawian chiefs increasingly took an anti-child marriage
stance. Theresa Kachindamoto, the senior chief of Dedza district, for instance won
national and international acclamation as the “terminator of child marriage” (“Fearsome
chief”, 2016). In addition, a number of chiefs publicly campaigned against the practice of
child marriage (Witmos, 2014). In 2014, chiefs launched a public campaign to increase the
marriage age to 21 (Masina, 2014). The likely origin of the chiefs’ commitment to end
child marriage is threefold: Just fourteen years ago, chiefs had been publicly derided as
“deathmongers” (Ligomeka, 2000) by the wife of the then Malawian president due to their
implications in both child marriage and polygyny, as well as the related sexual initiation
and cleansing rituals found in some parts of Malawi. Such “public shaming” may have
exerted pressure on these traditional authorities and their successors. Second, since then a
new generation of traditional chiefs has succeeded the older generation. Not only is this
new generation of chiefs younger and in many instances more educated, but the proportion
of female chiefs has risen as well (personal interviews 12; 17; Muriaas, Wang, Benstead,
Dulani, & Rakner, 2017). Finally, these chiefs received much assistance from national and
international NGOs and organisations to draft progressive by-laws forbidding child

Daniel Nowack
marriage in particular (personal interviews 12; 17). Set against this backdrop, elite
attitudes − particularly those of chiefs who are generally revered as “custodians of culture”
(personal interview 17) − were an important precondition for the passing for the Marriage,
Divorce, and Family Relations Act.
International norm diffusion
Changes in attitudes and behaviour concerning gender equality − and child marriage in
particular − also need to be put into the larger regional and international context of on-
going norm diffusion. The map in Figure 5 indicates which countries in Africa had
criminalised child marriage by 2015. It illustrates that by that year most countries in
Africa already had legislation in place that prohibited child marriage. In fact, Malawi
stands out as one of only thirteen African countries that had not yet prohibited child
marriage legislatively. 5 Even if one questions to what degree national child marriage
prohibitions are actually enforced − and, hence, taking into account that some legislation
is mere lip service − the map illustrates that anti-child marriage norms had already been
diffused regionally.
Figure 5: Criminalisation of child marriage in Africa in 2015
Source: Gender statistics retrieved from World Bank, 2018
5 At that time, Malawi was, of course, just about to criminalise child marriage as the Marriage, Divorce, and
Family Relations Act was passed in early 2015. Nevertheless, as discussed above, a certain ambiguity
about the validity of the new legislation existed until the constitutional amendment was passed two years
later.

Consequently, within the regional organisations in which Malawi is represented, such as

the AU and SADC, child marriage became an often discussed issue (personal interview 2).
Representatives of civil society organisations stated that Malawi’s reform of the law
gained the country reverence in regional fora (personal interviews 1; 2; 5), showing that
the fact that child marriage was on the agenda regionally rendered it more salient and
provided an incentive towards reform.
In addition, anti-child marriage norms were diffused through international agreements and
global campaigns initiated by international organisations and these, in turn, exerted
pressure on Malawian decision-makers. In February 2013, the international movement
“Girls not Brides” opened a chapter in Malawi. Two years later, UN Women declared
President Mutharika champion of the global “He For She” campaign. International
organisations such as the United Nations Population Fund (UNFPA) and Human Rights
Watch pressurised the Malawian government publicly to reform the legal marriage age
(Mwalwanda, 2014; Masina, 2014). Domestic actors not only adopted such campaigns but
launched ones of their own. In November 2014, for instance, the Gender Ministry jointly
with the Malawian NGO “Youth Net and Counselling” (YONECO) launched the “End
Child Marriage” campaign.
National CSOs and NGOs used regional and international norm diffusion as a leverage
point for pressuring the government. They participated in the drafting of regional rankings
to create “naming-and-shaming” effects (personal interview 2). The Malawian NGO
Gender Coordination Network, for instance actively participated in the Southern African
Gender Protocol Alliance and, in collaboration with other regional NGOs, drew up
country- and region-specific reports that monitored the implementation of the SADC
Gender Protocol (see, for example, Kaliya, 2015). Likewise, the Centre for Human Rights
and Rehabilitation (CHRR) submitted alternative and shadow reports to the United
Nations Human Rights Committee and, together with the Centre for the Development of
People (CEDEP), to the African Commission on Human and Peoples’ Rights (CHRR &
CEDEP [Centre for the Development of People], 2013; CHRR 2011). Similarly,
Malawian CSOs and NGOs pushed the amendment to change the age of the child in the
Constitution by petitioning the African Union. In this way, domestic civil society made
use of norm diffusion at the international and national level by pressing the Malawian
government to harmonise its domestic legislation and align it with international and
regional agreements and treaties. These provided vital focal points for domestic CSOs and
NGOs and thereby lowered their coordination and transactions costs.
International donors and organisations
International donors and organisations played a key role in pushing forward the Marriage,
Divorce, and Family Relations Act. Their importance played out especially in capacitating
domestic actors, both civil society and state actors, to mobilise for reform of the law. Their
contribution took the form of transfer of expertise and technical assistance, funding and
partnership-based financial support, as well as lobbying and on-the-ground mobilisation.
Conventional international donors, such as OECD donor countries, played a large role in
passive funding. Many domestic CSOs and NGOs as well as state agencies like the Law
Commission are to a very large degree funded by donor countries. It was therefore no
surprise that interviewees unanimously agreed that without these external funds their

Daniel Nowack
organisations would not have been able to do their work. Funding from donor countries
also reached domestic organisations and agencies through their partnership with
international organisations. In such instances, donor countries funded an international
organisation that would use these funds in their collaboration with domestic CSOs and
NGOs. UN Women, for example, received a grant from the Norwegian Embassy and
employed these funds in their cooperation with domestic NGOs and CSOs (personal
interviews 12; 14).
While traditional donors took a more passive role, international organisations such as the
UN agencies were free to adopt a more active role as they are often not traditional funding
organisations. This allows them to transfer technical assistance and expertise to state and
civil society organisations. CSOs, NGOs, and international organisations assisted many
chiefs in setting up progressive by-laws prohibiting child marriage. In turn, CSOs and
NGOs received background information about how similar law reforms had been
undertaken in other countries (often from the SADC region) as well as statistical
information from international organisations. International organisations also helped these
domestic organisations to analyse and create input for the Marriage, Divorce, and Family
Relations Bill. Finally, concerning decision-making in parliament, while in affluent
countries MPs can often rely on an extensive staff of assistants, MPs in Malawi do not
have offices, let alone assistants; in addition, they may be relatively uneducated or even
illiterate (personal interviews 2; 6). In collaboration with domestic CSOs and NGOs,
international organisations played an important role in this regard by setting up briefing
meetings and workshops with parliamentarians to clarify and interpret the draft bill
(personal interviews 6; 7; 12; 16).
Finally, international organisations also played a major role in lobbying for the Marriage,
Divorce, and Family Relations Act. They were crucial in engaging the media, both print and
− for popular mobilisation even more important − broadcast media. How sincere media
outreach was viewed is exemplified by UN Women contracting a Malawi-based PR firm to
manage some of its communications work during the period leading up to the Marriage,
Divorce, and Family Relations Act (personal interview 20). Besides media outreach,
lobbying also took the form of face-to-face lobbying with essential decision-makers. Hence,
country-representatives of international organisations had personal interaction with, for
instance, the president and ministers on the issue (personal interview 12). It is likely that
representatives of donor countries, too, referred to the bill in their communication with
decision-makers from higher echelons (personal interviews 14; 21). Finally, international
organisations also contributed a great deal by connecting all stakeholders and bringing them
together. They engaged CSOs and NGOs, ministries, and the parliament and convened
meetings for strategic planning and briefing. They also facilitated exchange within the
various different parliamentary groups, enabling the Women’s Caucus and the Social
Affairs Committee to pitch the Marriage, Divorce, and Family Relations Bill to other
parliamentary committees. Hence, changing exact contribution of international organisations
was twofold here: First, they facilitated coordination of different actors. Second, by funding
venues and other resources, they provided the logistical resources necessary to hold such
meetings (personal interviews 2; 6; 11; 12; 16).
Many interviewees mentioned the key role that UN Women took in these processes. While
changing reform of the law cannot ultimately be ascribed to the agency of only one
organisation, it seems as if UN Women was “at the right spot, at the right time”. It made a

major contribution by lowering transaction and coordination costs between the diversity of
actors that were already pushing for the criminalisation of child marriage. Although the time
was ripe, it seems as if these actors had hitherto not managed to form a broad coalition. The
failure of Parliament to pass the Marriage, Divorce, and Family Relations Bill in 2010
supports this view. This failure took place despite the fact that the parliament of that time
was new and keen to pass almost any legislation as the previous generation of
parliamentarians had fiercely opposed any legislative initiatives coming from the govern-
ment. The process leading up to the actual passing of the bill in 2015 gathered momentum
after UN Women became active in the country in 2013. After that, UN Women − as a major
coordinating player − collected up the “loose threads” and brought them together.
4 Appraising the hypotheses: explaining the effectiveness of support to

democracy
The hypotheses can be evaluated in light of the insights gained through the process
analyses. In each case, a number of factors contributed to the effectiveness of support for
democracy. But it should be kept in mind that each case presents contextual uniqueness,
especially with regard to differing time horizons. Donor activity in Muluzi’s third-term
attempt to a great degree took the form of an intervention on short notice. In contrast, in
the case of the reform of marriage law, donor activity was embedded in a lengthy process
of mobilisation and sensitisation. International donor activity, however, seems to have
been more decisive in the case of the Marriage, Divorce, and Family Relations Act.
4.1 Organisational resources are a necessary but not a sufficient condition for
effective democracy support
According to Hypothesis 1, the more organisational resources an actor group controls, the
more likely it is to achieve its aims. Only the case of the Marriage, Divorce, and Family
Relations Act supports this hypothesis directly, while the case of Muluzi’s third-term bid
refutes it. Taken together, the results emphasise the necessary trade-offs between an actor
group’s size and its capacities. This trade-off eventually means that resources are only a
necessary but not a sufficient condition for actor groups to achieve their aim.
In the case of Muluzi’s third-term attempt, the advantage of greater organisational

resources lay with the Muluzi government. Although the opposing actor group consisting
of civil society actors received their resources from international donors, the vote-buying
undertaken by the Muluzi government clearly shows that his faction was at greater liberty
to use resources – in this case in the form of money as the incumbent president Muluzi had
large amounts of the state’s resources at his disposal to further his own cause. In contrast,
the opposition faction relied on donor spending which was volatile and less flexibly
accessible. Against this background, the fact that Muluzi almost succeeded in getting the
Open Term Bill passed in 2002 testifies to the importance, but also to the limits, of
organisational resources.
Likewise, the case of the Marriage, Divorce, and Family Relations Act supports this
conclusion. Here the broad coalition of civil society and non-government organisations

Daniel Nowack
had an edge over the opposing faction in terms of resources. This advantage accrued
simply due to the size and scope, as the actor constellation diagram in Figure 4 illustrates.
Despite this edge over the opposing faction, however, the first attempt to get the Marriage,
Divorce, and Family Relations Bill passed in 2010 failed. In contrast, the bill was finally
passed in 2015. In between these two points in time, UN Women became operative in
Malawi in 2012/2013. As shown in the analysis, the organisation had a particularly key
role in rallying the broad coalition of international and domestic civil society and non-
government organisations, state institutions, and MPs. This further supports the fact that,
although organisational resources are crucial for actor groups in pursuing their goals,
resources alone do not suffice. In fact, resources need to be matched by effective
coordination within the actor group.
4.2 Popular attitudes mediate effective democracy support
Hypothesis 2 argues that of two opposing actor groups the one that advocates for a
position closer to domestic attitudes of the masses is more likely to succeed. Both cases
analysed here support this hypothesis.
In the first case, Muluzi’s third-term bid ran against the popular democratic mass attitudes
of Malawians. Muluzi’s bid came only ten years after Malawi had experienced a rare
peaceful transition from autocracy to democracy. Autocracy in Malawi had come about
precisely in a situation similar to Muluzi’s third-term bid when Kamuzu Hastings Banda
had himself declared “Lifelong President” in 1970. Consequently, as survey data shows,
many Malawians are wary of the accumulation of power and prefer limited presidential
terms to unlimited presidential terms (Table 1). Despite these popular attitudes, Muluzi
almost succeeded in pushing through his Open Term Bill. This illustrates the fact that
mobilisation is needed for popular attitudes to be expressed. It also raises questions about
historical contingency. Abolishing term limits completely is naturally a less opaque risk to
democratic political competition than extending the term limit by one year. If Muluzi had
not started his third-term bid off with the Open Term Bill, but with the later Third Term Bill
that eventually withered away, he might have had succeeded in securing another term. This
reasoning is counterfactual and, hence, should be taken with a grain of salt. However, it
suggests the further hypothesis that popular attitudes and their mobilisation play out more
strongly in less opaque issues than in issues where the consequences are hard to assess.
In the second case, clear support of Malawians in favour of gender equality attests to the
importance of popular attitudes. Comparing the issue of child marriage criminalisation
with the criminalisation of polygamy stresses this further. Although no clear-cut
attitudinal data on child marriage and polygamy exists, the process analysis of the
Marriage, Divorce, and Family Relations Act suggests that popular attitudes opposed
criminalising polygamy rather than criminalising child marriage. Because attempting to
pass the Marriage, Divorce, and Family Relations Act with the stipulation criminalising
polygamy still in it would likely have raised the stakes for decision-makers, the
criminalisation of polygamy was dropped from the draft bill in the course of the legislative
process. Although it remains speculative, it seems that the polygamy issue would have
trumped the child marriage issue. Hence, the case of the Marriage, Divorce, and Family
Relations Act illustrates how important it is for the effectiveness of democracy support to
factor in local attitudes and to contextualise them.

4.3 A value-fit between donor and recipient countries seems to condition the
effectiveness of instruments of the logic of consequences, but definitive
evidence is missing
The third hypothesis argues that the effective application of instruments that belong to the
logic of consequences, for instance material incentives and sanctions, is conditional on
whether there is a value-fit between donor and recipient country. The case of Malawi’s
third-term debate provides suggestive evidence for this, but similar evidence for the case
of the Marriage, Divorce, and Family Relations Act is much weaker.
In the case of Muluzi’s third-term bid, a value-fit between donors and broader society as
an aggregate existed. Donors clearly warned the Muluzi government to heed democratic
principles throughout the process, for example by warning it not to oppress
demonstrations violently. Malawians, civil society organisations, as well as state actors
such as the High Court in its ruling on the legality of demonstrations also defended
democratic rights and principles. Within this context, public statements, the threat, and the
actual withdrawal of aid proved to be effective, although they may have been only one
factor that played into the outcome of debate. Donor countries and organisations became
more active after the Open Term Bill had failed narrowly in Parliament in the summer of
2002. When Muluzi’s planned Third Term Bill was not tabled during the October session
of Parliament, some media ascribed this to international pressure from donor
organisations. Eventually, by the end of the year, support for Muluzi’s bid had fizzled out.
Although the ultimate reason why the Third Term Bill had not been tabled during the
October session of Parliament − as well as whether it would have been passed if it had
been tabled − is not clear, the sequence of events nevertheless provides suggestive
evidence that donor interventions played a role, at least to certain extent.
The case of the Marriage, Divorce, and Family Relations Act unfortunately does not serve
well as a comparison. For this, a value-misfit between donor and recipient would be
required, but whether such a misfit concerning gender equality existed is not as clear-cut as
it might seem at first glance. Donors certainly advocated for gender equality, and according
to survey data, Malawians too preferred gender equality to traditional gender roles (Table 2).
The limited opportunity for donors to use conditionality instruments in the case of the
Marriage, Divorce, and Family Relations Act complicates a direct comparison of the two
cases. At the time when the Act was eventually passed, donors had already suspended
direct budgetary support. This meant that, in contrast to the case of Muluzi’s third-term
bid, they did not have the opportunity to cut the kind of financial support that would have
felt most afflictive to the government.
Despite this, interviewees stated that mechanisms of reputation tied to development

support were working implicitly in favour of the criminalisation of child marriages. Such
reputation mechanisms worked through naming-and-shaming in regional and
international changing as well as through accusing the Malawian government for not
adhering to the agreements and treaties it had formally agreed to:
So, it is on that understanding that [SADC] member states will not be punished for
not doing it [complying with the SADC gender protocol], but they face bad
consequences for not doing it. The consequences can be many: the naming-and-
shaming, and when you start being named-and-shamed other countries or the

Daniel Nowack
development partners or those that are helping the country will start saying: ‘This
country is not a country we should be investing in’. (Personal interview 2)
Hence, even if not directly sanctioned, governments are still subject to the implicit
incentive to comply with international standards. This translates into incentives to produce
a record of progressive legislation such as the Marriage, Divorce, and Family Relations
Act. The moment of passing the Marriage, Divorce, and Family Relations Act provides
some support for this. As some interviewees stated, international pressure on Malawi was
rising in 2015. Direct budgetary support had already been suspended for two years,
Moreover Malawi was regularly presented as one of the countries with the highest
incidence of child marriages internationally, and the country had just failed to implement
the Millennium Development Goals. This time-specific context might offer one part of the
answer to the question of why the Marriage, Divorce, and Family Relations Act was
passed in 2015 but not earlier in 2010.
Nevertheless, because a value-fit existed in both cases, the analysis can only assess that
instruments of the logic of consequences do work when there is a value-fit, but cannot
verify this vis-à-vis the opposite case.
4.4 Instruments of the logic of appropriateness work if the donor enjoys high
esteem, but this is sensitive to context
Hypothesis 4 states that instruments of the logic of appropriateness are more effective
when they come from donors that enjoy higher esteem in the recipient countries. As it is
difficult to trace variation in attitudes towards donors in Malawi, the analysis provides
only tentative evidence for this hypothesis. However, the cases also show that the logic of
appropriateness works generally, but that its effectiveness is context-sensitive to the exact
issue at hand.
Aggregate Malawian attitudes towards donor countries and international organisations

are difficult to decipher. According to data collected by the Afrobarometer, most
Malawians have uninformed opinions on both the degree to which donors are assisting
their country and on whether international donors and NGOs should have more or less
influence (Tables 3 and 4).
Table 3: Attitudes towards donors in Malawi

International United European United African
donors/NGOs States Union Nations Union
Do nothing 3.02% 5.01% 5.15% 4.83% 5.56%
Do a little bit 5.9% 2.53% 2.52% 4.39% 5.08%
Do something 10.2% 11.27% 7.42% 9.31% 7.99%
Do a lot 16.26% 23.51% 24.03% 16.45% 12.29%
Don’t know 64.62% 57.6% 60.81% 65.02% 68.77%
Notes:
Respondents were asked:
“In your opinion, how much do each of the following do to help your country, or haven’t you heard enough to say?”
(Relates to international donors and NGOs apart from UN agencies).
Source: Afrobarometer, 2018, Round 4, 2008/2009

At best, the large proportion of participants who answered “Don’t know” indicates
indifference towards donor activity in Malawi. At worst, the response category hides
aversion towards donors. Focusing only on the remaining response categories, attitudes
towards donor activity seems to be balanced although tending towards a positive view of
donors. Thirty-five per cent of Malawians think that the United States assists their country
a lot or somewhat, followed by the EU (32 per cent), other international donors and NGOs
(27 per cent), the United Nations (26 per cent), and the AU (20 per cent). Similarly,
respondents are divided on whether international donors and NGOs should have more
influence. About one in seven Malawians argues that they should have more, less, or have
the right amount of influence (Table 4).
Table 4: Attitudes towards donors and NGOs’ influence

Don’t know 57.53%
About right 13.76%
Somewhat or far too much 14.63%
Somewhat or far too little 13.78%
Notes:
Respondents were asked: “Do you think that each of the following has too little, too much, or about the right amount
of influence over your government?” (Relating to international donors and NGOs).
Source: Afrobarometer, 2018, Round 4, 2008/2009
Assessing the hypothesis also proves difficult because the reactions of different donors
were similar to one another in both cases, thus restricting the variation in donor reaction.
In the case of Muluzi’s third-term attempt, most donors issued rhetorical condemnation in
the form of warnings in the summer of 2002. These warnings stated that Muluzi’s
government should respect democratic principles and processes and should not interfere
with them. These statements were echoed by popular attitudes. But underhand financing
of some domestic NGOs by donors also kindled anxiety over unjust donor interference
and thus reduced the esteem in which donors were held (Ligomeka & Kang’ombe, 2002).
Using instruments of the logic of appropriateness, international donors signalled to Muluzi
not to “overplay his hand”, and that he faced unified domestic and international
opposition. As argued further above, these donor interventions likely played some part in
why the Muluzi government did not introduce the Third Term Bill in the second half of
2002, but it was probably not the only decisive factor.
In the case of the law reform connected with gender, donor activity congruent with the
logic of appropriateness was applied much more than in Malawi’s third-term debate in
2002. However, in this case donor activity also focused less on public statements and
diplomacy but rather on assisting and working with domestic civil society and NGOs
through lobbying and mobilisation. Donors made use of such instruments as well as other
appropriateness-instruments, such as public statements, in the issue of gay rights in
Malawi. It is useful to contrast the perception of the donors’ role in this issue with their
role in the process of the Marriage, Divorce, and Family Relations Act.
In Malawi, homosexuality is a hotly debated issue. Only a very small number of NGOs
work on LGBTI rights issues, and important civil society actors such as the churches
condemn homosexuality. The arrest and prosecution in court of a gay couple in 2010
sparked broad international condemnation, including a personal intervention by the then UN

Daniel Nowack
Secretary-General Ban Ki Moon (Mwakasungula, 2013). In reaction to this international

condemnation, the prosecution of the gay couple was dropped, suggesting that donor esteem
does amplify the effectiveness of instruments that use the logic of appropriateness.
Besides this, however, the comparison between the homosexuality debate and the child
marriage issue in Malawi shows that donor esteem is context-sensitive. Based on field
research in northern Malawi, McNamara (2014) shows that the homosexuality debate in
Malawi is characterised by the perception that donors are imposing Western culture and
values on Malawians. Consequently, this perception nurtures much opposition towards
donor activity and so-called “interference” in the homosexuality debate. It is striking that in
the case of child marriage such perceptions are lacking or are at least not as strong as in the
homosexuality debate. One explanation for this could be found in the continuous
campaigning on and sensitisation to child marriage that donors undertook for years prior to
the Marriage, Divorce, and Family Relations Act. This suggests that the esteem they enjoy
varies with the issue in which donors become active. Being an esteemed donor does not
trump broad, popular-based opposition. Instead, a fit of attitudes and values between the
society of the recipient country and donors is more crucial to the effectiveness of support for
democracy. Such a fit can be reached through engaging debates on attitudes and values
through campaigns and sensitisation. Furthermore, donor esteem is not fixed but can suffer
if donors engage in issues in which external interference is perceived as unjust.
5 Conclusions and recommendations
The two cases analysed here show that there is no “silver bullet” for effective democracy
support. No grand scheme, no blueprint can direct its instruments. In contrast, democracy
has to be flexible to the specific situation it encounters. Ideally, policymakers could resort
to an adaptive procedure of deciding which instruments to use. The following conclusions
and recommendations may serve as a very first beginning in formulating such decision-
making heuristics.
Identify what leeway for democracy support exists in each issue
The fact that the effectiveness of support for democracy is context-sensitive is clearly
apparent from both of the cases analysed. The extent to which the instruments of both the
logic of consequences and the logic of appropriateness work is restricted by the particular
issues as well as by the specific local domestic attitudes they engage. Some issues are
marked by strong attitudes in partner countries receiving democracy support. In some
instances − for example regarding the issue of polygamy and homosexuality in Malawi −
local domestic attitudes are very well established, and donors may alienate the domestic
population by arguing forcefully against them. In other issues, attitudes are less strong or
are on the verge of changing, as for instance in the case of child marriage in Malawi. In
such instances, donors might find room for the possibility of supporting these attitudes.
Hence, the esteem in which donors are held and the effectiveness of democracy promotion
are often conditioned by the way in which donors engage in which issues. The issue of
child marriage versus polygamy in Malawi’s family law reform highlights this very well.
Moreover this insight urges donors to undertake fine discriminations between issues and
the local domestic attitudes tied to them. Donors must differentiate between upcoming

issues that offer an opportunity to support democracy and those that do not. They then
need to focus on those issues where democracy support has a high chance of being
effective. In contrast, insisting slavishly on the domestication of Western attitudes in all
regards would be the wrong course of action. Hence, this demands a degree of ambiguity
and tolerance on the part of the donors.
Seek open discourse if attitudes are unfavourable
In those cases where attitudes are entrenched and opposed to democracy support and in
which the issue at hand does not allow any space to manoeuvre, careful mobilisation and
sensitisation embedded in a process of societal dialogue can help. This opens up a process
of renegotiation of attitudes. To engage in this strategy, however, means that donors need
to partner up with domestic actors, often stemming from the sphere of civil society. It is
important that such an approach towards renegotiation is seen as a public discourse with
open results, lest it appears as an attempt at foreign control. This will require a lengthy
investment in time and must proceed in a step-by-step fashion.
Help domestic actors to mobilise and coordinate
Even if domestic attitudes are favourable, they may not translate directly into action.
Instead, actors with these attitudes need to be mobilised. Donors can play an important
role in this regard, as they have the resources that domestic actors lack. However, donors
need to bring more to the table than just these resources: The case of Malawi shows very
well that donors can have a pivotal role in coordinating domestic actor groups as well as in
backing them up with moral support.
Conditionality requires concurrence on general values and attitudes
Concerning the differing logics of social action − the logic of consequences and the logic
of appropriateness − the results of the analysis tend to underline the importance of the
logic of appropriateness. The logic of appropriateness yields more sustained effectiveness,
albeit only on a longer time horizon; on a short-time horizon, when critical junctures
suddenly materialise, it is not very effective. In the latter cases, the logic of consequences
is more applicable, but its sustained effectiveness is conditioned on whether there is
already concurrence concerning values and attitudes between donor and partner country.

References
African Church Information Service. (2002a, 1 April). Malawi: Bank withholds funds for development
assistance (Church news service). Retrieved from http://allafrica.com/stories/200203280779.html
African Church Information Service. (2002b, 17 June). Malawi: US opposes Muluzi’s third term bid for
continued rule. Retrieved from https://allafrica.com/stories/200206170169.html
African Church Information Service. (2002c, 14 October). Malawi: Church against change to extend
presidential term (Church news service). Retrieved from
https://allafrica.com/stories/200210140187.html
African Church Information Service. (2002d, 8 November). Malawi: Police disperse demo against third
presidential term (Church news service). Retrieved from
http://allafrica.com/stories/200211080269.html
Afrobarometer: A pan-African series of national public attitude surveys on democracy, governance, and
society. (2018). Website. Retrieved from http://www.afrobarometer.org/data
Allik, J., & Realo, A. (2004). Individualism-collectivism and social capital. Journal of Cross-Cultural
Psychology, 35(1), 24-49. Doi:10.1177/0022022103260381
Banda, M. (19 November 2001). Malawi: EU, USA, UK suspend aid. Daily Times. Retrieved from
Banda, M. (2002a, 30 January). Malawi: Danish government stop aid. Daily Times. Retrieved from
Banda, M. (2002b, 18 April). UDF forms 3rd term team. Daily Times. Retrieved from
Becker, G. S. (1983). A theory of competition among pressure groups for political influence. The Quarterly
Journal of Economics, 98(3), 371-400.
Beichelt, T. (2012). The research field of democracy promotion. Living Reviews in Democracy, 3, 1-13.
Bennett, A. & Checkel, J. T. (2015). Process tracing: From metaphor to analytic tool. Cambridge:
Cambridge University Press.
Boroş, S., Meslec, N., Curşeu, P. L., & Emons, W. (2010). Struggles for cooperation: Conflict resolution
strategies in multicultural groups. Journal of Managerial Psychology, 25(5), 539-554.
Börzel, T. A., & Risse, T. (2003). Conzeptualizing the domestic impact of Europe. In K. Featherstone & C.
M. Radaelli (Eds.), The Politics of Europeanization (pp. 57-87). Oxford: Oxford University Press.
Börzel, T. A., & Risse, T. (2009). The transformative power of Europe: The European Union and the
diffusion of ideas (KFG Working Paper 1/May). Berlin: Freie Universität Berlin.
Britain cautions Muluzi over governance. (31 January 2002). Daily Times. Retrieved from
Brown, S. (2005). ‘Born-again politicians hijacked our revolution!’: Reassessing Malawi’s transition to
democracy. Canadian Journal of African Studies, 38(3), 705-722.
Burnell, P. (2007). Does international democracy promotion work? (Discussion Paper 17/2007). Bonn:
German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE).
Cammack, D. (2017). Malawi’s political settlement: Crafting poverty and peace, 1994-2014. Journal of
International Development, 29(5), 661-677. Doi:10.1002/jid.3288
Carothers, T. (2002). The end of transition paradigm. Journal of Democracy, 13(1), 5-21.
Carroll, A., & Mendos, L. R. (2017). State-sponsored homophobia 2017: A world survey of sexual
orientation laws: Criminalisation, protection and recognition. Geneva: International Lesbian, Gay,
Trans, and Intersex Association (ILGA). Retrieved from https://www.ilga.org/map-sexual-orientation-
laws-overview-2017
Chafunya, B. (2002, 11 July). Malawi: Britain says aid to Lilongwe not lost. Daily Times. Retrieved from

Daniel Nowack
Checkel, J. T. (2001). Why comply? Social learning and European identity change. International
Organization, 55(3), 553-588.
Chinsinga, B. (2008). Malawi’s democracy project at a crossroads. In Konrad Adenauer Stiftung (Ed.),
Towards the consolidation of Malawi’s democracy (Occasional Paper 11). Lilongwe, Malawi: Editor.
Chiweza, A., Wang, V., & Maganga, A. (2016). Acting jointly on behalf of women? The cross-party
women’s caucus in Malawi (CMI Brief 15/8): Bergen, Norway: Chr. Michelsen Institute.
CHRR (Centre for Human Rights and Rehabilitation). (2011). Alternative report for the review of Republic
of Malawi by the Human Rights Committee. Lilongwe, Malawi: Author.
CHRR & CEDEP (Centre for the Development of People). (2013). A shadow report to the Malawi
government's first periodic report on the implementation of the international covenant on civil and
politics rights. Lilongwe, Malawi: Authors.
Collier, D. (2011). Understanding process tracing. Political Science and Politics, 44(4), 823-830.
doi:10.1017/S1049096511001429
Dahl, R. (1971). Polyarchy: Participation and opposition. New Haven, CT: Yale University Press.
Demographic and Health Survey. (2017). Malawi Demographic and Health Survey 2015-16. Zomba,
Malawi: National Statistical Office.
Donors fuel political feud. (26 October 2002). Malawi Insider. Retrieved from
Dulani, B. (2011). Democracy movements as bulwarks against presidential usurpation of power: Lessons
from the third-term bids in Malawi, Namibia, Uganda and Zambia. Stichproben. Wiener Zeitschrift für
kritische Afrikastudien, 11(20), 115-139.
Edwards, B., & McCarthy, J. D. (2007). Resources and social movement mobilization. In D. A. Snow, S. A.
Soule, & H. Kriesi (Eds.), The Blackwell Companion to Social Movements (pp. 116-152). Oxford:
Blackwell Publishing.
Fahamu. (21 April 2015). Malawi’s new marriage law condemned for LGBTI bias. Retrieved from
Fearsome chief, terminator of child marriages. (30 March 2016). Al Jazeera. Retrieved from
Feldman, S. (2013). Political ideology. In L. Huddy, D. O. Sears, & J. S. Levy (Eds.), The Oxford Handbook
of Political Psychology (Vol. 2, pp. 591-626). Oxford: Oxford University Press.
Fiedler, C. (2018). On the effects of international support to peace and democracy in Kenya and Kyrgyzstan
in the aftermath of interethnic violence. Journal of Asian and African Studies, 53(2), 314-329.
Fischer, R., & S. Schwartz. (2011). Whence differences in value priorities? Individual, cultural, or artifactual
sources. Journal of Cross-Cultural Psychology, 45(7), 1127-1144.
Fish, M. S. (2009). Encountering culture. In Z. Barany & R. G. Moser (Eds.), Is democracy exportable? (pp.
57-84). New York: Cambridge University Press.
Ganthu, J. (18 March 2002). Malawi: The IMF cautions government. The Chronicle. Retrieved from
Germany, Italy write-off part of Lilongwe’s debt. (3 July 2002). Daily Trust. Retrieved from
Glasman, L. R., & Albarracín, D. (2006). Forming attitudes that predict future behavior: A meta-analysis of
the attitude-behavior relation. Psychological Bulletin, 132(5), 778-822.
Government of Malawi. (2016). The Constitution of the Republic of Malawi. Lilongwe, Malawi: Author.
Grimm, S., & Leininger, J. (2012). Not all good things go together: Conflicting objectives in democracy
promotion. Democratization, 19(3), 391-414. doi:10.1080/13510347.2012.674355
Halman, L. (2007). Political values. In R. J. Dalton & H.-D. Klingemann (Eds.), The Oxford handbook of
political behavior. Oxford: Oxford University Press.

Ham, C. v. & Lindberg, S. I. (2015). From sticks to carrots: Electoral manipulation in Africa, 1986–2012.
Government and Opposition, 50(3), 521–548.
Harrison, G. (2007). Democratization. In M. Bevir (Ed.), Encyclopedia of Governance (pp. 212-215).
Thousand Oaks, CA: Sage Publications.
Henrich, J., & McElreath, R. (2003). The evolution of cultural evolution. Evolutionary Anthropology, 12(3),
123-135. doi:10.1002/evan.10110
Hofstede, G. (2003). Culture’s consequences: Comparing values, behaviors, institutions and organizations
across nations London: Sage Publications.
Human Rights Watch. (21 November 2016). Human Rights Watch submission to the committee on the rights
of the child concerning Malawi. Retrieved from http://allafrica.com/stories/201611220220.html
Hussein, M. (2004). Strenthening parliamentary democracy in SADC countries: Malawi country report.
Johannesburg: South African Institute of International Affairs (SAIIA).
Ihonvbere, J. O. (1997). From despotism to democracy: The rise of multiparty politics in Malawi. Third
World Quarterly, 18(2), 225-247.
IRIN (United Nations Integrated Regional Information Networks). (2002a, 1 March). Malawi: Problems
with the donors. Retrieved from http://allafrica.com/stories/200203010003.html
IRIN. (2002b, 20 June). Malawi: Donors cancel debt, but warn against third term. Retrieved from
IRIN. (2002c, 23 July). Malawi: Third term debate back on the agenda. Retrieved from
IRIN. (2002d, 23 September). Malawi: Muluzi slams donors over third term. Retrieved from
IRIN. (2002e, 24 October). Malawi: Third term divides ruling party. Retrieved from
Itzchakov, G., Uziel, L., & Wood, W. (2018). When attitudes and habits don’t correspond: Self-control
depletion increases persuasion but not behavior. Journal of Experimental Social Psychology, 75, 1-10.
Jacobs, A. M. (2015). Process tracing the effects of ideas. In A. Bennett & J. T. Checkel (Eds.), Process
tracing: From metaphor to analytic tool (pp. 31-74). Cambridge: Cambridge University Press.
Jamieson, P. (6 October 2002). Malawi: UDF MPs reject third term bid. The Chronicle. Retrieved from
https://allafrica.com/stories/200210080198.html
Jenkins, J. C. (1983). Resource mobilization theory and the study of social movements. Annual Review of
Socioloy, 9, 527-553.
Jerolmack, C., & Khan, S. (2014). Talk is cheap: Ethnography and the attitudinal fallcy. Sociological
Methods & Research, 43(2), 178-209. doi:10.1177/0049124114523396
Kaliya, E. (Ed.). (2015). SADC Gender Protocol Barometer 2015: Malawi. Lilongwe, Malawi: Malawi Human
Rights Resource Center (MHRRC).
Kanguru, J. (14 June 2002). Malawi: Muluzi unsure of seeking new term. The Nation. Retrieved from
Khakee, A. (2009). A long-lasting controversy: Western democracy promotion in Jordan. Amman, Jordan:
Mediterranean Academy of Diplomatic Studies, Malta and Al Urdun Al Jadid Research Center (UJRC).
Law Commission. (2006). Report of the Law Commission on the review of laws on marriage and divorce.
Lilongwe, Malawi: Author.
Lee-Rife, S., Malhotra, A., Warner, A., & McGonagle Glinski, A. (2012). What works to prevent child
marriage: A review of the evidence. Studies in Family Planning, 43(4), 287-303.
Leininger, J. (2010). “Bringing the outside in”: Illustrations from Haiti and Mali for the re-conceptualization
of democracy promotion. Contemporary Politics, 16(1), 63-80. doi:10.1080/13569771003593888
Leininger, J. (2016). Contextualizing democracy promotion. Comparative Democratization, 14(3), 7-11.

Daniel Nowack
Ligomeka, B. (29 November 2000). Malawian chiefs are deathmongers, says First Lady. African Eye News
Service. Retrieved from http://allafrica.com/stories/200011290312.html
Ligomeka, B. & Kang’ombe, P. (26 October 2002). Donors fuel political feud. Malawi Insider. Retrieved
from http://allafrica.com/stories/200210280440.html
Lijphart, A. (2012). Patterns of democracy: Government forms and performance in thirty-six countries. New
Haven, CT: Yale University Press.
Mahoney, J. (2012). The logic of process tracing tests in the social sciences. Sociological Methods &
Research, 41(4), 570-597. doi:0.1177/0049124112437709
Maleki, A., & Hendriks, F. (2015). The relation between cultural values and models of democracy: A cross-
national study. Democratization, 22(6), 981-1010. doi:10.1080/13510347.2014.893426
March, J. G., & Olsen, J. P. (1989). Rediscovering institutions: The organizational basis of politics. New
York: The Free Press, A Division of Macmillan, Inc.
March, J. G., & Olsen, J. P. (2009). The logic of appropriateness (ARENA Working Paper 04/09). Oslo:
University of Oslo, ARENA Centre for European Studies. Retrieved from
https://www.sv.uio.no/arena/english/research/publications/arena-working-papers/2001-
2010/2004/wp04_9.pdf
Marcus, J., & Le, H. (2013). Interactive effects of levels of individualism – collectivism on cooperation: A
meta-analysis. Journal of Organizational Behavior, 34(6), 813-834. doi:10.1002/job.1875
Masina, L. (7 March 2014). Human Rights Watch pushes Malawi to end early marriages. Voice of America.
Retrieved from http://allafrica.com/stories/201403071183.html
McCarthy, J. D., & Zald, M. N. (1977). Resource mobilization and social movements: A partial theory.
American Journal of Sociology, 82(6), 1212-1241.
McNamara, T. (2014). Not the Malawi of our parents: Attitudes toward homosexuality and perceived
westernisation in Northern Malawi. African Studies, 73(1), 84-106.
doi:10.1080/00020184.2014.887747
Meinhardt, H., & Patel, N. (2003). Malawi’s process of democratic transition: An analysis of political
developments between 1990 and 2003. Lilongwe, Malawi: Konrad-Adenauer Stiftung (KAS).
Merkel, W., & Croissant, A. (2004). Conclusion: Good and defective democracies. Democratization, 11(5),
199-213. doi:10.1080/13510340412331304651
MHRC (Malawi Human Rights Commission). (2006). Cultural practices and their impact on the enjoyment
of human rights, particulary the rights of women and children in Malawi. Lilongwe, Malawi: Author.
Mnela, M. (1 April 2002). Malawi: UDF third term bid snubbed. The Chronicle. Retrieved from
Moorman, R. H., & Blakely, G. L. (1995). Individualism-collectivism as an individual difference predictor
of organizational citizenship behavior. Journal of Organizational Behavior, 16(2), 127-142.
doi:10.1002/job.4030160204
Morrow, S. (2006). Toxic mushrooms? The presidential third-term debate in Malawi. In R. Southall & H.
Melber (Eds.), Legacies of Power: Leadership change and former presidents in African politics. (pp.
151-174). Cape Town: Human Sciences Research Council (HSRC).
Muriaas, R., Wang, V., Benstead, L. J., & Rakner, L. (2017). It takes a female chief: Gender and effective
policy advocacy in Malawi (Working Paper 11). Bergen, Norway: Chr. Michelsen Institute and
University of Gothenburg, Sweden, Program on Governance and Local Development.
Mwakasungula, U. (2013). The LGBT situation in Malawi: An activist perspective. In C. Lennox & M. Waites
(Eds.), Human rights, sexual orientation and gender identity in the Commonwealth: Struggles for
decriminalisation and change. London: University of London, School of Advanced Study.
Mwale, F. A. (2012). Law reform as a tool for strengthening legal processes for gender equality in Africa:
The case of Malawi. Unpublished manuscript presented at the Africa Region Colloquium for Women
Leaders as Agents of Change, 28-29 March, Accra, Ghana.

Mwalwanda, G. (16 July 2014). Raise girls’ marrying age, pleads UNFPA. Malawi News Agency. Retrieved
Mwanyongo, T., & Kang’ombe, P. (3 October 2002). Malawi’s religious leaders divided on third term.
Malawi Insider. Retrieved from http://allafrica.com/stories/200210040554.html
Mwase, L. (2002a, 23 September). Malawi: Muluzi, churches clash on 3rd term. The Chronicle. Retrieved
Mwase, L. (2002b, 21 October). Malawi: Muluzi’s 3rd term letter leaked. The Chronicle. Retrieved from
Mweninguwe, R. (25 October 2002). Malawi: Political anarchy besets Malawi. Mail & Guardian. Retrieved
Namingha, M. (2002a, 2 September). Malawi: Third term non starter! The Chronicle. Retrieved from
Namingha, M. (2002b, 11 November). Malawi: Third term on this December. The Chronicle. Retrieved
New marriage law deemed invalid. (26 October 2015). Nyasa Times. Retrieved from
Nunn, N. (2012). Culture and the Historical Process. Economic History of Developing Regions, 27, Supplement
1, 108-126.
O’Donnell, G. (2004). Why the rule of law matters. Journal of Democracy, 15(4), 32-46.
OECD (Organisation for Economic Co-operation and Development. (2014). Social Institutions and Gender
Index: 2014 synthesis report. Paris: Author.
Patel, N., & Wahman, M. (2015). The presidential, parliamentary and local elections in Malawi, May 2014.
Africa Spectrum, 50(1), 79-92.
Poppe, A. E., & Wolff, J. (2013). The normative challenge of interaction: Justice conflicts in democracy
promotion. Global Constitutionalism 2(3), 373-406. doi:10.1017/S204538171200024X
Posner, D. N., & Young, D. J. (2007). The institutionalization of political power in Africa. Journal of
Democracy, 18(3), 126-140.
Resnick, D. (2012). Two steps forward, one step back: The limits of foreign aid on Malawi’s democratic
consolidation (WIDER Working Paper 28). Helsinki: United Nations University World Institute for
Developoment (UNU-WIDER).
Schumpeter, J. A. (2003). Capitalism, socialism and democracy. London: Taylor & Francis.
Schwartz, S. H. (1992). Universals in the content and structure of values: Theoretical advances and
empirical tests in 20 countries. Advances in Experimental Social Psychology, 25, 1-62.
Schwartz, S. H. (2006). A theory of cultural value orientations: Explication and applications. Comparative
Sociology, 5(2), 137-182.
Schwartz, S. H., & Boehnke, K. (2004). Evaluating the structure for human values with confirmatory factor
analysis. Journal of Research in Personality, 38(3), 230-255.
The Catholic Commission for Justice and Peace Statement on the third term bill. (21 October 2002). The
Chronicle. Retrieved from http://allafrica.com/stories/200210210561.html
Tull, D. M., & Simons, C. (2017). The institutionalism of power revisited: Presidential term limits in Africa.
Africa Spectrum, 52(2), 79-102.
UNOHCHR (United Nations Office of the High Commissioner of Human Rights). (1989). United Nations
Convention on the Rights of the Child. Retrieved from
https://www.ohchr.org/en/professionalinterest/pages/crc.aspx
Van Hoorn, A. (2014). Individualist-collectivist culture and trust radius: A multilevel approach. Journal of
Cross-Cultural Psychology, 46(2), 269-276. doi:10.1177/0022022114551053

Daniel Nowack
Vanhanen, T. (2000). A new dataset for measuring democracy, 1810-1998. Journal of Peace Research,
37(2), 251-265.
Wagner III, J. A. (1995). Studies of individualism-collectivism: Effects on cooperation in groups. Academy
of Management Journal, 38(1), 152-172. doi:10.2307/256731
Warren, M. E. (2011). Civil society and democracy. In M. Edwards (Ed.), The Oxford handbook of civil
society. Oxford: Oxford University Press.
Witmos, B. (22 February 2014). Chief Kwataine in campaing against early marriages. Malawi News Agency.
Retrieved from http://allafrica.com/stories/201402240831.html
Wolff, J., & Zimmermann, L. (2016). Between Banyans and battle scenes: Liberal norms, contestation, and
the limits of critique. Review of International Studies, 42(3), 513-534.
doi:10.1017/S0260210515000534
World Bank. (2018). DataBank: Gender statistics. Retrieved from
http://databank.worldbank.org/Data/indicator/SG.LAW.CHMR?id=2ddc971b&report_name=Gender_I
ndicators_Report&populartype=series
Wroe, D. (2012). Briefing: Donors, dependency, and political crisis in Malawi. African Affairs, 111(442),
135-144. doi:10.1093/afraf/adr076

Appendix 1
Table A1: Timeline of Muluzi’s third-term attempt

Date Event
Muluzi wins Malawi’s first democratic presidential election and enters his first term
1994
as president
Presidential election: Muluzi gets re-elected as president; first rumours circulate that
1999 - mid-year
Muluzi plans to secure himself a third term
Withdrawal of direct budgetary support by the EU, UK, and USA due to corruption
2001
allegations
Catholic bishops publicise a pastoral letter that cautions against manipulations of
2001 - 25 March
Malawi’s constitution
2001 - April CCAP issues a pastoral letter emphasising Malawi’s democratic culture
2002 - January Denmark suspends all aid to Malawi
Failed attempt to change the constitutional provision for amendments to the
2002 - beginning
Constitution from two-thirds to simple majority
Catholic church warns against amendments to the Constitution without public
2002 - April
consultation in its Lentine Pastoral Letter
2002 - May Muluzi’s government issues a ban on demonstrations
High Court rules the demonstrations ban unconstitutional; Muluzi counters this ruling
2002 - early June
publicly, and the High Court overturns it
2002 - 7 June US issues a public statement that opposes the Open Term Bill
EU publicly voices concern over how Muluzi’s government handles the Open Term
2002 - mid-June
Bill debate
Germany and Italy cancel parts of Malawi’s debt, but urge Muluzi’s government to
2002 - end-June
adhere to democratic practices
2002 - July The Open Term Bill is tabled in parliament, but is narrowly defeated
Muluzi’s party (UDF) meets and concludes to table the Third Term Bill later in the
2002 - July
year
2002 - August ELCM publicly condemns attempts to secure a third presidential term
2002 - 6
The Third Term Bill is gazetted
September
IMF and World Bank are discussing further aid programmes with the Muluzi
2002 - September government and allegedly demanded clarification of Muluzi’s position on the Third
Term Bill
2002 - early Muluzi government institutes another ban on rallies and demonstrations against the
October Third Term
2002 - 26 October CCJP issues a statement that condemns the Third Term Bill
2002 - early Street fights between police and protesters against the Third Term Bill take place in
November Blantyre
Third Term Bill is sent to the parliamentary Legal Affairs Committee for revision
2003 - beginning
(that is, de facto withdrawal)
2003 - February PAC petitions Legal Affairs Committee to withdraw the Third Term Bill officially
Muluzi announces his nomination of Bingu wa Mutharika as UDF candidate in the
2003 - March
upcoming presidential elections

Daniel Nowack
Table A2: Timeline of the Marriage, Divorce, and Family Relations Act
Date Event
1994 Malawi’s new democratic Constitution includes a provision on gender equality
4th UN Conference on Women passes the Beijing Declaration and Platform for
1995
Action
Within Malawi’s parliament, female MPs officially found the cross-party Women’s
1996
Caucus
Malawi draws up the National Platform for Action to implement the Beijing
1997
Declaration
Civil rights activists supported by UNDP found the Gender Coordination Network, an
1998
umbrella-NGO that cooperates extensively with the parliamentary Women’s Caucus
2001 - September Special Law Commission on Gender (Gender Commission) is instituted
Gender Commission conducts workshops and consultations to lay down a

2002
programme for legal reform
UNFPA (United Nations Population Fund) issues its first report on the occurrence of
2003 - June
fistula in Sub-Saharan Africa
2003 UNFPA launches the worldwide “End Fistula Week”
Gender Commission publishes its report on a legal reform of marriage law,
2006 accompanying a draft for the Marriage, Divorce, and Family Relations Bill that
harmonises existing law and criminalises child marriage and polygamy
Malawi’s Ministry of Gender launches the “Fistula Week”, a week-long campaign to
2008
sensitise people on the health issue of fistula
2008 - September Malawian NGO Girls Empowerment Network (GENET) founded
Malawi’s annual “Children’s Parliament”, an institutionalised national assembly of

2009
youth people demands that the legal age of marriage be raised to 18
“Chidyamakanda Bill”, which raises the legal age of the child from 15 to 16, is
2009 tabled in parliament, and passed; civil society groups successfully pressure the
president not to assent to the bill
Marriage, Divorce, and Family Relations Bill is introduced to parliament, but sent
2010
back to the Gender Commission for revision
Child Care, Justice, and Protection Act on children rights and parental obligations is
2010
passed
Amendment to the Constitution allow specified sections of the Constitution to be
2010
changed through a two-thirds parliamentary vote instead of through a referendum
GENET, in cooperation with international partners, launches the “Stop Child
2011
Marriage Campaign”
2013 - February The international movement “Girls Not Brides” opens a Malawi chapter
2013 UN Women opens office and becomes operative in Malawi

2013 - mid-year Issue of reforming marriage age is discussed in media
2013 - June Marriage, Divorce, and Family Relations Bill on media agenda
Traditional Authorities (chiefs) launch a campaign lobbying for setting the marriage
2014 - February
age at 21

Human Rights Watch Report on child marriage in Malawi calls on government to

2014 - March
raise the legal marriage age
Marriage, Divorce, and Family Relations Bill is presented to cabinet, and subsequently
2014 - October
introduced to parliament; MPs express concern about polygamy provision
Ministry of Gender in cooperation with Malawian NGO launches the “End Child
2014 - November
Marriage Campaign”
On an initiative of UN Women, the president and parliamentarians officially join the
2015 - February
global “He For She” campaign
Marriage, Divorce, and Family Relations Bill is tabled in parliament, not containing
2015 - February the polygamy provision anymore, and is passed, but is still in conflict with
constitutional provisions
2015 - April President assents to the Marriage, Divorce, and Family Relations Act
NGOs and CSOs lodge complaint with the AU because of Malawi’s failure not to
2016
implement international agreements to criminalise child marriage
2016 - November Malawian government pledges to the AU that it will criminalise child marriage
Parliament passes the constitutional amendment to raise the age of the child to under
2017 - February
18 years
2017 - April President assents to the constitutional amendment

Daniel Nowack
Appendix 2
List of interviews
Interview No. Type of actor
1 Malawian CSO
2 Malawian CSO
3 Government of Malawi
4 Malawian CSO
5 Malawian CSO
6 Parliament of Malawi
7 International organisation
9 Malawian CSO
10 Malawian CSO
13 Malawian CSO
14 OECD donor
16 Parliament of Malawi
19 Malawian CSO
20 Malawian private firm
21 OECD donor

Appendix 3
The two empirical types of cases we identified were mapped onto two different
dimensions of cultural/value orientations as found in the existing literature. Cases of type
One, the third-term cases, were mapped unto the individualism-collectivism cultural
dimension (Hofstede, 2003), while cases of type Two, cases on law reform, were mapped
onto the egalitarian-hierarchy dimension (Schwartz, 2006).
Items: Third-term debate => individualism-collectivism
Since Hofstede’s original conceptualisation, individualism-collectivism has been

investigated from various vantage points. While the concept of individualism, and
individualistic value orientations, emphasises loose ties between individuals, self-reliance,
and the importance of the self above the collective (or group), collectivism puts more
emphasis on the collective, in contrast to the self, and on the integration of individuals in
strong, cohesive and united groups. Importantly in relation to our cases, while
individualism favours competition, collectivism favours consensus and cooperation.
Earlier studies established that collectivism is associated with better in-group coordination
(for example, Wagner, 1995) as people seem to be more likely to act pro-socially out of
inherent motivations without direct external incentives (Moorman & Blakely, 1995).
However, the research literature has not found such coordination payoffs on a societal
level at which it could be argued that collectivism would thus support a better coordinated
opposition to presidents seeking an end to term limits (Marcus & Le, 2013). Hence,
individualism − with its emphasis on competition − can be hypothesised to raise the stakes
against a possible lifting of presidential term limits as such term limits hamper political
competition substantially. The associations found by the research literature between
individualism and greater social capital, and radiuses of interpersonal trust (Allik & Realo,
2004; Van Hoorn, 2014) support this hypothesis: a politically more engaged population
may express opposition against lifting a presidential term limit more directly and strongly.
Unfortunately, it is difficult to re-construct the individualism-collectivism dimension

using the Afrobarometer data. Hence, we identified only three out of eight possible items
that could be used to construct a rough individualism-collectivism index (all items taken
from Afrobarometer, Round 2):
1.) “Which of the following statements is closest to your view? Choose Statement A or
Statement B:
A: Since everyone is equal under the law, leaders should not favour their own family or
group.
B: Once in office, leaders are obliged to help their own family or group.”
2.) “Let’s talk for a moment about the kind of society we would like to have in this
country. Which of the following statements is closest to your view? Choose Statement A or
Statement B:
A: People should look after themselves and be responsible for their own success in life.
B: The government should bear the main responsibility for the well-being of people.”
Statement B:

Daniel Nowack
A: In order to make decisions in our community, we should talk until everyone agrees.
B: Since we will never agree on everything, we must learn to accept differences of opinion
within our community.”
Response categories to all items are semantic differentials of the following kind:
Agree very strongly with A
Agree with A
Agree with B
Agree very strongly with B
Agree with neither
Items: Law reform => egalitarian-hierarchy
The law reform cases we identified all focus on political equality. We hence chose to link
the law reform cases up to value orientations that more closely focus on the issue of
equality than the individualism-collectivism dimension does. Following Schwartz (2006)
we chose his egalitarianism-hierarchy dimension. According to Schwartz’ own
conceptualisation, the hierarchy dimension describes
hierarchical systems of ascribed roles to insure responsible, productive behaviour.

It defines the unequal distribution of power, roles, and resources as legitimate.
People are socialized to take the hierarchical distribution of roles for granted and to
comply with the obligations and rules attached to their roles. Values like social
power, authority, humility, and wealth are highly important in hierarchical cultures.
(Schwartz, 2006, p. 141)
In contrast, egalitarian value-orientations
induce people to recognise one another as moral equals who share basic interests as human
beings. People are socialized to internalize a commitment to cooperate and to feel concern
for everyone’s welfare. They are expected to act for the benefit of others as a matter of
choice. Important values in such cultures include equality, social justice, responsibility,
help, and honesty. (Schwartz, 2006, p. 140)
Again we identified three items of the Afrobaromter (Round 3) that can be used to construct
a measure indicating Schwartz’ egalitarianism-hierarchy dimension. We attempted to map
them onto the sub-dimensions/concepts that make up Schwartz’ egalitarianism-hierarchy
(see Schwartz & Boehnke, 2004; Schwartz, 1992). In contrast to the World Values Survey,
the Afrobarometer does not aim at producing a Schwartz Index, hence, only a very limited
number of the 11 sub-dimensions making up the egalitarianism-hierarchy dimension up
could be covered:

Statement B.
A: In our country, women should have equal rights and receive the same treatment as men do.
B: Women have always been subject to traditional laws and customs, and should remain
so.”
(Mapped onto “equality” sub-dimension/concept)
2.) “Let’s talk for a moment about the kind of society we would like to have in this country.
Which of the following statements is closest to your view? Choose Statement A or Statement B.
A: As citizens, we should be more active in questioning the actions of our leaders.
B: In our country these days, we should show more respect for authority.”
(Mapped onto “authority” sub-dimension/concept)
Statement B.
A: It is better to have free schooling for our children, even if the quality of education is low.
B: It is better to raise educational standards, even if we have to pay school fees.”
(Mapped onto “social justice” sub-dimension/concept)
Again, response categories to all three items were given as semantic differentials:
Agree very strongly with A
Agree with A
Agree with B
Agree very strongly with B
Agree with neither
Methodology of index construction & results
The indices were constructed using principal components analysis (PCA). All items were
recoded, so that their response categories became aligned. The ultimate PCA was done on a
polychoric correlation matrix to account for the ordinal scales of the items. In both cases,
only one component was retrieved which explained only about 40 per cent of overall
variation between the items, and scores for each respondent were calculated and normalised
to range between 0 (egalitarian/individualist) and 1 (hierarchical/collectivist) (see also
STATA-do file for more details).

Daniel Nowack
Country means are shown in Table A3, and the boxplots below inform on the distribution:
Table A3: Mean index values for four selected countries

Index Malawi Uganda Mali Senegal
Egalitarianism
.395 .36 .418 .335
versus hierarchy
Individualism
versus .404 .542 .414 .353
collectivism
Source: Own computation, based on Afrobarometer, 2018

Publications of the German Development Institute /
Deutsches Institut für Entwicklungspolitik (DIE)
Discussion Papers
26/2018 Hulse, Merran. Cultural values, popular attitudes and democracy promotion: How values
mediate the effectiveness of donor support for term limits and LGBT+ rights in Uganda
(45 pp.). ISBN 978-3-96021-082-5. DOI: 10.23661/dp26.2018.
25/2018 El-Haddad, Amirah. Exporting for growth: Identifying leading sectors for Egypt and
Tunisia using the Product Space Methodology (49 pp.). ISBN 978-3-96021-075-7. DOI:
10.23661/dp25.2018.
24/2018 Chmiel, Oskar. The engagement of Visegrad countries in EU-Africa relations (40 pp.).
ISBN 978-3-96021-081-8. DOI: 10.23661/dp24.2018.
23/2018 Belke, Ansgar, & Ulrich Volz. Capital flows to emerging market and developing
economies: Global liquidity and uncertainty versus country-specific pull factors (57 pp.).
ISBN 978-3-96021-080-1. DOI: 10.23661/dp23.2018.
22/2018 Baydag, Rena Melis, Stephan Klingebiel, & Paul Marschall. Shaping the patterns of aid
allocation: A comparative analysis of seven bilateral donors and the European Union
(42 pp.). ISBN 978-3-96021-079-5. DOI: 10.23661/dp22.2018.
21/2018 Wehrmann, Dorothea. Incentivising and regulating multi-actor partnerships and private
sector engagement in development cooperation (50 pp.). ISBN 978-3-96021-078-8.
DOI:10.23661/dp21.2018.
20/2018 Scheumann, Waltina, & George Phiri. Coordination: The key to governing the water-land-
food nexus in Zambia? (47 pp.). ISBN 978-3-96021-077-1. DOI: 10.23661/dp20.2018
19/2018 Grimm, Sven, Mareike Magdalena Gensch, Johanna Hauf, Julia Prenzel, Nitja Rehani,
Sarah Senz, & Olivier Vogel. The interface between research and policy-making in South
Africa: Exploring the institutional framework and practice of an uneasy relationship (45
pp.). ISBN 978-3-96021-076-4. DOI: 10.23661/dp19.2018.
18/2018 Griffith-Jones, Stephany, & Samuel Leistner. Mobilising capital for sustainable
infrastructure: The cases of AIIB and NDB (55 pp.). ISBN 978-3-96021-074-0. DOI:
10.23661/dp18.2018
17/2018 Eger, Jens, Hannes Öhler, & Alexandra Rudolph. Is the sectoral aid allocation within
countries need-oriented? (15 pp.). ISBN 978-3-96021-073-3. DOI:10.23661/dp17.2018
16/2018 Groß, Lisa. Assessing the impact of governance programmes: GIZ support to citizen
participation in local governance in Benin (59 pp.). ISBN 978-3-96021-072-6.
DOI:10.23661/dp16.2018
15/2018 Bohnet, Michael, Stephan Klingebiel, & Paul Marschall. Die Struktur der deutschen
öffentlichen Entwicklungszusammenarbeit: Hintergründe, Trends und Implikationen für
das BMZ und andere Bundesressorts (84 pp.). ISBN 978-3-96021-071-9.
DOI:10.23661/dp15.2018
14/2018 Högl, Maximilian. Enabling factors for cooperation in the climate negotiations – a
comparative analysis of Copenhagen 2009 and Paris 2015 (70 pp.). ISBN 978-3-96021-
070-2. DOI: 10.23661/dp14.2018
[Price: EUR 6.00; publications may be ordered from the DIE or through bookshops.]
For a complete list of DIE publications:
www.die-gdi.de

Notes

Uploaded by

Copyright:

Available Formats

Notes

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Notes

Uploaded by

Copyright:

Available Formats

Discussion Paper 27/2018

Cultural Values, Attitudes, and

How Values Mediate the Effectiveness of Donor

Die deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie;

ISBN 978-3-96021-084-9 (printed edition)

Daniel Nowack is a researcher in the research programme “Transformation of Political (Dis-)Order:

© Deutsches Institut für Entwicklungspolitik gGmbH

Bonn, 25 September 2018 Daniel Nowack

These findings prompt a number of recommendations for donors wishing to promote

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 1

2 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

Democracy promotion is a highly normative undertaking. This expresses itself in repeated

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 3

2 Research design: theory and method

2.1 Discussion of concepts: democracy, democratisation, democracy promotion,

In its broadest understanding, the term “democratisation” describes an “expansion of a set of

Political competition and equality appear to be essential conceptual characteristics in most

Strengthening these two fundamental principles of democracy in recipient countries is the

4 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

actors and institutions in the recipient country, or indirectly by establishing favourable

There is no established and widely regarded conceptualisation or definition of values and

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 5

Hofstede’s (2003) individualism-collectivism dimension describes a value continuum

A similar well-established value orientation is provided by Schwartz’ cultural value theory

6 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

2.2 Actor constellations, values, and the effectiveness of democracy promotion

As noted above, democracy support as an element of development cooperation is seen as

Instead of making an external-internal distinction, this paper follows Leininger (2010) in

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 7

2.2.1 Organisational resources

2.2.2 Congruence of attitudes

8 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 9

logic of consequences and the logic of appropriateness provides for an answer. As

The outcome variable of interest is the effectiveness of democracy promotion.

To capture the effectiveness of democracy promotion, instances in which democracy

10 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 11

2.4 Case selection

12 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

To facilitate the process-tracing analysis, we collected and analysed primary newspapers

3 Country analysis: democratic reforms in Malawi

3.1 The political background of Malawi, 1992 to 2017

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 13

2 For a list of personal interviews, see Appendix 2.

14 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

Figure 2: Value indices for Malawi: egalitarianism-hierarchy and individualism-collectivism

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 15

3.2 Case 1: Muluzi’s attempt at a third term

3.2.1 Process description

Table A1 in Appendix 1 provides an overview of the process of Muluzi’s third-term

16 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 17

3.2.2 Explanatory factors

Civil society & popular attitudes

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 19

From the very beginning, church organisations cooperated with non-governmental

20 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE)

Table 1: Attitudes on term limits and one-man rule

German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 21